Deep neural networks have established themselves as the dominant approach for image restoration problems such as deblurring and denoising. However, the high computational complexity of state-of-the-art designs prevents their effective use in resource-constrained scenarios such as edge devices, where power-efficient inference is key. In this paper, we present a state-of-the-art backbone neural network design for image restoration, called MuFIR (Multiplication-Free Image Restoration), that is entirely devoid of multiplication operations. When coupled with suitable hardware implementations, the proposed concept enables fast and low-complexity inference by requiring only integer additions and bitshifts. This is made possible by several ingredients proposed in this work, namely ternary weight quantization to eliminate multiplications in the main network layers, careful use of novel normalizations to ensure stability of the ternarized architecture, and quantization of specific parameters and activations to combinations of powers of two, to remove the remaining multiplications. This is coupled with an annealed training procedure which progressively transforms a conventional network into our multiplication-free design. We experimentally show that, despite the all-integer operations and the lack of multiplications, MuFIR achieves performance close to that of full-precision models in terms of deblurring and denoising quality.

Personalized image generation requires text-to-image generative models that capture the core features of a reference subject to allow for controlled generation across different contexts. Existing methods face challenges due to complex training requirements, high inference costs, limited flexibility, or a combination of these issues. In this paper, we introduce DreamCache, a scalable approach for efficient and high-quality personalized image generation. By caching a small number of reference image features from a subset of layers and a single timestep of the pretrained diffusion denoiser, DreamCache enables dynamic modulation of the generated image features through lightweight, trained conditioning adapters. DreamCache achieves state-of-the-art image and text alignment, utilizing an order of magnitude fewer extra parameters, and is both more computationally effective and versatile than existing models.1

We present stochastic Gaussian splatting (SGS): the first framework for uncertainty estimation using Gaussian splatting (GS). GS recently advanced the novel-view synthesis field by achieving impressive reconstruction quality at a fraction of the computational cost of neural radiance fields (NeRFs). However, contrary to the latter, it still lacks the ability to provide information about the confidence associated with their outputs. To address this limitation, in this brief, we introduce a variational inference (VI)-based approach that seamlessly integrates uncertainty prediction into the common rendering pipeline of GS. In addition, we introduce the area under sparsification error (AUSE) as a new term in the loss function, enabling optimization of uncertainty estimation alongside image reconstruction. Experimental results on the three different datasets demonstrate that our method outperforms existing approaches in terms of both image rendering quality and uncertainty estimation accuracy. Overall, our framework equips practitioners with valuable insights into the reliability of synthesized views, facilitating safer decision-making in real-world applications.

Learning deep super-resolution models without the need for ground truth data at a higher resolution is critical for satellite imaging applications. This is either due to the lack of existing images at better resolution for certain target wavelengths or the existence of significant domain gaps between the images of different satellites. In this paper, we propose a method and neural network architecture for a multi-image super-resolution problem, where each image in the input stack might be affected by a different degradation process. A test-time finetuning procedure allows to dynamically account for the degradations observed for a specific set of LR inputs, improving over baseline results.

AI-based compression is gaining popularity for traditional photos and videos. However, such techniques do not typically scale well to the task of compressing hyperspectral images, and may have computational requirements in terms of memory usage and total floating point operations that are prohibitive for usage onboard of satellites. In this paper, we explore the design of a predictive compression method based on a novel neural network design, called LineRWKV. Our neural network predictor works in a line-by-line fashion limiting memory and computational requirements thanks to a recurrent inference mechanism. However, in contrast to classic recurrent networks, it relies on an attention operation that can be parallelized for training, akin to Transformers, unlocking efficient training on large datasets, which is critical to learn complex predictors. In our preliminary results, we show that LineRWKV significantly outperforms the state-of-the-art CCSDS-123 standard and has competitive throughput.

The 26th IEEE International Workshop on Multimedia Signal Processing (MMSP 2024), organized by the Multimedia Signal Processing Technical Committee (MMSP-TC) of IEEE Signal Processing Society (SPS), was held at Purdue University, West Lafayette, Indiana, U.S.A., from October 2 - 4, 2024.

Generating videos with realistic and physically plausible motion is one of the main recent challenges in computer vision. While diffusion models are achieving compelling results in image generation, video diffusion models are limited by heavy training and huge models, resulting in videos that are still biased to the training dataset. In this work we propose MotionCraft, a new zero-shot video generator to craft physics-based and realistic videos. MotionCraft is able to warp the noise latent space of an image diffusion model, such as Stable Diffusion, by applying an optical flow derived from a physics simulation. We show that warping the noise latent space results in coherent application of the desired motion while allowing the model to generate missing elements consistent with the scene evolution, which would otherwise result in artefacts or missing content if the flow was applied in the pixel space. We compare our method with the state-of-the-art Text2Video-Zero reporting qualitative and quantitative improvements, demonstrating the effectiveness of our approach to generate videos with finely-prescribed complex motion dynamics.

Mobile traffic forecasting is an important topic for mobile operators as accurate predictions are essential for en-hancing network efficiency. However, predicting mobile demand is difficult because it requires to distinguish different types of patterns. Recently, deep learning based approaches have been proven effective because of their ability to learn complex non-linear dependencies. Among different architectures, the models built on Recurrent Neural Network (RNN) show impressive performance in sequence modelling and they are widely used in mobile traffic forecasting; however, their rather long training time raises the issue of a high training cost for mobile operators. To better capture the latest trend of mobile demand, mobile op-erators would retrain the forecasting model every week using the recent observations; the training is very expensive and consumes a considerable amount of energy considering that massive cells are managed by mobile operators. For mobile operators, it is always desirable to reduce the training cost without reducing the forecasting performance. To address this challenge, in this paper we propose a pure Fe-layer deep learning predictor allowing the mobile operators to obtain excellent forecasting performance along with very low training cost. The network is composed of three feature extraction layers, containing RNN -inspired blocks learning features that are robust to unexpected local variations; these features are then used to make predictions in the last two layers. Extensive experiments are conduct on a real-world dataset, showing that our model obtains similar prediction performance as the state-of-the-art RNN-based baseline model, while reuuiring only 1 % of its trainina time.

Abstract Recently, biometric recognition has become a significant field of research. The concept of cancelable biometrics (CB) has been introduced to address security concerns related to the handling of sensitive data. In this paper, we address unconstrained face verification by proposing a deep cancelable framework called BiometricNet+ that employs random projections (RP) to conceal face images and compressive sensing (CS) to reconstruct measurements in the original domain. Our lightweight design enforces the properties of unlinkability, revocability, and non-invertibility of the templates while preserving face recognition accuracy. We compare facial features by learning a regularized metric: at training time, we jointly learn facial features and the metric such that matching and non-matching pairs are mapped onto latent target distributions; then, for biometric verification, features are randomly projected via random matrices changed at every enrollment and query and reconstructed before the latent space mapping is computed. We assess the face recognition accuracy of our framework on challenging datasets such as LFW, CALFW, CPLFW, AgeDB, YTF, CFP, and RFW, showing notable improvements over state-of-the-art techniques while meeting the criteria for secure cancelable template design. Since our method requires no fine-tuning of the learned features, it can be applied to pre-trained networks to increase sensitive data protection.

Deep learning methods have traditionally been difficult to apply to compression of hyperspectral images onboard spacecrafts due to the large computational complexity needed to achieve adequate representational power, as well as the lack of suitable datasets for training and testing. In this article, we depart from the traditional autoencoder approach, and we design a predictive neural network, called line receptance weighted key value (LineRWKV), which works recursively line by line to limit memory consumption. In order to achieve that, we adopt a novel hybrid attentive-recursive operation that combines the representational advantages of Transformers with the linear complexity and recursive implementation of recurrent neural networks (RNNs). The compression algorithm performs the prediction of each pixel using LineRWKV, followed by entropy coding of the residual. Experiments on multiple datasets show that LineRWKV is highly memory-efficient, significantly outperforms state-of-the-art deep learning methods, and is the first deep learning approach to outperform CCSDS-123.0-B-2 at lossless and near-lossless compression. Promising throughput results are also evaluated on a 7-W embedded system.

Developing deep learning techniques for super-resolving bursts of images acquired by mobile cameras is a topic that has recently gained significant interest. This topic fits the general problem of learning-based multi-image super-resolution (SR), which, contrary to its sibling single-image SR, has so far received little attention despite its potential. In this work, we introduce a neural network architecture for burst SR, called MLB-FuseNet (Multi-Level Burst Fusion Network), that is capable of extracting features in a manner that is invariant to permutations in the burst and to progressively condition features extracted from a reference image. Permutation invariance is desirable as it is known that the order of images in a burst does not matter in this problem, but its study has so far been neglected. Moreover, we also introduce a module exploiting a polyphase decomposition to improve feature extraction from mosaiced raw images. Results show an improvement over the state of the art on the BurstSR dataset – a recent and popular benchmark for this problem.

Hyperbolic neural networks are emerging as an effective technique to better capture hierarchical representations of many data types, from text to images and, recently, point clouds. In this paper, we extend our earlier work, that showed how to use regularizers in the hyperbolic space to improve performance of point cloud classification models, to the problem of part segmentation. This requires careful modeling of the hierarchical relationships between parts and whole point cloud to properly control the hyperbolic geometry of the feature space produced by the neural network. We show how the proposed method improves the performance of commonly used neural network architectures, reaching state-of-the-art performance on the part segmentation task.

Mobile traffic forecasting plays a key role for optimizing the configuration of network cells. Network operators are very interested in predicting upcoming mobile traffic peaks in an accurate way, in order to improve quality of service via efficient resource allocation. However, forecasting potential peaks is very challenging considering that many peaks occur suddenly for no apparent reason; hence, it is very difficult to determine if a peak will occur in near future based on the temporal dynamics of mobile traffic, potentially leading to inaccurate predictions. To improve the performance of peak prediction, we propose a novel deep learning model called Mixture of Quantiles (MoQ). MoQ employs a mixture of experts model featuring a manager to fuse the predictions of multiple experts. In order to overcome the problem of overly smooth predictions on peaks, the experts are designed to have differentiated forecasting styles from conservative to aggressive. A cooperation mechanism is established through a carefully designed training process, whereby conservative experts are responsible for the forecasting of the off-peak region, and the employed experts are switched to the aggressive ones once the potential increasing trend is detected by manager, which leads to significantly improved peak predictions. Extensive experiments on real-world dataset showcase the effectiveness of the proposed MoQ model, which outperforms all the benchmarks and shows its superior performance in peak forecasting along with excellent interpretability.

In this paper, we explore the onboard processing capabilities of an optical Earth observation instrument operating under the principles of compressed sensing, currently under preliminary study. In particular, we focus on two main aspects for onboard operations: i) how to process measurements in a computationally-efficient way to obtain previews of the reconstructed image that can be easily used by downstream inference algorithms; ii) the possibility of having simultaneous compression and encryption by proper management of the pseudorandom patterns used for the sensing matrix and measurements

Compressive sensing has established itself as a novel imaging paradigm. In this paper, we analyze the behavior of a a compressive instrument based on spatial light modulators (SLM), operating in the mid-infrared. We show that, contrary to the well-studied visible and near-infrared wavelengths, mid-infrared poses modeling challenges due to non-negligible SLM diffraction effects. We show a way to model such effect analytically and to account for them in the reconstruction process, leading to improved reconstruction quality.

Multi-image super-resolution (MISR) is a technique used to increase the spatial resolution of images acquired by remote sensing platforms by combining the images acquired through multiple revisits. Supervised training of MISR models requires collecting high-resolution images to be used as ground truth. Except for a few special cases, this involves acquiring images from a different satellite, resulting in a shift in the optical and radiometric characteristics with respect to the sensor to be super-resolved. In this paper, we explore the use of Sentinel-2 images to train a MISR model for Proba-V images and highlight the challenges of this pursuit.

Multi-temporal super-resolution (SR) whereby a number of images of the same scene acquired at different times are fused to enhance its spatial resolution has recently enjoyed great success thanks to advances in deep learning methods. However, the literature has so far focused on supervised training approaches that require the availability of high-resolution (HR) images at the target resolution. This is a significant limitation because such imagery may not exist, might be difficult to source or exhibit domain gaps such as different spectral bands or radiometric characteristics. Unsupervised training approaches that do not require imagery beyond the input low resolution are needed to overcome this limitation. This paper presents a first analysis of the problem, taking inspiration from the literature on blind single-image SR, but also focusing on the uniqueness of multi-temporal satellite images. Our preliminary results show that it is indeed possible to develop accurate deep learning models for multi-temporal SR without HR images.

Mobile traffic forecasting is crucial for optimizing the network configuration and improving Quality-of-Service (QoS); accurate mobile traffic predictions can help the network operators better configure the network and adapt to the trend of mobile demand. However, improving mobile traffic forecasting is challenging as traffic patterns exhibit strong periodicity while also retaining a certain level of randomness; this makes it difficult to model the sequence correlation. To obtain better mobile traffic predictions, we proposed a deep learning-based predictor called Temporal Dynamics Aware Network (TDANet). TDANet is carefully designed to extract both the global and the local temporal patterns employing recurrent and attention components; a linear module is employed to make the predictions more sensitive to the magnitude of mobile traffic, which makes the model better able to capture fast dynamics. Extensive experiments are conducted on a real-world mobile traffic dataset, and the results show that TDANet outperforms all the baseline models on all evaluation metrics, including both accuracy and complexity.

Abstract In this work, we present the Gaussian Class-Conditional Simplex (GCCS) loss: a novel approach for training deep robust multiclass classifiers that improves over the state-of-the-art in terms of classification accuracy and adversarial robustness, with little extra cost for network training. The proposed method learns a mapping of the input classes onto Gaussian target distributions in a latent space such that a hyperplane can be used as the optimal decision surface. Instead of maximizing the likelihood of target labels for individual samples, our loss function pushes the network to produce feature distributions yielding high inter-class separation and low intra-class separation. The mean values of the learned distributions are centered on the vertices of a simplex such that each class is at the same distance from every other class. We show that the regularization of the latent space based on our approach yields excellent classification accuracy. Moreover, GCCS provides improved robustness against adversarial perturbations, outperforming models trained with conventional adversarial training (AT). In particular, our model learns a decision space that minimizes the presence of short paths toward neighboring decision regions. We provide a comprehensive empirical evaluation that shows how GCCS outperforms state-of-the-art approaches over challenging datasets for targeted and untargeted gradient-based, as well as gradient-free adversarial attacks, both in terms of classification accuracy and adversarial robustness.

Existing end-to-end signal compression schemes using neural networks are largely based on an autoencoder-like structure, where a universal encoding function creates a compact latent space and the signal representation in this space is quantized and stored. Recently, advances from the field of 3D graphics have shown the possibility of building implicit representation networks, i.e., neural networks returning the value of a signal at a given query coordinate. In this paper, we propose using neural implicit representations as a novel paradigm for signal compression with neural networks, where the compact representation of the signal is defined by the very weights of the network. We discuss how this compression framework works, how to include priors in the design, and highlight interesting connections with transform coding. While the framework is general, and still lacks maturity, we already show very competitive performance on the task of compressing point cloud attributes, which is notoriously challenging due to the irregularity of the domain, but becomes trivial in the proposed framework.

Inverse problems consist in reconstructing signals from incomplete sets of measurements and their performance is highly dependent on the quality of the prior knowledge encoded via regularization. While traditional approaches focus on obtaining a unique solution, an emerging trend considers exploring multiple feasibile solutions. In this paper, we propose a method to generate multiple reconstructions that fit both the measurements and a data-driven prior learned by a generative adversarial network. In particular, we show that, starting from an initial solution, it is possible to find directions in the latent space of the generative model that are null to the forward operator, and thus keep consistency with the measurements, while inducing significant perceptual change. Our exploration approach allows to generate multiple solutions to the inverse problem an order of magnitude faster than existing approaches; we show results on image super-resolution and inpainting problems.

Multi-image super-resolution from multi-temporal satellite acquisitions of a scene has recently enjoyed great success thanks to new deep learning models. In this paper, we go beyond classic image reconstruction at a higher resolution by studying a super-resolved inference problem, namely semantic segmentation at a spatial resolution higher than the one of sensing platform. We expand upon recently proposed models exploiting temporal permutation invariance with a multi-resolution fusion module able to infer the rich semantic information needed by the segmentation task. The model presented in this paper has recently won the AI4EO challenge on Enhanced Sentinel 2 Agriculture.

In this paper we explore the recent topic of point cloud completion, guided by an auxiliary image. We show how it is possible to effectively combine the information from the two modalities in a localized latent space, thus avoiding the need for complex point cloud reconstruction methods from single views used by the state-of-the-art. We also investigate a novel self-supervised setting where the auxiliary image provides a supervisory signal to the training process by using a differentiable renderer on the completed point cloud to measure fidelity in the image space. Experiments show significant improvements over state-of-the-art supervised methods for both unimodal and multimodal completion. We also show the effectiveness of the self-supervised approach which outperforms a number of supervised methods and is competitive with the latest supervised models only exploiting point cloud information.

Point clouds of 3D objects exhibit an inherent compositional nature where simple parts can be assembled into progressively more complex shapes to form whole objects. Explicitly capturing such part-whole hierarchy is a long-sought objective in order to build effective models, but its tree-like nature has made the task elusive. In this paper, we propose to embed the features of a point cloud classifier into the hyperbolic space and explicitly regularize the space to account for the part-whole hierarchy. The hyperbolic space is the only space that can successfully embed the tree-like nature of the hierarchy. This leads to substantial improvements in the performance of state-of-art supervised models for point cloud classification.

Semi-supervised learning techniques are gaining popularity due to their capability of building models that are effective, even when scarce amounts of labeled data are available. In this paper, we present a framework and specific tasks for self-supervised pretraining ofmultichannelmodels, such as the fusion of multispectral and synthetic aperture radar images. We show that the proposed self-supervised approach is highly effective at learning features that correlate with the labels for land cover classification. This is enabled by an explicit design of pretraining tasks which promotes bridging the gaps between sensing modalities and exploiting the spectral characteristics of the input. In a semi-supervised setting, when limited labels are available, using the proposed self-supervised pretraining, followed by supervised finetuning for land cover classification with SAR and multispectral data, outperforms conventional approaches such as purely supervised learning, initialization from training on ImageNet and other recent self-supervised approaches.

Information extraction from synthetic aperture radar (SAR) images is heavily impaired by speckle noise, and hence, despeckling is a crucial preliminary step in scene analysis algorithms. The recent success of deep learning envisions a new generation of despeckling techniques that could outperform classical model-based methods. However, current deep learning approaches to despeckling require supervision for training, whereas clean SAR images are impossible to obtain. In the literature, this issue is tackled by resorting to either synthetically speckled optical images, which exhibit different properties with respect to true SAR images, or multitemporal SAR images, which are difficult to acquire or fuse accurately. In this article, inspired by recent works on blind-spot denoising networks, we propose a self-supervised Bayesian despeckling method. The proposed method is trained by employing only noisy SAR images and can, therefore, learn features of real SAR images rather than synthetic data. Experiments show that the performance of the proposed approach is very close to the supervised training approach on synthetic data and superior on real data in both quantitative and visual assessments.

Recent advances have shown how deep neural networks can be extremely effective at super-resolving remote sensing imagery, starting from a multitemporal collection of low-resolution images. However, existing models have neglected the issue of temporal permutation, whereby the temporal ordering of the input images does not carry any relevant information for the super-resolution task and causes such models to be inefficient with the, often scarce, ground truth data that available for training. Thus, models ought not to learn feature extractors that rely on temporal ordering. In this paper, we show how building a model that is fully invariant to temporal permutation significantly improves performance and data efficiency. Moreover, we study how to quantify the uncertainty of the super-resolved image so that the final user is informed on the local quality of the product. We show how uncertainty correlates with temporal variation in the series, and how quantifying it further improves model performance. Experiments on the Proba-V challenge dataset show significant improvements over the state of the art without the need for self-ensembling, as well as improved data efficiency, reaching the performance of the challenge winner with just 25% of the training data.

In this paper, we present a deep learning model that exploits the power of self-supervision to perform 3D point cloud completion, estimating the missing part and a context region around it. Local and global information are encoded in a combined embedding. A denoising pretext task provides the network with the needed local cues, decoupled from the high-level semantics and naturally shared over multiple classes. On the other hand, contrastive learning maximizes the agreement between variants of the same shape with different missing portions, thus producing a representation which captures the global appearance of the shape. The combined embedding inherits category-agnostic properties from the chosen pretext tasks. Differently from existing approaches, this allows to better generalize the completion properties to new categories unseen at training time. Moreover, while decoding the obtained joint representation, we better blend the reconstructed missing part with the partial shape by paying attention to its known surrounding region and reconstructing this frame as auxiliary objective. Our extensive experiments and detailed ablation on the ShapeNet dataset show the effectiveness of each part of the method with new state of the art results. Our quantitative and qualitative analysis confirms how our approach is able to work on novel categories without relying neither on classification and shape symmetry priors, nor on adversarial training procedures.

In modern society, the ever-increasing demand for Earth Observation products in a large variety of sectors is exposing the limitations of traditional satellite data chain architectures. The European Union Horizon 2020 EO-ALERT project aims at overcoming the existing bottlenecks by leveraging the performance of state-of-the-art commercial off-the-shelf devices to move the critical elements of data processing on the flight segment without sacrificing processing performance. This paper introduces the architecture of the EO-ALERT CPU Scheduling, Compression, Encryption and Data Handling Subsystem, responsible for coordinating the onboard optical and Synthetic Aperture Radar data chains, as well as providing data compression, encryption, and storage services. The performance obtained by a reference implementation of the proposed architecture is also presented, showing an extremely low contribution to the overall system latency that allows real-time Earth Observation product delivery to the end user in less than 5 min.

This paper presents the architecture and results achieved by the EO-ALERT H2020 project. EO-ALERT proposes the definition and development of the next-generation Earth Observation (EO) data processing chain, based on a novel flight segment architecture that moves optimised key EO data processing elements from the ground segment to onboard the satellite, with the aim of delivering the EO products to the end user with very low latency (in almost real-time). This paper presents the EO-ALERT architecture, its performance and hardware. Performances are presented for two reference user scenarios; ship detection and extreme weather nowcasting/monitoring. The hardware testing results show that, when implemented using Commercial Off-The-Shelf (COTS) components and available communication links, the proposed architecture can deliver EO products and alerts to the end user with a latency lower than one-point-five minutes, for both SAR and Optical Very High Resolution (VHR) missions, demonstrating the viability of the EO-ALERT concept and architecture.

Due to a growing interest for imagery with high spatial and spectral resolution, Earth Observation sensors are producing increasing amounts of data. This poses a severe challenge in terms of computational, memory and transmission requirements. In order to overcome these limitations, a fascinating approach is the implementation of a compressive sensing architecture. In this paper, we present an instrumental concept based on the use of a spatial light modulator to implement a super-resolved, compressive demonstrator of an instrument aimed at Earth Observation in the visible and medium infrared spectral regions from geostationary platform.

Transmitting images from earth observation satellites to ground is a major challenge, and a compression/encryption stage is actually mandatory. Development of hardware accelerators is highly recommended, both to relieve the software from such demanding task, and to improve performance, aiming at quasi-real-time data processing. To this end, we discuss the design, development, deployment and test of a FPGA-based accelerator, featuring a lossless and lossy (near-lossless) compression, including the data encryption too. Its architecture is well suited for different image types, including single- and multi-band optical and SAR images and can be fully run-time configurable. Measured performance showed a throughput of 10 Msamples/s, in agreement with related state-of-the-art works, focused on lossless compression only.

Point clouds are an important type of geometric data generated by 3D acquisition devices, and have widespread use in computer graphics and vision. However, learning representations for point clouds is particularly challenging due to their nature as being an unordered collection of points irregularly distributed in 3D space. Recently, supervised and semisupervised problems for point clouds leveraged graph convolution, a generalization of the convolution operation for data defined over graphs. This operation has been shown to be very successful at extracting localized features from point clouds. In this paper, we study the unsupervised problem of a generative model exploiting graph convolution. Employing graph convolution operations in generative models is not straightforward and it poses some unique challenges. In particular, we focus on the generator of a GAN, where the graph is not known in advance as it is the very output of the generator. We show that the proposed architecture can learn to generate the graph and the features simultaneously. We also study the problem of defining an upsampling layer in the graph-convolutional generator, proposing two methods that respectively learn to exploit a multi-resolution or self-similarity prior to sample the data distribution.

Deep learning has shown outstanding performance in several applications including image classification. However, deep classifiers are known to be highly vulnerable to adversarial attacks, in that a minor perturbation of the input can easily lead to an error. Providing robustness to adversarial attacks is a very challenging task especially in problems involving a large number of classes, as it typically comes at the expense of an accuracy decrease. In this work, we propose the Gaussian class-conditional simplex (GCCS) loss: a novel approach for training deep robust multiclass classifiers that provides adversarial robustness while at the same time achieving or even surpassing the classification accuracy of state-of-the-art methods. Differently from other frameworks, the proposed method learns a mapping of the input classes onto target distributions in a latent space such that the classes are linearly separable. Instead of maximizing the likelihood of target labels for individual samples, our objective function pushes the network to produce feature distributions yielding high inter-class separation. The mean values of the distributions are centered on the vertices of a simplex such that each class is at the same distance from every other class. We show that the regularization of the latent space based on our approach yields excellent classification accuracy and inherently provides robustness to multiple adversarial attacks, both targeted and untargeted, outperforming state-of-the-art approaches over challenging datasets.

In this paper we propose a lossless and lossy onboard compression algorithm for multispectral and hyperspectral images, based on the recent CCSDS-123.0-B-2 standard, which takes advantage of cloud screening in order to perform data volume reduction, by avoiding to transmit pixels that are covered by clouds. In particular, we develop methods addressing two problems: i) how to signal the cloud mask in the compressed file, and ii) how to handle cloudy pixels in order to maximize the amount of compression. Experimental results on a set of LANDSAT 8 ETM+ and AVIRIS images show a significant data volume reduction with respect to the plain use of the CCSDS-123.0-B-2 standard.

Deep learning methods for super-resolution of a remote sensing scene from multiple unregistered low-resolution images have recently gained attention thanks to a challenge proposed by the European Space Agency. This paper presents an evolution of the winner of the challenge, showing how incorporating non-local information in a convolutional neural network allows to exploit self-similar patterns that provide enhanced regularization of the super-resolution problem. Experiments on the dataset of the challenge show improved performance over the state-of-the-art, which does not exploit non-local information.

SAR despeckling is a problem of paramount importance in remote sensing, since it represents the first step of many scene analysis algorithms. Recently, deep learning techniques have outperformed classical model-based despeckling algorithms. However, such methods require clean ground truth images for training, thus resorting to synthetically speckled optical images since clean SAR images cannot be acquired. In this paper, inspired by recent works on blind-spot denoising networks, we propose a self-supervised Bayesian despeckling method. The proposed method is trained employing only noisy images and can therefore learn features of real SAR images rather than synthetic data. We show that the performance of the proposed network is very close to the supervised training approach on synthetic data and competitive on real data.

Coronal Mass Ejections (CMEs) are massive releases of plasma from the solar corona. When the charged material is ejected towards the Earth, it can cause geomagnetic storms and severely damage electronic equipment and power grids. Early detection of CMEs is therefore crucial for damage containment. In this paper, we study detection of CMEs from sequential images of the solar corona acquired by a satellite. A low-complexity deep neural network is trained to process the raw images, ideally directly on the satellite, in order to provide early alerts.

Colorization of near-infrared (NIR) images is a challenging problem due to the different material properties at the infared wavelenghts, thus reducing the correlation with visible images. In this paper, we study how graph-convolutional neural networks allow exploiting a more powerful inductive bias than standard CNNs, in the form of non-local self-similiarity. Its impact is evaluated by showing how training with mean squared error only as loss leads to poor results with a standard CNN, while the graph-convolutional network produces significantly sharper and more realistic colorizations.

The increasing amount of data generated by space applications poses several challenges due to limited resources available onboard: power, memory, computation, data rate. In this paper, we propose Compressed Sensing (CS) as the key tool to face those challenges via compressive imaging. This signal processing technique, only recently applied to space applications, dramatically simplifies the image acquisition featuring native compression/encryption and enabling onboard image analysis, allowing to design simpler and lighter optical systems. In this paper, we try to answer the following question: To what extent are the potential benefits of CS going to materialize in a realistic “space big data” application scenario? To this purpose, we first review compressive imaging techniques and already existing prototypes and concepts, critically discussing the technological issues involved. Then, we propose a set of instrument concepts in the application domains of space science, planetary exploration and earth observation, most suitable for a CS-based application. For the most promising of them, we go deeper into the analysis showing preliminary reconstruction performance tests.

Recently, convolutional neural networks (CNNs) have been successfully applied to many remote sensing problems. However, deep learning techniques for multi-image super-resolution (SR) from multitemporal unregistered imagery have received little attention so far. This article proposes a novel CNN-based technique that exploits both spatial and temporal correlations to combine multiple images. This novel framework integrates the spatial registration task directly inside the CNN, and allows one to exploit the representation learning capabilities of the network to enhance registration accuracy. The entire SR process relies on a single CNN with three main stages: shared 2-D convolutions to extract high-dimensional features from the input images; a subnetwork proposing registration filters derived from the high-dimensional feature representations; 3-D convolutions for slow fusion of the features from multiple images. The whole network can be trained end-to-end to recover a single high-resolution image from multiple unregistered low-resolution images. The method presented in this article is the winner of the PROBA-V SR challenge issued by the European Space Agency (ESA).

Compressed sensing (CS) has recently emerged as an effective and efficient way to encrypt data. Under certain conditions, it has been shown to provide some secrecy notions. In theory, it could be considered to be a perfect match for constrained devices needing to acquire and protect the data with computationally cheap operations. However, the theoretical results on the secrecy of compressive cryptosystems only hold under the assumption of infinite precision representation. With this work, we aim to close this gap and lay the theoretical foundations to support this practical framework. We provide theoretical upper bounds on the distinguishability of the measurements acquired through finite precision sensing matrices and experimentally validate them. Our main result is that the secrecy of a CS cryptosystem can be exponentially increased with a linear increase in the representation precision. This result confirms that the CS can be an effective secrecy layer and provides tools to use it in practical settings.

Non-local self-similarity is well-known to be an effective prior for the image denoising problem. However, little work has been done to incorporate it in convolutional neural networks, which surpass non-local model-based methods despite only exploiting local information. In this paper, we propose a novel end-to-end trainable neural network architecture employing layers based on graph convolution operations, thereby creating neurons with non-local receptive fields. The graph convolution operation generalizes the classic convolution to arbitrary graphs. In this work, the graph is dynamically computed from similarities among the hidden features of the network, so that the powerful representation learning capabilities of the network are exploited to uncover self-similar patterns. We introduce a lightweight Edge-Conditioned Convolution which addresses vanishing gradient and over-parameterization issues of this particular graph convolution. Extensive experiments show state-of-the-art performance with improved qualitative and quantitative results on both synthetic Gaussian noise and real noise.

Transforms incorporating directional information are appealing in a wide range of applications. In this paper, we introduce a new framework that allows to define a directional transform starting from any two-dimensional separable transform. The proposed method is highly general and it can be of interest in many areas of signal processing. We also show an example of possible application. We define a directional integer DCT and DST and we show their application in video compression by integrating them in the HEVC video coding standard.

Recovering an image from a noisy observation is a key problem in signal processing. Recently, it has been shown that data-driven approaches employing convolutional neural networks can outperform classical model-based techniques, because they can capture more powerful and discriminative features. However, since these methods are based on convolutional operations, they are only capable of exploiting local similarities without taking into account non-local self-similarities. In this paper we propose a convolutional neural network that employs graph-convolutional layers in order to exploit both local and non-local similarities. The graph-convolutional layers dynamically construct neighborhoods in the feature space to detect latent correlations in the feature maps produced by the hidden layers. The experimental results show that the proposed architecture outperforms classical convolutional neural networks for the denoising task.

We propose a novel architecture for generic biometric authentication based on deep neural networks: RegNet. Differently from other methods, RegNet learns a mapping of the input biometric traits onto a target distribution in a well-behaved space in which users can be separated by means of simple and tunable boundaries. More specifically, authorized and unauthorized users are mapped onto two different and well behaved Gaussian distributions. The novel approach of learning the mapping instead of the boundaries further avoids the problem encountered in typical classifiers for which the learnt boundaries may be complex and difficult to analyze. RegNet achieves high performance in terms of security metrics such as Equal Error Rate (EER), False Acceptance Rate (FAR) and Genuine Acceptance Rate (GAR). The experiments we conducted on publicly available datasets of face and fingerprint confirm the effectiveness of the proposed system.

This paper addresses the problem of training a deep neural network for satellite image segmentation so that it can be deployed over images whose statistics differ from those used for training. For example, in postdisaster damage assessment, the tight time constraints make it impractical to train a network from scratch for each image to be segmented. We propose a convolutional encoder-decoder network able to learn visual representations of increasing semantic level as its depth increases, allowing it to generalize over a wider range of satellite images. Then, we propose two additional methods to improve the network performance over each specific image to be segmented. First, we observe that updating the batch normalization layers' statistics over the target image improves the network performance without human intervention. Second, we show that refining a trained network over a few samples of the image boosts the network performance with minimal human intervention. We evaluate our architecture over three data sets of satellite images, showing the state-of-the-art performance in binary segmentation of previously unseen images and competitive performance with respect to more complex techniques in a multiclass segmentation task.

Compression of hyperspectral images onboard of spacecrafts is a tradeoff between the limited computational resources and the ever-growing spatial and spectral resolution of the optical instruments. As such, it requires low-complexity algorithms with good rate-distortion performance and high throughput. In recent years, the Consultative Committee for Space Data Systems (CCSDS) has focused on lossless and near-lossless compression approaches based on predictive coding, resulting in the recently published CCSDS 123.0-B-2 recommended standard. While the in-loop reconstruction of quantized prediction residuals provides excellent rate-distortion performance for the near-lossless operating mode, it significantly constrains the achievable throughput due to data dependencies. In this paper, we study the performance of a faster method based on the prequantization of the image followed by a lossless predictive compressor. While this is well known to be suboptimal, one can exploit powerful signal models to reconstruct the image at the ground segment, recovering part of the suboptimality. In particular, we show that convolutional neural networks can be used for this task and that they can recover the whole SNR drop incurred at a bit rate of 2 bits per pixel.

In recent years, the Compressed Sensing (CS) framework has been shown to be an effective private key cryptosystem. If infinite precision is available, then it has been shown that spherical secrecy can be achieved. However, despite its theoretically proven secrecy properties, the only practically feasible implementations involve the use of Bernoulli sensing matrices. In this work, we show that different distributions employing a much larger finite alphabet can be considered. More in detail, we consider the use of quantized Gaussian sensing matrices and experimentally show that, besides being suitable for practical implementation, they can achieve higher secrecy with respect to Bernoulli sensing matrices. Furthermore, we show that this approach can be used to tune the secrecy of the CS cryptosystems based on the available machine precision.

Recent advent of graph signal processing (GSP) has spurred intensive studies of signals that live naturally on irregular data kernels described by graphs (e.g., social networks, wireless sensor networks). Though a digital image contains pixels that reside on a regularly sampled 2-D grid, if one can design an appropriate underlying graph connecting pixels with weights that reflect the image structure, then one can interpret the image (or image patch) as a signal on a graph, and apply GSP tools for processing and analysis of the signal in graph spectral domain. In this paper, we overview recent graph spectral techniques in GSP specifically for image/video processing. The topics covered include image compression, image restoration, image filtering, and image segmentation.

A well-known result [1, Lemma 3.4] states that, without noise, it is better to overestimate the support of a sparse signal, since, if the estimated support includes the true support, the reconstruction is perfect. In this paper, we investigate whether this result holds also in the presence of noise. First, we derive the covariance matrix of the signal estimate when the observation matrix is Gaussian, generalizing existing results. Then, we show that, even in the noisy case, overestimating the support length is the preferred solution, as the error incurred by missing some signal components dominates the overall error variance. Finally, an upper bound of the estimated support length is provided to avoid excessive noise amplification.

In this paper, we propose new gradient-based methods for image reconstruction from partial Fourier measurements, which are commonly used in magnetic resonance imaging (MRI) or synthetic aperture radar. Compared to classical gradient recovery methods, a key improvement is obtained by formulating the gradient recovery problem as a compressed sensing problem with the additional constraint that the curl of the gradient field must be zero. Moreover, we formulate the image recovery problem as an inverse problem on graphs. Iteratively reweighted ℓ1recovery methods are proposed to recover these relative differences and the structure of the similarity graph. Finally, the image is recovered from the compressed Fourier measurements using least squares estimation. Numerical experiments demonstrate that the proposed approach outperforms the state-of-the-art image recovery methods.

In this paper, we propose a new method for the recovery of a sparse signal from few linear measurements using a reference signal as side information. Modeling the signal coefficients with a double Laplace mixture model, and assuming that the distribution of the components of the prior information differs slightly from the unknown signal, the problem is formulated as a weighted ℓ1minimization problem. We derive sufficient conditions for perfect recovery and we show that our method is able to reduce significantly the number of measurements required for reconstruction. Numerical experiments demonstrate that the proposed approach outperforms the best algorithms for compressed sensing with prior information and is robust in imperfect scenarios.

We present an Automatic License Plate Recognition system designed around Convolutional Neural Networks (CNNs) and trained over synthetic plate images. We first design CNNs suitable for plate and character detection, sharing a common architecture and training procedure. Then, we generate synthetic images that account for the varying illumination and pose conditions encountered with real plate images and we use exclusively such synthetic images to train our CNNs. Experiments with real vehicle images captured in natural light with commodity imaging systems show precision and recall in excess of 93% despite our networks are trained exclusively on synthetic images.

Fine-grained vehicle classification is a challenging task due to the subtle differences between vehicle classes. Several successful approaches to fine-grained image classification rely on part-based models, where the image is classified according to discriminative object parts. Such approaches require however that parts in the training images be manually annotated, a labor-intensive process. We propose a convolutional architecture realizing a transform network capable of discovering the most discriminative parts of a vehicle at multiple scales. We experimentally show that our architecture outperforms a baseline reference if trained on class labels only, and performs closely to a reference based on a part-model if trained on loose vehicle localization bounding boxes.

Predictive coding is attractive for compression of hyperspectral images onboard of spacecrafts in light of the excellent rate-distortion performance and low complexity of recent schemes. In this letter, we propose a rate control algorithm and integrate it in a lossy extension to the CCSDS-123 lossless compression recommendation. The proposed rate algorithm overhauls our previous scheme by being orders of magnitude faster and simpler to implement, while still providing the same accuracy in terms of output rate and comparable or better image quality.

Directional transforms have recently raised a lot of interest thanks to their numerous applications in signal compression and analysis. In this letter, we introduce a generalization of the discrete Fourier transform (DFT), called steerable DFT (SDFT). Since the DFT is used in numerous fields, it may be of interest in a wide range of applications. Moreover, we also show that the SDFT is highly related to other well-known transforms, such as the Fourier sine and cosine transforms and the Hilbert transforms.

We use some of the largest order statistics of the random projections of a reference signal to construct a binary embedding that is adapted to signals correlated with such signal. The embedding is characterized from the analytical standpoint and shown to provide improved performance on tasks such as classification in a reduced-dimensionality space.

Multifactor user authentication systems enhance security by augmenting passwords with the verification of additional pieces of information such as the possession of a particular device. This paper presents an innovative user authentication scheme that verifies the possession of one’s smartphone by uniquely identifying its camera. High-frequency components of the photo-response nonuniformity of the optical sensor are extracted from raw images and used as a weak physical unclonable function. A novel scheme for efficient transmission and server-side verification is also designed based on adaptive random projections and on an innovative fuzzy extractor using polar codes. The security of the system is thoroughly analyzed under different attack scenarios both theoretically and experimentally.

In image compression, classical block-based separable transforms tend to be inefficient when image blocks contain arbitrarily shaped discontinuities. For this reason, transforms incorporating directional information are an appealing alternative. In this paper, we propose a new approach to this problem, namely, a discrete cosine transform (DCT) that can be steered in any chosen direction. Such transform, called steerable DCT (SDCT), allows to rotate in a flexible way pairs of basis vectors, and enables precise matching of directionality in each image block, achieving improved coding efficiency. The optimal rotation angles for SDCT can be represented as solution of a suitable rate-distortion (RD) problem. We propose iterative methods to search such solution, and we develop a fully fledged image encoder to practically compare our techniques with other competing transforms. Analytical and numerical results prove that SDCT outperforms both DCT and state-of-the-art directional transforms.

In this paper, we introduce new gradient-based methods for image recovery from a small collection of spectral coefficients of the Fourier transform, which is of particular interest for several scanning technologies, such as magnetic resonance imaging. Since gradients of a medical image are much more sparse or compressible than the corresponding image, classical ℓ1-minimization methods have been used to recover these relative differences. The image values can then be obtained by integration algorithms imposing boundary constraints. Compared with classical gradient recovery methods, we propose two new techniques that improve reconstruction. First, we cast the gradient recovery problem as a compressed sensing problem taking into account that the curl of the gradient field should be zero. Second, inspired by the emerging field of signal processing on graphs, we formulate the gradient recovery problem as an inverse problem on graphs. Iteratively reweighted ℓ1recovery methods are proposed to recover these relative differences and the structure of the similarity graph. Once the gradient field is estimated, the image is recovered from the compressed Fourier measurements using least squares estimation. Numerical experiments show that the proposed approach outperforms the state-of-the-art image recovery methods.

In this paper, we propose a method for estimating the sparsity of a signal from its noisy linear projections without recovering it. The method exploits the property that linear projections acquired using a sparse sensing matrix are distributed according to a mixture distribution whose parameters depend on the signal sparsity. Due to the complexity of the exact mixture model, we introduce an approximate two-component Gaussian mixture model whose parameters can be estimated via expectation-maximization techniques. We demonstrate that the above model is accurate in the large system limit for a proper choice of the sensing matrix sparsifying parameter. Moreover, experimental results demonstrate that the method is robust under different signal-to-noise ratios and outperforms existing sparsity estimation techniques.

Iterative shrinkage-thresholding algorithms provide simple methods to recover sparse signals from compressed measurements. In this paper, we propose a new class of iterative shrinkage-thresholding algorithms which preserve the computational simplicity and improve iterative estimation by incorporating a soft support detection. Indeed, at each iteration, by learning the components that are likely to be nonzero from the current signal estimation using Bayesian techniques, the shrinkage-thresholding step is adaptively tuned and optimized. Unlike other adaptive methods, we are able to prove, under suitable conditions, the convergence of the proposed methods. Moreover, we show through numerical experiments that the proposed methods outperform classical shrinkage-thresholding in terms of rate of convergence and of sparsity-undersampling tradeoff.

In image compression, block-based transforms tend to be inefficient when blocks contain arbitrarily shaped discontinuities. For this reason, transforms incorporating directional information are an appealing alternative. Starting from the graph Fourier transform, in this paper we present a new transform, called Subspace-Sparsifying Steer-able DCT, that can be obtained by rotating the basis vectors of the 2D DCT using a set of angles that best matches the block to be encoded. In particular, this new transform chooses the set of angles providing the sparsest image representation in the transform domain, yielding a matrix of transform coefficients that is triangular. In this way, it nearly halves the number of coefficients that need to be transmitted, obtaining a significant coding gain in comparison to the classical DCT.

Recently, distributed algorithms have been proposed for the recovery of sparse signals in networked systems, e.g. wireless sensor networks. Such algorithms allow large networks to operate autonomously without the need of a fusion center, and are very appealing for smart sensing problems employing low-power devices. They exploit local communications, where each node of the network updates its estimates of the sensed signal also based on the correlated information received from neighboring nodes. In the literature, theoretical results and numerical simulations have been presented to prove convergence of such methods to accurate estimates. Their implementation, however, raises some concerns in terms of power consumption due to iterative inter-node communications, data storage, computation capabilities, global synchronization, and faulty communications. On the other hand, despite these potential issues, practical implementations on real sensor networks have not been demonstrated yet. In this paper we fill this gap and describe a successful implementation of a class of randomized, distributed algorithms on a real low-power wireless sensor network test bed with very scarce computational capabilities. We consider a distributed compressed sensing problem and we show how to cope with the issues mentioned above. Our tests on synthetic and real signals show that distributed compressed sensing can successfully operate in a real-world environment.

Predictive lossy compression has been shown to represent a very flexible framework for lossless and lossy onboard compression of multispectral and hyperspectral images with quality and rate control. In this paper, we improve predictive lossy compression in several ways, using a standard issued by the Consultative Committee on Space Data Systems, namely CCSDS-123, as an example of application. First, exploiting the flexibility in the error control process, we propose a constant-signal-to-noise-ratio algorithm that bounds the maximum relative error between each pixel of the reconstructed image and the corresponding pixel of the original image. This is very useful to avoid low-energy areas of the image being affected by large errors. Second, we propose a new rate control algorithm that has very low complexity and provides performance equal to or better than existing work. Third, we investigate several entropy coding schemes that can speed up the hardware implementation of the algorithm and, at the same time, improve coding efficiency. These advances make predictive lossy compression an extremely appealing framework for onboard systems due to its simplicity, flexibility, and coding efficiency.

In this paper, the security of the compressed sensing (CS) framework as a form of data confidentiality is analyzed. Two important properties of one-time random linear measurements acquired using a Gaussian independent identically distributed matrix are outlined: 1) the measurements reveal only the energy of the sensed signal and 2) only the energy of the measurements leaks information about the signal. An important consequence of the above facts is that CS provides information theoretic secrecy in a particular setting. Namely, a simple strategy based on the normalization of the Gaussian measurements achieves, at least in theory, perfect secrecy, enabling the use of CS as an additional security layer in privacy preserving applications. In the generic setting in which CS does not provide information theoretic secrecy, two alternative security notions linked to the difficulty of estimating the energy of the signal and distinguishing equal-energy signals are introduced. Useful bounds on the mean square error of any possible estimator and the probability of error of any possible detector are provided and compared with the simulations. The results indicate that CS is in general not secure according to cryptographic standards, but may provide a useful built-in data obfuscation layer.

Compressed sensing (CS) is a fast and efficient way to obtain compact signal representations. Oftentimes, one wishes to extract some information from the available compressed signal. Since CS signal recovery is typically expensive from a computational point of view, it is inconvenient to first recover the signal and then extract the information. A much more effective approach consists in estimating the information directly from the signal's linear measurements. In this paper, we propose a novel framework for compressive estimation of autoregressive (AR) process parameters based on ad hoc sensing matrix construction. More in detail, we introduce a compressive least square estimator for AR(p) parameters and a specific AR(1) compressive Bayesian estimator. We exploit the proposed techniques to address two important practical problems. The first is compressive covariance estimation for Toeplitz structured covariance matrices where we tackle the problem with a novel parametric approach based on the estimated AR parameters. The second is a block-based compressive imaging system, where we introduce an algorithm that adaptively calculates the number of measurements to be acquired for each block from a set of initial measurements based on its degree of compressibility. We show that the proposed techniques outperform the state-of-the-art methods for these two problems.

In this paper, we propose a mathematical model to estimate the sparsity degree k of exactly k-sparse signals acquired through Compressed Sensing (CS). Our method does not need to recover the signal to estimate its sparsity, and is based on the use of sparse sensing matrices. We exploit this model to propose a CS acquisition system where the number of measurements is calculated on-the-fly depending on the estimated signal sparsity. Experimental results on block-based CS acquisition of black and white images show that the proposed adaptive technique outperforms classical CS acquisition methods where the number of measurements is set a priori.

In this paper, we analyze a new class of iterative re-weighted least squares (IRLS) algorithms and their effectiveness in signal recovery from incomplete and inaccurate linear measurements. These methods can be interpreted as the constrained maximum likelihood estimation under a two-state Gaussian scale mixture assumption on the signal. We show that this class of algorithms, which performs exact recovery in noiseless scenarios under suitable assumptions, is robust even in presence of noise. Moreover these methods outperform classical IRLS for ℓτ-minimization with τ ∈ (0; 1] in terms of accuracy and rate of convergence.

Recently, we demonstrated that random projections can provide an extremely compact representation of a camera fingerprint without significantly affecting the matching performance. In this paper, we propose a new construction that makes random projections of camera fingerprints scale-robust. The proposed method maps the compressed fingerprint of a rescaled image to the compressed fingerprint of the original image, rescaled by the same factor. In this way, fingerprints obtained from rescaled images can be directly matched in the compressed domain, which is much more efficient than existing scale-robust approaches. Experimental results on the publicly available Dresden database show that the proposed technique is robust to a wide range of scale transformations. Moreover, robustness can be further improved by providing reference scales in the database, with a small additional storage cost.

In this work, we propose a new method of graph construction for graph-based image compression. In particular, because of the overhead incurred by graph transmission to the receiver, we focus our attention to develop an efficient method to construct and to code the graph representation of the image. The proposed method employs innovative edge metrics, quantization and prediction techniques, leading to a compact yet high-quality graph, corresponding to a very efficient transform that performs very well on natural as well as piece-wise smooth images. We have tested our method on different images and, compared to the standard DCT, it provides an average quality gain of 1.6 dB.

Block-based compression tends to be inefficient when blocks contain arbitrary shaped discontinuities. Recently, graph-based approaches have been proposed to address this issue, but the cost of transmitting graph topology often overcome the gain of such techniques. In this work we propose a new Superpixel-driven Graph Transform (SDGT) that uses clusters of superpixels, which have the ability to adhere nicely to edges in the image, as coding blocks and computes inside these homogeneously colored regions a graph transform which is shape-adaptive. Doing so, only the borders of the regions and the transform coefficients need to be transmitted, in place of all the structure of the graph. The proposed method is finally compared to DCT and the experimental results show how it is able to outperform DCT both visually and in term of PSNR.

Image retrieval is the process of finding images from a large collection, satisfying a user-specified criterion. Content-based retrieval has been the traditional paradigm, in which one wishes to find images whose content is similar to a query. In this paper we explore a novel criterion for image search, based on forensic principles. We address the problem of retrieving all the photos in a collection that have been acquired by a specific device which is presented to the system as a query. This is an important forensic problem, whose solution could be very useful for detecting improper usage of pictures. We do not rely on metadata such as Exif headers because they can be unavailable, or easily manipulated, and in most cases cannot identify the specific device. We rely instead on a forensic tool called Photo Response Non-Uniformity (PRNU), which constitutes a reliable fingerprint of a camera sensor. We examine recent advances in compression of such fingerprints, which allow to address the previously unexplored image retrieval problem on large scales.

The compression algorithm developed for METIS arises from the standard CCSDS 123.0-r-1 [1], a lossless data compressor suitable for multispectral and hyperspectral imagers and sounders. This paper presents an extension of the standard to handle lossless compression specifically tailored to solar corona images. The main contributions of this paper are the following: the adaptation to the radial geometry of solar corona images through a remapping process called “radialization” and the ability to exploit three-dimensional image compression of the standard through the compression of a multi-temporal data cube composed by successive acquisitions. The development of the algorithm took care of many aspects based on the geometry and the nature of the expected acquisitions; this adaptation process leads to a significant increase in compression performance through routines resulting very light from a computational point of view, which is a key aspect for on-board applications.

In this paper, we consider the problem of RSS-fingerprinting localization in wireless sensor networks. In particular, inspired by the recent advances in sparse approximation and compressive sensing theory, we propose a localization scheme based on the dictionary design of block-sparse signals. We show via numerical simulations and real experiments that the proposed technique outperforms traditional fingerprinting methods.

Block-based separable transforms tend to be inefficient when blocks contain arbitrarily shaped discontinuities. For this reason, transforms incorporating directional information are an appealing alternative. In this paper, we propose a new approach to this problem, designing a new transform that can be steered in any chosen direction and that is defined in a rigorous mathematical way. This new steerable DCT allows to rotate in a flexible way pairs of basis vectors, enabling precise matching of directionality in each image block, and thereby achieving improved coding efficiency. We tested the proposed transform on several images and the results show that it provides a significant performance gain compared to the DCT. Moreover, the mathematical framework on which the steerable DCT is based allows to generalize the transform to more complex steering patterns than a single pure rotation.

The choice for image descriptor in a visual navigation system is not straightforward. Descriptors must be distinctive enough to allow for correct localization while still offering low matching complexity and short descriptor size for real-time applications. MPEG Compact Descriptor for Visual Search is a low complexity image descriptor that offers several levels of compromises between descriptor distinctiveness and size. In this work we describe how these trade-offs can be used for efficient loop-detection in a typical indoor environment.

The research of affine scale space is to create a more general approach to the affine invariant image scale representation by modifying the corresponding Gaussian filters in order to cope with the specific change of view point. It has the purpose to retain a linear relationship with the transiting of the view point. With this linear relationship, the affine scale space could be established as a more general approach for the affine invariant image retrieval, including affine feature detection and affine feature descriptor. The scope of this paper is to discuss the accessible to the affine scale space, its performance and a practical implementation to construct it in order to cope with the high complexity brought in by the scale space and the affine adaptation.

Compressed sensing (CS) is an emerging paradigm for acquisition of compressed representations of a sparse signal. Its low complexity is appealing for resource-constrained scenarios like sensor networks. However, such scenarios are often coupled with unreliable communication channels and providing robust transmission of the acquired data to a receiver is an issue. Multiple description coding (MDC) effectively combats channel losses for systems without feedback, thus raising the interest in developing MDC methods explicitly designed for the CS framework, and exploiting its properties. We propose a method called Graded Quantization (CS-GQ) that leverages the democratic property of compressive measurements to effectively implement MDC, and we provide methods to optimize its performance. A novel decoding algorithm based on the alternating directions method of multipliers is derived to reconstruct signals from a limited number of received descriptions. Simulations are performed to assess the performance of CS-GQ against other methods in presence of packet losses. The proposed method is successful at providing robust coding of CS measurements and outperforms other schemes for the considered test metrics.

Heterogeneous systems on a single chip composed of a central processing unit, graphics processing unit (GPU), and field-programmable gate array (FPGA) are expected to emerge in the near future. In this context, the system on chip can be dynamically adapted to employ different architectures for execution of data-intensive applications. Motion estimation (ME) is one such task that can be accelerated using FPGA and GPU for high-performance H.264/Advanced Video Coding encoder implementation. This paper presents an inherent parallel low-complexity rate-distortion (RD) optimized fast ME algorithm well suited for parallel implementations, eliminating various data dependencies caused by a reliance on spatial predictions. In addition, this paper provides details of the GPU and FPGA implementations of the parallel algorithm by using OpenCL and Very High Speed Integrated Circuits (VHSIC) Hardware Descriptive Language (VHDL), respectively, and presents a practical performance comparison between the two implementations. The experimental results show that the proposed scheme achieves significant speedup on GPU and FPGA, and has comparable RD performance with respect to sequential fast ME algorithm.

The ℓ0/ℓ1-regularized least-squares approach is used to deal with linear inverse problems under sparsity constraints, which arise in mathematical and engineering fields. In particular, multiagent models have recently emerged in this context to describe diverse kinds of networked systems, ranging from medical databases to wireless sensor networks. In this paper, we study methods for solving ℓ0/ℓ1-regularized leastsquares problems in such multiagent systems. We propose a novel class of distributed protocols based on iterative thresholding and input driven consensus techniques, which are well-suited to work in-network when the communication to a central processing unit is not allowed. Estimation is performed by the agents themselves, which typically consist of devices with limited computational capabilities. This motivates us to develop low-complexity and low-memory algorithms that are feasible in real applications. Our main result is a rigorous proof of the convergence of these methods in regular networks. We introduce a suitable distributed, regularized, least-squares functional, and we prove that our algorithms reach their minima using results from dynamical systems theory. Furthermore, we propose numerical comparisons with the alternating direction method of multipliers and the distributed subgradient methods, in terms of performance, complexity, and memory usage. We conclude that our techniques are preferable for their good memory-accuracy tradeoff.

Retrieving pictures from large collections according to a specific criterion is an increasingly relevant task. An important , but so far overlooked, such criterion is the retrieval of pictures acquired by a specific camera. Instead of relying on metadata , which can be absent or easily manipulated, a forensic tool is exploited, namely the photo response non-uniformity (PRNU) of the camera sensor. Recent works showed that random projections can be used to significantly compress the PRNU, enabling operation on very large scales, previously impossible due to the size of the PRNU and to the complexity of the matching operations. In this paper, we propose efficient techniques for management and retrieval of images employing the PRNU, and test them on a database of 1174 cameras and half a million pictures downloaded from the Internet.

In this paper, we analyze the security of compressed sensing (CS) as a cryptosystem. We demonstrate that random linear measurements acquired using a Gaussian i.i.d. matrix reveal only the energy of the sensed signal, and that only the energy of the measurements leaks information about the signal. We provide useful bounds for assessing the information leakage about the energy, linking those bounds to the minimum mean square error achievable by practical estimators. Moreover, we propose a simple strategy based on the normalization of the measurements which achieves, at least in theory, perfect secrecy, enabling the use of CS-based encryption in practical cryptosystems.

A sparse or compressible signal can be recovered from a certain number of noisy random projections, smaller than what dictated by classic Shannon/Nyquist theory. In this paper, we derive the closed-form expression of the mean square error performance of the oracle receiver, knowing the sparsity pattern of the signal. With respect to existing bounds, our result is exact and does not depend on a particular realization of the sensing matrix. Moreover, our result holds irrespective of whether the noise affecting the measurements is white or correlated. Numerical results show a perfect match between equations and simulations, confirming the validity of the result.

In this paper, we address the problem of distributed support detection of multiple sparse signals with common support. Specifically, signals are acquired by the individual nodes of a network according to the so-called Joint Sparsity Model 2 (JSM-2). By leveraging on this model, we propose a distributed scheme for in-network signal recovery, i.e. not requiring data gathering and processing at a fusion center, based on distributed iterative thresholding and consensus strategies. For the proposed scheme, whose convergence properties we rigorously prove, no a priori knowledge on the non-zero number of entries in the signal vector is required.

Compressed Sensing (CS) is suitable for remote acquisition of hyperspectral images for earth observation, since it could exploit the strong spatial and spectral correlations, allowing to simplify the architecture of the onboard sensors. Solutions proposed so far tend to decouple spatial and spectral dimensions to reduce the complexity of the reconstruction, not taking into account that onboard sensors progressively acquire spectral rows rather than acquiring spectral channels. For this reason, we propose a novel progressive CS architecture based on separate sensing of spectral rows and joint reconstruction employing Total Variation. Experimental results run on raw AVIRIS and AIRS images confirm the validity of the proposed system.

In this paper, we present a new recovery algorithm for innetwork compressed sensing from measurements acquired in multi-agent systems. Each agent has to recover a common signal taking advantage of local communication and simple computations. Such distributed problem typically incurs a high energy cost due to inter-node communications. In this paper we propose an iterative distributed algorithm to address this problem, featuring pairwise gossip communications and updates. We propose some theoretical results on its dynamics and numerical comparisons with the most recent approaches proposed in literature. The performance turns out to be competitive in terms of reconstruction accuracy, complexity, and energy consumption required for convergence.

Compressive sensing achieves effective dimensionality reduction of signals, under a sparsity constraint, by means of a small number of random measurements acquired through a sensing matrix. In a signal processing system, the problem arises of processing the random projections directly, without first reconstructing the signal. In this paper, we show that circulant sensing matrices allow to perform a variety of classical signal processing tasks such as filtering, interpolation, registration, transforms, and so forth, directly in the compressed domain and in an exact fashion, i.e., without relying on estimators as proposed in the existing literature. The advantage of the techniques presented in this paper is to enable direct measurement-to-measurement transformations, without the need of costly recovery procedures.

In this paper we present F2OMP, a recovery algorithm for Compressed Sensing over finite fields. Classical recovery algorithms do not exploit the fact that a signal may belong to a finite alphabet, while we show that this information can lead to more efficient reconstruction algorithms. As an application, we use the proposed algorithm to recover sparse grayscale images, showing that performing CS operation over a finite field can outperform classical recovery algorithms from visual quality, memory occupation and complexity point of view.

In this paper we propose an efficient architecture for onboard implementation of rate-controlled predictive lossy compression of hyperspectral and multispectral images. In particular, we consider the recent state-of-the-art rate control algorithm for onboard predictive compression [1], and propose an architecture addressing two fundamental aspects of its hardware implementation. Specifically, this architecture overcomes the serial nature of the algorithm, as well as the large memory requirements of the entropy coding stage, achieving a pipelined implementation suitable for high-throughput onboard implementation, at a negligible cost in terms of coding efficiency.

We consider correlated and distributed sources without cooperation at the encoder. For these sources, we derive the best achievable performance in the rate-distortion sense of any distributed compressed sensing scheme, under the constraint of high-rate quantization. Moreover, under this model we derive a closed-form expression of the rate gain achieved by taking into account the correlation of the sources at the receiver and a closed-form expression of the average performance of the oracle receiver for independent and joint reconstruction. Finally, we show experimentally that the exploitation of the correlation between the sources performs close to optimal and that the only penalty is due to the missing knowledge of the sparsity support as in (non distributed) compressed sensing. Even if the derivation is performed in the large system regime, where signal and system parameters tend to infinity, numerical results show that the equations match simulations for parameter values of practical interest.

Predictive coding is attractive for compression on board of spacecraft due to its low computational complexity, modest memory requirements, and the ability to accurately control quality on a pixel-by-pixel basis. Traditionally, predictive compression focused on the lossless and near-lossless modes of operation, where the maximum error can be bounded but the rate of the compressed image is variable. Rate control is considered a challenging problem for predictive encoders due to the dependencies between quantization and prediction in the feedback loop and the lack of a signal representation that packs the signal's energy into few coefficients. In this paper, we show that it is possible to design a rate control scheme intended for onboard implementation. In particular, we propose a general framework to select quantizers in each spatial and spectral region of an image to achieve the desired target rate while minimizing distortion. The rate control algorithm allows achieving lossy near-lossless compression and any in-between type of compression, e.g., lossy compression with a near-lossless constraint. While this framework is independent of the specific predictor used, in order to show its performance, in this paper, we tailor it to the predictor adopted by the CCSDS-123 lossless compression standard, obtaining an extension that allows performing lossless, near-lossless, and lossy compression in a single package. We show that the rate controller has excellent performance in terms of accuracy in the output rate, rate-distortion characteristics, and is extremely competitive with respect to state-of-the-art transform coding.

A key problem in network coding (NC) lies in the complexity and energy consumption associated with the packet decoding processes, which hinder its application in mobile environments. Controlling and hence limiting such factors has always been an important but elusive research goal, since the packet degree distribution, which is the main factor driving the complexity, is altered in a non-deterministic way by the random recombinations at the network nodes. In this paper we tackle this problem with a new approach and propose Band Codes (BC), a novel class of network codes specifically designed to preserve the packet degree distribution during packet encoding, recombination and decoding. BC are random codes over GF(2) that exhibit low decoding complexity, feature limited and controlled degree distribution by construction, and hence allow to effectively apply NC even in energy-constrained scenarios. In particular, in this paper we motivate and describe our new design and provide a thorough analysis of its performance. We provide numerical simulations of the BC performance in order to validate the analysis and assess the overhead of BC with respect to a conventional random NC scheme. Moreover, experiment in a real-world application, namely peer-to-peer mobile media streaming using a random-push protocol, show that BC reduce the decoding complexity by a factor of two with negligible increase of the encoding overhead, paving the way for the application of NC to power-constrained devices.

Network coding (NC) has been shown to be very effective for collaborative media streaming applications. A pivotal issue in media streaming with NC lies in the packet scheduling policy at the network nodes, which affects the perceived media quality. In this paper, we address the problem of finding the packet scheduling policy that maximizes the number of media segments recovered in the network. We cast this as a distributed minimization problem and propose heuristic solutions that make the proposed framework robust to infrequent or inaccurate feedback information. Moreover, the proposed framework accounts for the properties of layered and multiple description encoded media to provide graceful quality degradation in case of packet losses or lack of upload bandwidth. Experimental results on a local testbed as well as PlanetLab suggest that our scheduling framework achieves better media quality, lower playback delay, and lower bandwidth consumption than a random-push scheme.

This paper presents a VLSI architecture for a low complexity motion estimation algorithm, referred to as Slim264, for multiview video coding extension of H.264. Algorithmic modifications are introduced to obtain a fully parallel computational structure able to meet the throughput requirements of high resolution and high frame rate videos. High parallelism is achieved by predicting small blocks, i.e. 4x4 pixel blocks, in parallel and then adding them up in order to get Sum of Absolute Differences (SADs) of large block sizes. The predictor is able to support high resolution videos i.e. 1080p. The modified algorithm shows promising PSNR results with respect to full search algorithm. The predictor is synthesized with a clock frequency of 200 MHz, occupying an area of 0.49 mm2, on 90-nm Standard Cell ASIC technology.

In this paper, we address the problem of distributed sparse recovery of signals acquired via compressed measurements in a sensor network. We propose a new class of distributed algorithms to solve Lasso regression problems, when the communication to a fusion center is not possible, e.g., due to communication cost or privacy reasons. More precisely, we introduce a distributed iterative soft thresholding algorithm (DISTA) that consists of three steps: an averaging step, a gradient step, and a soft thresholding operation. We prove the convergence of DISTA in networks represented by regular graphs, and we compare it with existing methods in terms of performance, memory, and complexity.

In this work, we propose a feedback scheme to drive the packet recombination process at the network nodes in a collaborative Network Coding (NC) scenario. Our scheme addresses the issue of determining which symbols are more helpful at the receivers to recover the message and how to accordingly recombine the received packets at the intermediate nodes where the original symbols are not available. We experimentally demonstrate that our scheme increases the coding efficiency and reduces the computational complexity at the decoder in a video communication scenario without using explicit feedback messages.

We present a distributed packet scheduling scheme for pushbased Peer-to-Peer (P2P) video streaming with Network Coding (NC) over unstructured random overlays. While previous research has shown the potentials of random-push NC for P2P, little attention has been given to the problem of scheduling the packet transmissions at the network nodes. The proposed scheduling scheme exploits the knowledge of the status of the network links and nodes to maximize the number of nodes that are able to recover the media content prior to its playout deadline. Our experiments show a large performance gain with respect to random-push scheduler in terms of better media quality.

The compressed sensing paradigm allows to efficiently represent sparse signals by means of their linear measurements. However, the problem of transmitting these measurements to a receiver over a channel potentially prone to packet losses has received little attention so far. In this paper, we propose novel methods to generate multiple descriptions from compressed sensing measurements to increase the robustness over unreliable channels. In particular, we exploit the democracy property of compressive measurements to generate descriptions in a simple manner by partitioning the measurement vector and properly allocating bit-rate, outperforming classical methods like the multiple description scalar quantizer. In addition, we propose a modified version of the Basis Pursuit Denoising recovery procedure that is specifically tailored to the proposed methods. Experimental results show significant performance gains with respect to existing methods.

Previous research has shown the benefits of random-push Network Coding (NC) for P2P video streaming. On the other hand, scalable video coding provides graceful quality adaptation to heterogeneous network conditions. Nevertheless, packet scheduling for scalable media streaming with P2P NC is still a largely unexplored problem. Our ongoing research aims at designing a packet scheduling scheme that maximizes the quality of the video with minimal coordination among peers. In this work, we provide a preliminary description of our scheduling scheme and preliminary performance measurements.

In this paper we address the problem of visual quality of images reconstructed from block-wise random projections. Independent reconstruction of the blocks can severely affect visual quality, by displaying artifacts along block borders. We propose a method to enforce smoothness across block borders by modifying the sensing and reconstruction process so as to employ partially overlapping blocks. The proposed algorithm accomplishes this by computing a fast preview from the blocks, whose purpose is twofold. On one hand, it allows to enforce a set of constraints to drive the reconstruction algorithm towards a smooth solution, imposing the similarity of block borders. On the other hand, the preview is used as a predictor of the entire block, allowing to recover the prediction error, only. The quality improvement over the result of independent reconstruction can be easily assessed both visually and in terms of PSNR and SSIM index.

Compressive imaging is an emerging application of compressed sensing, devoted to acquisition, encoding and reconstruction of images using random projections as measurements. In this paper we propose a novel method to provide a scalable encoding of an image acquired by means of compressed sensing techniques. Two bit-streams are generated to provide two distinct quality levels: a low-resolution base layer and full-resolution enhancement layer. In the proposed method we exploit a fast preview of the image at the encoder in order to perform inter-layer prediction and encode the prediction residuals only. The proposed method successfully provides resolution and quality scalability with modest complexity and it provides gains in the quality of the reconstructed images with respect to separate encoding of the quality layers. Remarkably, we also show that the scheme can also provide significant gains with respect to a direct, non-scalable system, thus accomplishing two features at once: scalability and improved reconstruction performance.

A limitation of many compressive imaging architectures lies in the sequential nature of the sensing process, which leads to long sensing times. In this paper we present a novel architecture that uses fewer detectors than the number of reconstructed pixels and is able to acquire the image in a single acquisition. This paves the way for the development of video architectures that acquire several frames per second. We specifically address the diffraction problem, showing that deconvolution normally used to recover diffraction blur can be replaced by convolution of the sensing matrix, and how measurements of a 0/1 physical sensing matrix can be converted to -1/1 compressive sensing matrix without any extra acquisitions. Simulations of our architecture show that the image quality is comparable to that of a classic Compressive Imaging camera, whereas the proposed architecture avoids long acquisition times due to sequential sensing. This one-shot procedure also allows to employ a fixed sensing matrix instead of a complex device such as a Digital Micro Mirror array or Spatial Light Modulator. It also enables imaging at bandwidths where these are not efficient.

We present PISTA, a GPU-accelerated Iterative Soft Thresholding (IST) algorithm for sparse image recovery in Compressive Sensing applications. As the time required to recover an image increases with the number of pixels, GPU-acceleration enables to recover even large images in reasonable time. With respect to equivalent methods, IST-like algorithms have lower computational complexity per-iteration and lower memory requirements, plus the operations are inherently suitable for parallelization. Our experiments show that our algorithm enables a significant reduction in the time required to recover an image even over a highly-optimized CPU-only reference.

Recently, the parallel processing capability of the Graphics Processing Unit (GPU) has been employed for accelerating motion estimation in H.264/AVC encoder implementations. However, while implementing parallel motion estimation on GPU, the bit rate cost of the motion vector is generally ignored. This is due to the unavailability of the spatially predicted motion vectors, which leads to rate-distortion performance degradation. This paper presents a fast parallel motion estimation algorithm implemented on GPU using OpenCL to tackle this problem. The predicted motion vectors are estimated from temporally predicted motion vectors and used for evaluating the bit rate cost of the motion vectors simultaneously. The experimental results show that the proposed scheme achieves significant speedup and has comparable rate-distortion performance with respect to sequential fast motion estimation algorithm.

While every network node only relays messages in a traditional communication system, the recent network coding (NC) paradigm proposes to implement simple in-network processing with packet combinations in the nodes. NC extends the concept of “encoding” a message beyond source coding (for compression) and channel coding (for protection against errors and losses). It has been shown to increase network throughput compared to traditional networks implementation, to reduce delay and to provide robustness to transmission errors and network dynamics. These features are so appealing for multimedia applications that they have spurred a large research effort towards the development of multimedia-specific NC techniques. This paper reviews the recent work in NC for multimedia applications and focuses on the techniques that fill the gap between NC theory and practical applications. It outlines the benefits of NC and presents the open challenges in this area. The paper initially focuses on multimedia-specific aspects of network coding, in particular delay, in-network error control, and media-specific error control. These aspects permit to handle varying network conditions as well as client heterogeneity, which are critical to the design and deployment of multimedia systems. After introducing these general concepts, the paper reviews in detail two applications that lend themselves naturally to NC via the cooperation and broadcast models, namely peer-to-peer multimedia streaming and wireless networking.

Compressed sensing (CS) is an innovative technique allowing to represent signals through a small number of their linear projections. In this paper we address the application of CS to the scenario of progressive acquisition of 2D visual signals in a line-by-line fashion. This is an important setting which encompasses diverse systems such as flatbed scanners and remote sensing imagers. The use of CS in such setting raises the problem of reconstructing a very high number of samples, as are contained in an image, from their linear projections. Conventional reconstruction algorithms, whose complexity is cubic in the number of samples, are computationally intractable. In this paper we develop an iterative reconstruction algorithm that reconstructs an image by iteratively estimating a row, and correlating adjacent rows by means of linear prediction. We develop suitable predictors and test the proposed algorithm in the context of flatbed scanners and remote sensing imaging systems. We show that this approach can significantly improve the results of separate reconstruction of each row, providing very good reconstruction quality with reasonable complexity.

We present Band Codes (BC), a novel class of rate less codes that makes possible to control the computational complexity of Network Coding (NC). NC increases throughput of the networks via packet recombinations at the network nodes. In a NC scenario based on rate less codes, the recombinations at the nodes alter the packet degree distribution selected at the source and increase the computational complexity of the packet decoding process. Unlike other classes of rate less codes, BC preserve the degree distribution of the encoded packets through the recombinations at the nodes. Furthermore, BC enable to control the decoding complexity of each network node independently from the rest of the network. We evaluate BC in a P2P scenario using a purposely designed random-push protocol for live video streaming. The experiments show that BC achieve high encoding efficiency, enable nodes with different computational capabilities to coexist within the same network and reduce the processor load on a real mobile device by nearly 50%.

The adoption of distributed databases calls for storing data at two or more sites, in order to address application-specific requirements including, e.g., redundancy, data locality, and so on. When visual data (including images, videos and their corresponding descriptors) need to be stored, synchronization across different sites might require significant bandwidth resources. In this paper we explore the use of distributed source coding to encode local SIFT descriptors extracted from static images. The key tenet is to exploit, at the decoder side, the correlation between matching pairs of descriptors extracted, respectively, from the out-of-date and up-to-date image. Preliminary results show that a coding efficiency gain up to 1 bit/descriptor can be achieved in the case of ideal lossless coding. In the case of distributed source coding with LDPC codes, a practical average gain of 0.19 bit/descriptor is observed.

We introduce randomized multicast trees (RMT), an overlay topology designed for low-delay media streaming using network coding. RMTs improve on tree-based overlays in terms of start-up delay. We develop a push-based streaming system that leverages network coding to efficiently distribute the information in the overlay without using buffer maps, followed by a short pull stage to recover from packet losses, and appropriate management procedures to handle ungraceful peers departures.We report performance results of the proposed system, and compare it with an optimized pull system, and with an existing peer-to-peer system employing network coding, showing a significant performance improvement in terms of delay and resiliency to peers’ dynamics and packet losses.

Lossy compression of hyperspectral and ultraspectral images is traditionally performed using 3D transform coding. This approach yields good performance, but its complexity and memory requirements are unsuitable for onboard compression. In this paper we propose a low-complexity lossy compression scheme based on prediction, uniform threshold quantization, and rate-distortion optimization. Its performance is competitive with that of state-of-the-art 3D transform coding schemes, but the complexity is immensely lower. The algorithm is able to limit the scope of errors, and is amenable to parallel implementation, making it suitable for onboard compression at high throughputs.

We present a novel architecture for complexity-adaptive Random Network Coding (RNC) and its application to Peer-to-Peer (P2P) video streaming. Network coding enables the design of simple and effective P2P video distribution systems, however it relies on computationally intensive packet coding operations that may exceed the computational capabilities of power constrained devices. It is hence desirable that the complexity of network coding can be adjusted at every node according to its computational capabilities, so that different classes of nodes can coexist in the network. To this end, we model the computational complexity of network coding as the sum of a packet decoding cost, which is centrally minimized at the encoder, and a packet recoding cost, which is locally controlled by each node. Efficient network coding is achieved exploiting the packet decoding process as a packet pre-recoding stage, hence increasing the chance that transmitted packets are innovative without increasing the recoding cost. Experiments in a P2P video streaming framework show that the proposed design enables the nodes of the network to operate at a wide range of computational complexity levels, while a higher number of low complexity nodes are able to join the network and experience high-quality video.

We propose a novel standard-compliant multiple description coding (MDC) method that exploits the H.264/AVC redundant slice tool, performing rate-distortion optimization at the slice level. The strategy to allocate redundancy to each slice jointly takes into account its contribution to distortion, its position in the GOP, the effect of decoder error concealment, and the transmission conditions. This makes the algorithm more accurate with respect to previous frame-based solutions, and experimental results show that it compares favorably with other state-of-the-art standard-compliant MDC techniques.

Lossy compression of hyperspectral and ultraspectral images is traditionally performed using 3D transform coding. This approach yields good performance, but the complexity and memory requirements make it unsuitable for onboard compression. In this paper we propose a low-complexity lossy compression scheme based on prediction, quantization and rate-distortion optimization. The scheme employs coset codes coupled with the newconcept of “informed quantization”, and requires no entropy coding. The performance of the resulting algorithm is competitive with that of state-of-the-art 3D transform coding schemes, but the complexity is immensely lower, making it suitable for onboard compression at high throughputs.

We address distributed source coding with decoder side information, when the decoder observes the source through a noisy channel. Existing approaches employ syndromeor parity-based channel codes. We propose a new approach based on distributed arithmetic coding (DAC).We introduce a DAC with forbidden symbol, which allows to tune the redundancy according to the amount of channel noise. We propose a novel sequential decoder that employs the known side information to decode the corrupted codeword. Experimental results show that the proposed scheme is better than parity-based turbo codes at relatively short block lengths.

Streaming on-demand video services can be provided to an end user by transmitting video data as a sequence of Internet Protocol (IP) packets over the network. In order to maintain a sufficient video quality at the end user, video packets must be protected against erasures by means of a suitable form of error control. In this contribution we make a comparison of 2 classes of Forward Error Correction (FEC) codes: the Reed-Solomon (RS) codes and the Raptor codes. We present the decoding complexity analysis of these codes and compare their erasure decoding performance. For a given decoding complexity, the codewords of the Raptor code can be taken much longer than those of the RS code, because the decoding of the former code involves only XOR operations. For a performance target of less than one decoding error event in 4 hours, a video bitrate of 7.3 Mbit/s, a latency constraint of 10 seconds and a transmission overhead of 20%, equal erasure probabilities at the decoder input are allowed for the RS(254,212) code and for the Raptor code of size (8760,7300), whereas this RS code has a 12 times higher decoding complexity.

Network coding is known to provide increased throughput and reduced delay for communications over networks. In this paper we propose a peer-to-peer video streaming system that exploits network coding in order to achieve low start-up delay, high streaming rate, and high resiliency to peers' dynamics. In particular, we introduce the concept of random multicast trees as overlay topology. This topology offers all benefits of tree-based overlays, notably a short start-up delay, but is much more efficient at distributing data and recovering from ungraceful peers departures. We develop a push-based streaming system that leverages network coding to efficiently distribute the information in the overlay without using buffer maps. We show performance results of the proposed system and compare it with an optimized pull systems based on Coolstreaming, showing significant improvement.

In this paper, we propose a lossless compression algorithm for hyperspectral images inspired by the distributed-source-coding (DSC) principle. DSC refers to separate compression and joint decoding of correlated sources, which are taken as adjacent bands of a hyperspectral image. This concept is used to design a compression scheme that provides error resilience, very low complexity, and good compression performance. These features are obtained employing scalar coset codes to encode the current band at a rate that depends on its correlation with the previous band, without encoding the prediction error. Iterative decoding employs the decoded version of the previous band as side information and uses a cyclic redundancy code to verify correct reconstruction. We develop three algorithms based on this paradigm, which provide different tradeoffs between compression performance, error resilience, and complexity. Their performance is evaluated on raw and calibrated AVIRIS images and compared with several existing algorithms. Preliminary results of a field-programmable gate array implementation are also provided, which show that the proposed algorithms can sustain an extremely high throughput.

Digital fountain codes have emerged as a low-complexity alternative to Reed-Solomon codes for erasure correction. The applications of these codes are relevant especially in the field of wireless video, where low encoding and decoding complexity is crucial. In this paper, we introduce a new class of digital fountain codes based on a sliding-window approach applied to Raptor codes. These codes have several properties useful for video applications, and provide better performance than classical digital fountains. Then, we propose an application of sliding-window Raptor codes to wireless video broadcasting using scalable video coding. The rates of the base and enhancement layers, as well as the number of coded packets generated for each layer, are optimized so as to yield the best possible expected quality at the receiver side, and providing unequal loss protection to the different layers according to their importance. The proposed system has been validated in a UMTS broadcast scenario, showing that it improves the end-to-end quality, and is robust towards fluctuations in the packet loss rate.

The objective of this paper is to survey recent developments of network coding, with specific focus on multimedia streaming. Network coding allows nodes to create and forward ldquocombinationsrdquo of incoming messages, which has been shown to increase throughput. While network coding has been invented in the information theory field, its potential benefits are spurring new research on its multimedia applications. We first review the concept of network coding, and briefly describe its potential benefits from a multimedia communication perspective. Then, we discuss the specific issues imposed by media delay constraints on network coding algorithms. Finally, we review some recent works that develop network coding principles in media streaming applications.

SEACAST is a peer-to-peer live streaming protocol developed at Politecnico di Torino, which aims at improving current systems in two key areas. The first is the use of fullfledged flow control using RTP/UDP and session signaling. The second is the use of multiple description coding to handle error resilience and user heterogeneity. In this paper we overview SEACAST, highlighting its main innovations, and providing a short summary of performance evaluation over a local testbed at Politecnico di Torino. The results show a definite performance improvement with respect to existing systems, and point out the usefulness of multiple description coding in the peer-to-peer context.

Hyperspectral images exhibit significant spectral correlation, whose exploitation is crucial for compression. In this paper, we investigate the problem of predicting a given band of a hyperspectral image using more than one previous band. We present an information-theoretic analysis based on the concept of conditional entropy, which is used to assess the available amount of correlation and the potential compression gain. Then, we propose a new lossless compression algorithm that employs a Kalman filter in the prediction stage. Simulation results are presented on Airborne Visible Infrared Imaging Spectrometer, Hyperspectral Digital Imagery Collection Experiment, and Hyperspectral Mapper scenes, showing competitive performance with other state-of-the-art compression algorithms.

The H.264/SVC video codec extends the H.264/AVC standard with scalability features. In this paper we introduce a traffic prioritization algorithm suitable for the transmission of both H.264/SVC and H.264/AVC video over 802.11e ad hoc wireless networks. The proposed algorithm exploits the traffic prioritization capabilities offered by 802.11e to provide better protection to the most perceptually important parts of a video while achieving efficient network resource usage. We evaluate the algorithm by simulating video transmissions in an ad hoc network scenario. Results show that the H.264/SVC codec particularly benefits from the proposed algorithm, which enables a graceful video quality degradation in congested network conditions, as well as PSNR gains up to 2 dB with respect to the H.264/AVC codec using the same amount of network resources.

We propose a distributed source coding system for data collected by sensor networks. It uses a feedback channel between the sensors and the gateway node (i.e., the joint decoder) but, unlike previous systems, the encoding process is driven by the decoder. Compression is performed using distributed arithmetic coding, which is extended to adaptively estimate the source probabilities. Specifically, the decoder estimates marginal and conditional probabilities, and sends them back to the sensors to drive the distributed arithmetic coding process. This reduces the decoding delay, and potentially eliminates the need of rate-compatible Slepian-Wolf codes.

We propose a compression algorithm for hyperspectral images featuring both lossy and near-lossless compression. The algorithm is based on JPEG 2000 and provides better near-lossless compression performance than 3D-CALIC. We also show that its effect on the results of selected applications is negligible and, in some cases, better than JPEG 2000.

In this paper a novel conditional access scheme for the distribution of H.264/AVC video is presented. The algorithm permits to cypher the full quality video, while guaranteeing free access to a reduced quality video that can be used as preview. The scheme is based on the novel coding options, introduced in the H.264 standard, and therefore is fully compliant with the standard syntax. The secured video stream exhibits a very small rate overhead and requires limited supplementary computational cost at coding time.

Digital fountain codes are becoming increasingly important for multimedia communications over networks subject to packet erasures. These codes have significantly lower complexity than Reed-Solomon ones, exhibit high erasure correction performance, and are very well suited to generating multiple equally important descriptions of a source. In this paper we propose an innovative scheme for streaming multimedia contents by using digital fountain codes applied over sliding windows, along with a suitably modified belief-propagation decoder. The use of overlapped windows allows one to have a virtually extended block, which yields superior performance in terms of packet recovery. Simulation results using LT codes show that the proposed algorithm has better performance in terms of efficiency, reliability and memory with respect to fixed-window encoding.

We propose a new scheme for symmetric Slepian-Wolf coding of correlated binary sources. Unlike previous designs that employ capacity-achieving channel codes, the proposed scheme is based on arithmetic codes with error correction capability. We define a time-sharing version of a distributed arithmetic coder, and a soft joint decoder. Experimental results on two sources show that, for short block length, the proposed scheme outperforms the symmetric turbo code design in (Stankovic et al., 2006).

We propose a new lossy compression algorithm for hyperspectral images, which is based on the spectral Karhunen-Loeve transform, followed by spatial JPEG 2000, which employs a model of anomalous pixels during the compression process. Results on Airborne Visible/Infrared Imaging Spectrometer scenes show that the new algorithm provides better rate-distortion performance, as well as improved anomaly detection performance, with respect to the state of the art.

Transform-based lossy compression has a huge potential for hyperspectral data reduction. Hyperspectral data are 3-D, and the nature of their correlation is different in each dimension. This calls for a careful design of the 3-D transform to be used for compression. In this paper, we investigate the transform design and rate allocation stage for lossy compression of hyperspectral data. First, we select a set of 3-D transforms, obtained by combining in various ways wavelets, wavelet packets, the discrete cosine transform, and the Karhunen–Loève transform (KLT), and evaluate the coding efficiency of these combinations. Second, we propose a low-complexity version of the KLT, in which complexity and performance can be balanced in a scalable way, allowing one to design the transform that better matches a specific application. Third, we integrate this, as well as other existing transforms, in the framework of Part 2 of the Joint Photographic Experts Group (JPEG) 2000 standard, taking advantage of the high coding efficiency of JPEG 2000, and exploiting the interoperability of an international standard. We introduce an evaluation framework based on both reconstruction fidelity and impact on image exploitation, and evaluate the proposed algorithm by applying this framework to AVIRIS scenes. It is shown that the scheme based on the proposed low-complexity KLT significantly outperforms previous schemes as to rate-distortion performance. As for impact on exploitation, we consider multiclass hard classification, spectral unmixing, binary classification, and anomaly detection as benchmark applications.

In this paper we address the problem of providing conditional access to video sequences, namely, to generate a low-quality video to be used as preview, which can be decoded at full quality if a decryption key is obtained. We propose and investigate the performance of two different techniques, based on smoothing and separate encoding in the compressed domain, and motion vector perturbation. We show that these techniques are able to provide conditional access to different quality levels of H.264/AVC video with very small rate overhead, and that their combination can provide different levels of security towards malicious attacks

In this letter we propose a new technique for progressive coding of hyperspectral data. Specifically, we employ a hybrid three-dimensional wavelet transform for spectral and spatial decorrelation in the framework of Part 2 of the JPEG 2000 standard. Both onboard and on-the-ground compression are addressed. The resulting technique is compliant with the JPEG 2000 family of standards and provides competitive performance with respect to state-of-the-art techniques.

JPEG 2000 is the novel ISO standard for image and video coding. Besides its improved coding efficiency, it also provides a few error resilience tools in order to limit the effect of errors in the codestream, which can occur when the compressed image or video data are transmitted over an error-prone channel, as typically occurs in wireless communication scenarios. However, for very harsh channels, these tools often do not provide an adequate degree of error protection. In this paper, we propose a novel error-resilience tool for JPEG 2000, based on the concept of ternary arithmetic coders employing a forbidden symbol. Such coders introduce a controlled degree of redundancy during the encoding process, which can be exploited at the decoder side in order to detect and correct errors. We propose a maximum likelihood and a maximum a posteriori context-based decoder, specifically tailored to the JPEG 2000 arithmetic coder, which are able to carry out both hard and soft decoding of a corrupted code-stream. The proposed decoder extends the JPEG 2000 capabilities in error-prone scenarios, without violating the standard syntax. Extensive simulations on video sequences show that the proposed decoders largely outperform the standard in terms of PSNR and visual quality.

We propose a novel multimedia security framework based on a modification of the arithmetic coder, which is used by most international image and video coding standards as entropy coding stage. In particular, we introduce a randomized arithmetic coding paradigm, which achieves encryption by inserting some randomization in the arithmetic coding procedure; notably, and unlike previous works on encryption by arithmetic coding, this is done at no expense in terms of coding efficiency. The proposed technique can be applied to any multimedia coder employing arithmetic coding; in this paper we describe an implementation tailored to the JPEG 2000 standard. The proposed approach turns out to be robust towards attempts to estimating the image or discovering the key, and allows very flexible protection procedures at the code-block level, allowing to perform total and selective encryption, as well as conditional access

The novel H.264/AVC video coding specification provides a significant improvement in terms of coding efficiency compared to previous standards. In certain streaming scenarios, the problem of whole-frame loss concealment can arise when using H.264/AVC. Two different algorithms have been proposed to conceal a lost image, based on the optical flow concept. In a real streaming scenario, an encoder usually introduces some kind of intra refresh policy that can insert trouble concealment algorithms leaving gaps in the motion vector field used for reference. We discuss the advantages of introducing a motion vector estimator for such intra coded regions.

In remote sensing systems, on-board data compression is a crucial task that has to be carried out with limited computational resources. In this paper we propose a novel lossless compression scheme for multispectral and hyperspectral images, which combines low encoding complexity and high-performance. The encoder is based on distributed source coding concepts, and employs Slepian-Wolf coding of the bitplanes of the CALIC prediction errors to achieve improved performance. Experimental results on AVIRIS data show that the proposed scheme exhibits performance similar to CALIC, and significantly better than JPEG 2000.

A first attempt to exploit distributed source coding (DSC) principles for the lossless compression of hyperspectral images is presented. The DSC paradigm is exploited to design a very light coder which minimizes the exploitation of the correlation between the image bands. In this way we managed to move the computational complexity from the encoder to the decoder, thus matching the needs of classical acquisition system where compression is achieved on board of the aerial platform and decoding at the ground station. Though the encoder does not explicitly exploit inter-band correlation, the achieved bit rate is about 1 bit/pixel lower than classical 2D schemes such as JPEG-LS or CALID 2D, and only about 1 b/p higher than the best performing, and much more complex, 3D schemes.

Hyperspectral image compression has recently attracted a remarkable interest for remote sensing applications. In this paper we propose a unified embedded lossy-to-lossless compression framework based on the JPEG 2000 standard. In particular, we exploit the multicomponent transformation feature of Part 2 of JPEG 2000 to devise a compression framework based on a spectral decorrelating transform followed by JPEG 2000 compression of the transformed coefficients. We evaluate several possible choices for the spectral transform, including a floating-point DCT, an integer DCT, and a wavelet transform. The final version of the proposed algorithm has been compared to 3D-SPIHT in the lossy case, and to several state-of-the-art compression algorithms including JPEG-LS and 3D-CALIC in the lossless case. Experimental results on AVIRIS data show that the proposed technique exhibits very competitive performance for both reversible and irreversible compression, with significantly lower complexity than DPCM-based methods, and memory requirements compatible with typical onboard processing subsystems of remote sensing platforms.

In this paper a novel scalable video coder, based on the principle of distributed source coding with side information is proposed. Coding scalability is achieved by means of bitplane coding in the wavelet domain. The distributed coding paradigm is applied to encode the wavelet coefficients of a given frame, by considering those of the previous frame as side information. LDPC syndrome encoding with proper context modeling of the frame correlation allowed us to significantly outperform intra coding obtained with JPEG 2000. Moreover, the proposed approach permits to perform motion compensation at the decoder side, thus opening a new perspective in the field of scalable video coding

In this paper, we propose an efficient method for the allocation of Reed-Solomon codes to source symbols, for unequal loss protection. The proposed formulation recasts the multivariate optimization problem into a univariate one, dramatically reducing the computational complexity. Results are shown for image transmission over lossy packet networks, employing the JPEG2000 and SPIHT encoders. The proposed algorithm exhibits performance equivalent to previous methods, while providing a significant complexity reduction.

In low bit-rate packet-based video communications, video frames may have very small size, so that each frame fills the payload of a single network packet; thus, packet losses correspond to whole-frame losses, to which the existing error concealment algorithms are badly suited and generally not applicable. In this paper, we deal with the problem of concealment of whole frame-losses, and propose a novel technique which is capable of handling this very critical case. The proposed technique presents other two major innovations with respect to the state-of-the-art: i) it is based on optical flow estimation applied to error concealment and ii) it performs multiframe estimation, thus optimally exploiting the multiple reference frame buffer featured by the most modern video coders such as H.263+ and H.264. If data partitioning is employed, by e.g., sending headers, motion vectors, and coding modes in prioritized packets as can be done in the DiffServ network model, the algorithm is capable of exploiting the motion vectors to improve the error concealment results. The algorithm has been embedded in the H.264 test model software, and tested under both independent and correlated packet loss models with parameters typical of the wireless environment. Results show that the proposed algorithm significantly outperforms other techniques by several dBs in peak signal-to-noise ratio (PSNR), provides good visual quality, and has a rather low complexity, which makes it possible to perform real-time operation with reasonable computational resources.

In this paper a novel error resilient MQ coder for reliable JPEG 2000 image delivery is designed. The proposed coder uses a forbidden symbol in order to force a given amount of redundancy in the codestream. At the decoder side, the presence of the forbidden symbol allows for powerful error correction. Moreover the added redundancy can be easily controlled and the proposed coder is kept backward compatible with MQ. In this work excellent improvements in the case of image transmission across both BSC and AWGN channels are obtained by means of a maximum a posteriori estimation technique.

A new error resilience tool is proposed for robust JPEG 2000 imaging over noisy channels. In particular, a modified encoder, based on an MQ arithmetic coder with forbidden symbol, is introduced, along with a maximum likelihood error-correcting MQ decoder. The proposed technique features error detection, error concealment and error correction capability, thus adding new useful functionalities to JPEG 2000. Experimental results show that this technique largely outperforms the standard JPEG 2000 error resilience tools for error concealment and hard/soft channel decoding.

The application of digital watermarking to remote sensing images requires a careful quality assessment in order to understand how the data quality is affected by the watermark. We propose a watermarking approach that minimizes the watermark impact on image classification, based on the idea of modulating the insertion coefficient in each channel so as to preserve to a larger extent the channels which classification is most sensitive to; we also propose a simplified procedure for estimating cluster displacement due to watermarking, leading to a low-complexity insertion approach. Experimental results on Landsat 7 ETM+ and IKONOS images show that the proposed MIC approach is able to significantly reduce classification errors, and to keep them within the intrinsic classification error

In this paper, we propose a joint despeckling and edge detection algorithm based on the Mumford-Shah functional, which accomplishes the image filtering and segmentation as a result of an analytical variational problem. This approach turns out to be well suited to jointly despeckle and segment SAR image data; the experimental results demonstrate that the proposed technique yields high quality despeckling without impairing critical image features and with the additional advantage to provide a detailed edge map

In this paper we present a new watermarking algorithm for joint near-lossless compression and authentication of remote sensing images. The adopted compression algorithm is the standard JPEG-LS algorithm. Our methodology has been designed by integrating into the standard JPEG-LS compression algorithm, by means of a stripe approach, a known authentication technique derived from Fridrich. This procedure points out two advantages: firstly, the produced bit-stream is perfectly compliant with the JPEG-LS standard, secondly, when the image has been decoded, it is always authenticated because information has been embedded in the reconstructed values. Near-lossless coding does not harm authentication procedure and robustness against different attacks is preserved

We describe a novel multimedia security framework based on a modification of the arithmetic coder, which is used by most international image and video coding standards as entropy coding stage. In particular, we propose a randomized arithmetic coding paradigm, which achieves encryption by randomly swapping the intervals of the least and most probable symbols in arithmetic coding; moreover, we describe an implementation tailored to the JPEG 2000 standard. The proposed approach turns out to be robust towards attempts to discover the key, and allows very flexible procedures for insertion of redundancy at the codeblock level, allowing to perform total and selective encryption, conditional access, and encryption of regions of interest.

We propose a new lossless and near-lossless compression algorithm for hyperspectral images based on context-based adaptive lossless image coding (CALIC). Specifically, we propose a novel multiband spectral predictor, along with optimized model parameters and optimization thresholds. The resulting algorithm is suitable for compression of data in band-interleaved-by-line format; its performance evaluation on Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) data shows that it outperforms 3-D-CALIC as well as other state-of-the-art compression algorithms.

We present hybrid loss protection as a new channel coding and packetization scheme for image transmission over nonprioritized lossy packet networks. The scheme employs an interleaver-based structure, and attempts to maximize the expected peak signal-to-noise ratio (PSNR) at the receiver given the constraint that the probability of failure, i.e., the probability that the PSNR of the decoded image is below a given threshold, is upper-bounded by a user-defined value. A new code-allocation algorithm is proposed, which employs Gilbert-Elliot modeling of the network statistics. Experimental results are provided in the case of transmission of images encoded by SPIHT and JPEG 2000 over a wireline, as well as a wireless UMTS-based Internet connection.

We propose a spatio-temporal error concealment algorithm for video transmission in an error-prone environment. The proposed technique employs motion vector estimation, edge-preserving interpolation, and texture analysis/synthesis. It has two main advantages with respect to existing methods, namely: (i) it aims at optimizing the visual quality of the restored video, and not only PSNR; and (ii) it employs an automatic mode selection algorithm in order to decide, on a macroblock basis, whether to use the spatial restoration, the temporal one, or a combination thereof. The algorithm has been applied to H.26L video, providing satisfactory performance over a large set of operating conditions.

We propose a reduced complexity technique for the rate-distortion optimization in JVT/H.26L in the presence of packet erasures. It is named "few decoders in the encoder", and is based on the idea of generating a selected number of error patterns in the encoder, so that a limited number of co-decoding processes can be implemented to estimate the transmission distortion term. The correlation amongst packet erasures is taken into account by employing a binary Gilbert model. The proposed algorithm exhibits competitive performance in terms of average PSNR and probability of decoding failure, with very affordable complexity and memory requirements.

A known problem in video streaming is that loss of a packet usually results into loss of a whole video frame. In this paper we propose an error concealment algorithm specifically designed to handle this sort of losses. The technique exploits information in a few past frames (namely the motion vectors) in order to estimate the forward motion vectors of the last received frame. This information is used to project the last frame onto an estimate of the missing frame. The algorithm has been tested on MPEG-2 video, providing very satisfactory results, and outperforming by several dBs in PSNR the concealment technique based on repetition of the last received frame.

We proposed a spatio-temporal error concealment algorithm for video transmission in an error-prone environment. The proposed technique employs motion vector estimation, edge-preserving interpolation, and texture analysis/synthesis. It has two main advantages with respect to existing methods, namely: i) it aims at optimizing the visual quality of the restored video, and not only PSNR; and ii) it employs an automatic mode selection algorithm in order to decide, on a macroblock basis, whether to use the spatial restoration, the temporal one, or a combination thereof. The algorithm has been applied to H26L video, providing satisfactory performance over a large set of operating conditions.

In this paper we study the problem of transmitting coded video over a UMTS network. We first discuss the statistical characteristics of packet losses in case of RTP/UDP/IP wireless video communication. Then, we propose a new rate allocation algorithm for H.264 video, based on a Gilbert-Elliot model of the packet losses. We compare the proposed algorithm with several other allocation strategies, showing that it achieves satisfactory performance in terms of PSNR. Moreover, we outline the limits of the employed distortion model and outline possible solutions to overcome them.

We study the problem of compression of videosurveillance sequences collected by a wireless sensor network. In particular, we propose a low-complexity coding framework based on change detection and JPEG-like compression of regions of interest, along with a suitable low-complexity change detection algorithm. We show that on typical videosurveillance sequences the performance of the proposed compression algorithm is similar to that of MPEG-2, at a much less computational cost. Energy profiling results on a TMS320VC5204 board validate the coder design for the proposed application.

Compression technology is becoming increasingly important in geospatial information systems. In this paper we address some of the most relevant compression issues for remote sensing applications, and highlight the potential benefits of the JPEG set of standards. In particular, we review the JPEG, JPEG 2000, and JPEG-LS compression standards, and the JPIP protocol for interactive image retrieval. Finally, we discuss the use of compressed-domain processing, along with the use of flexible file formats for efficient storage and access to metadata.

In this paper, we propose to employ predictive coding for lossy compression of synthetic aperture radar (SAR) raw data. We exploit the known result that a blockwise normalized SAR raw signal is a Gaussian stationary process in order to design an optimal decorrelator for this signal. We show that, due to the statistical properties of the SAR signal, an along-range linear predictor with few taps is able to effectively capture most of the raw signal correlation. The proposed predictive coding algorithm, which performs quantization of the prediction error, optionally followed by entropy coding, exhibits a number of advantages, and notably an interesting performance/complexity trade-off, with respect to other techniques such as flexible block adaptive quantization (FBAQ) or methods based on transform-coding; fractional output bit-rates can also be achieved in the entropy-constrained mode. Simulation results on real-world SIR-C/X-SAR as well as simulated raw and image data show that the proposed algorithm outperforms FBAQ as to SNR, at a computational cost compatible with modern SAR systems.

In this paper we propose a novel error concealment algorithm for video transmission over wireless networks potentially subject to packet erasures. In particular, we develop a technique for the replenishment of missing macroblocks, which aims at minimizing the impact of the lost data on the resulting video with respect to the human visual system. The proposed algorithm operates three reconstruction stages at different scales, by first recovering smooth large-scale patterns, then large-scale structures, and finally local edges in the lost macroblock. Experimental results show that the proposed algorithm achieves improved visual quality of the reconstructed frames with respect to other state-of-the-art techniques, as well as better PSNR results.

The lifting scheme is a very well-known computationally efficient alternative to the filter bank scheme for evaluating the discrete wavelet transform of signals and images. However, the actual computational saving is still a matter of debate. On one hand, theoretical results in the literature report an asymptotic upper-bound of two for very long wavelet filters. On the other hand, it is worth wondering to what extent the architecture of the processor used can actually bias this gain. In this paper we tackle this problem from an implementation perspective, and profile the execution time of the two algorithms on a digital signal processor. Both the real-valued and the integer versions of the wavelet transform are considered. The quantitative results are used to gain some insight on the way the processor architecture affects the algorithms.

In the context of joint source and channel coding, unequal loss protection is often used to make image data more robust to possible packet losses. The allocation of source and code symbols is customarily done so as to maximize the expected PSNR at the receiver. We propose a new objective function, attempting to maximize PSNR given a constraint on the system probability of failure, so that PSNR is constrained to be above a given threshold with a given probability. This leads to the definition of a hybrid loss protection scheme, and the related allocation algorithm, which is able to satisfy this constraint. Experimental results are reported, related to the transmission of JPEG2000-compressed images over the Internet. It is shown that the proposed hybrid approach outperforms existing algorithms in terms of PSNR, while requiring less computational resources.

We propose near-lossless digital watermarking for copyright protection of remote sensing images. In particular, we show that, by forcing a maximum absolute difference between the original and watermarked scene, the near-lossless paradigm makes it possible to decrease the effect of watermarking on remote sensing applications to be carried out on the images. As an example, the effect of near-lossless watermarking on image classification is analyzed.

In this article we evaluate the use of predictive coding for compression of SAR phase history data. We first show that the data are mainly correlated along range lines. Then, we exploit this result to define a new DPCM-based compression algorithm named RDPCM-BAQ. The performance of this algorithm is compared with that of BAQ on SIR-C/X-SAR data, showing a significant improvement in signal-to-noise ratio of up to 2 dB with respect to BAQ.

In this paper we compare two compression methods for SAR raw data, based on the discrete wavelet transform (DWT). In the former, the data are subject to blockwise normalization prior to being transformed by means of the DWT processor; then, an optimal rate allocation is performed for each subband. The latter employs the well known JPEG 2000 to perform the DWT and the subsequent quantization, rate allocation and coding steps; both the cases of normalized and non normalized data are considered. The performance of the algorithms have been tested on SIR-C/X-SAR data, and FBAQ is employed as the term of comparison for both quality and compression ratio. The obtained results show that the if the samples are normalized, the performance of the DWT-based algorithms is more predictable, and less subject to statistical fluctuations related to the characteristics of the input data.

Wireless video-surveillance networks are gaining increasing popularity due to the number of applications they make possible. We address the problem of designing a wireless video-surveillance network so as to optimize its performance. In particular, we investigate the possible trade-offs between energy consumption and image quality. We provide simulation results showing that video compression can be very beneficial in reducing data transmission costs, provided that the energy cost of video compression is low. Moreover, we discuss the impact of compression on image delay.

This paper deals with the design and implementation of an image transform coding algorithm based on the integer wavelet transform (IWT). First of all, criteria are proposed for the selection of optimal factorizations of the wavelet filter polyphase matrix to be employed within the lifting scheme. The obtained results lead to the IWT implementations with very satisfactory lossless and lossy compression performance. Then, the effects of finite precision representation of the lifting coefficients on the compression performance are analyzed, showing that, in most cases, a very small number of bits can be employed for the mantissa keeping the performance degradation very limited. Stemming from these results, a VLSI architecture is proposed for the IWT implementation, capable of achieving very high frame rates with moderate gate complexity.

The Hough transform is a widely used tool for line detection mainly due to its robustness to noise; on the other hand, it is also known to be computationally expensive, thus often preventing real-time operation. In this paper we evaluate the performance of an all-integer version of the Hough transform, implemented without floating point operations. We show that the integer transform is 2 to 3.5 times faster than the standard one on most platforms, while its performance loss is negligible.

We propose an on-board selection scheme for aerial and space images, based on linear feature detection in a feature hyperspace. The detection task is performed by means of the Radon transform (RT) and the wavelet transform; a fast algorithm for the RT computation is described, and counteractions against the discretization errors are proposed. A new, wavelet-based algorithm is introduced, which performs a fine analysis of the waveforms of the RT peaks, yielding a possibly error-free detection in images corrupted by a high level of noise. A technique, based on the feature hyperspace, is proposed, able to significantly exploit all the available pieces of information on these peaks. Results of the tests on synthetic and real images are reported, which show that this method achieves satisfactory results, making the detection task highly reliable in the presence of both noise and clutter.

In this paper we deal with the problem of finding high performance factorizations of wavelet filters to be employed within the lifting scheme framework to yield the integer wavelet transform (IWT). A method is proposed, based on the search for the factorization yielding the minimally non-linear iterated graphic function. Results are reported, referring to a set of popular wavelet filters, which show that the obtained implementations lead to IWTs achieving very satisfactory results for both lossy and lossless image compression.

This paper investigates the robustness of the wavelet transform, implemented by means of the lifting scheme (LS), with respect to numerical errors in the representation and calculation of transformed coefficients. The study promises to offer important contributions for the understanding of the LS capabilities when specific implementations are considered. This is a topic of growing interest as the new standard JPEG 2000, based on the wavelet transform, is being finalized. Taking into account the effect of finite precision representation can drive both software and hardware implementations with optimized trade off between complexity and performance; moreover the robustness to numerical errors could be an important feature, not usually considered, in order to select the best wavelet filters.

In many Earth observation missions, a large amount of data are collected by the on-board sensors, and must be transmitted to ground through a channel with limited capacity; in this case, besides lossy compression, one often has to select a subset of the original images for ground transmission. It is then desirable that the discarded images are those with a minor information content. In this paper we propose an algorithm for on-board image selection, which is fully integrated within a wavelet-based image compression scheme. The algorithm selects images possibly containing straight patterns, and uses the Hough transform, performed in the wavelet domain, for the detection task. It is shown that this method achieves a notable complexity reduction at the expense of a limited performance degradation, strongly increasing the sustainable real-time data throughput.