The development of antimicrobial peptides (AMPs) presents a promising approach to addressing antibiotic-resistant pathogens. Computational methods, such as Feedback Generative Adversarial Networks (FBGANs), have demonstrated strong performance in optimizing AMP design. FBGAN operates as a classifier-guided Generative Adversarial Network (GAN), refining training data by replacing them with the classifier's most accurate predictions based on a predefined threshold. However, this method may introduce bias and constrain the diversity and quality of the generated peptides. To address these limitations, we propose a novel classifier-driven GAN (cdGAN) framework that seamlessly integrates classifier predictions into the generative model's loss function. This enables an adaptive, end-to-end learning process that enhances AMP generation without requiring explicit data modifications. By embedding classifier guidance within the loss computation, cdGAN dynamically optimizes both peptide diversity and functionality. Comparative studies indicate that cdGAN outperforms conventional guided-GAN architectures, such as Conditional GANs and Auxiliary Classifier GANs, while achieving performance comparable to or exceeding established AMP design methods. Additionally, cdGAN's flexible architecture allows for the simultaneous optimization of multiple peptide attributes. To demonstrate this capability, we introduce a multi-task classifier based on the Evolutionary Scale Modeling 2 (ESM2) model, enabling cdGAN to assess both antimicrobial activity and peptide structural properties in parallel. This enhancement improves the likelihood of generating viable therapeutic candidates with enhanced antimicrobial effectiveness and reduced toxicity.

The increasing scale and complexity of remote sensing observations demand distributed processing to effectively manage the vast volumes of data generated. However, distributed processing presents significant challenges, including bandwidth limitations, high latency, and privacy concerns, especially when transmitting high-resolution images. To address these issues, we propose a novel scheme leveraging the encoder of a masked autoencoder to generate associated embedding (CLS tokens) from masked images which enable training deep learning models under Federated Learning scenarios. This approach enables the transmission of compact image patches instead of full images to processing nodes, drastically reducing bandwidth usage. On the processing nodes, classifiers are trained with the CLS tokens, and model weights are aggregated using FedAvg and FedProx Federated Learning algorithms. Experimental results on benchmark datasets demonstrate that the proposed approach significantly reduces data transmission requirements while maintaining and even surpassing the accuracy of systems with access to full data.

The Age of Incorrect Information (AoII) is a recently proposed metric for real-time remote monitoring systems. In particular, AoII measures the time the information at the monitor is incorrect, weighted by the magnitude of this incorrectness, thereby combining the notions of freshness and distortion. This paper addresses the definition of an AoII-optimal transmission policy in a discrete-time communication scheme with a resource constraint and a hybrid automatic repeat request (HARQ) protocol. Considering an N-ary symmetric Markov source, the problem is formulated as an infinite-horizon average-cost constrained Markov decision process (CMDP). Interestingly, it is proved that, under some conditions, the optimal transmission policy is to never transmit. This reveals a region of the source dynamics where communication is inadequate in reducing the AoII. Elsewhere, there exists an optimal transmission policy, which is a randomized mixture of two discrete threshold-based policies that randomize on at most one state. The optimal threshold and the randomization component are derived analytically. Numerical results illustrate the impact of the source dynamics, channel conditions, and resource constraints on the average AoII.

Antimicrobial peptides (AMPs) play a significant role in guiding drug design, advancing targeted therapies, and cancer treatment research. The function of peptides is highly associated with their three-dimensional structure. AMPs particularly favor alpha-helical structures, or alpha-folds, due to their ability to disrupt the protective layers that surround cells effectively and their structural stability. Existing classifiers mainly identify AMPs but overlook their structural fold which can provide valuable insights into their function. To address this limitation, we introduce an innovative multitask classifier that recognizes AMPs and predicts their alphahelical folds simultaneously. Our approach employs k-mers and Transformer networks for efficient, accurate multitask classification. Results on the datasets indicate comparable performance compared to single-task methods in half the time and complexity.

This paper presents an enhanced methodology for Recurrence Quantification Analysis (RQA) designed specifically for video analysis. By utilizing image quality metrics, with a focus on the Peak Signal-to-Noise Ratio (PSNR), we determine meaningful values for the RQA threshold $\varepsilon$, a critical factor for successful image processing. Utilizing the False Nearest Neighbors (FNN) technique, we identify the optimal embedding dimension D for each patch within the video frames. Our approach produces a heatmap that visualizes temporal recurrence information for each video patch.

The substantial volume of continuously gathered remote sensing data can serve as a valuable information source in mitigating the effects of natural disasters. This involves identifying changes in the time series of observations. Considering that the precise location of the changes may not be available in real-world scenarios, we propose an unsupervised method for detecting extreme events in multi-temporal satellite images. Specifically, we learn a basis matrix of each dimension of the feature space of the images using the tensor decomposition learning method. Then, each new image is represented in the feature space by expressing it as a multilinear combination of the learned tensor decomposition factors. The predicted changes can be obtained by comparing and thresholding the distance of the corresponding extracted features of the images before and after the event. Experimental results on real Sentinel-2 multi-temporal images demonstrate that the proposed method can efficiently detect the effects of fires and floods with low complexity.

Retrieval of Surface Soil Moisture (SSM) over large scales at high spatial resolution is crucial for numerous applications. Existing solutions rely on the analysis of remote sensing platforms or in situ measurements that are either too coarse in their resolution or too localized to address the aforementioned need. In this work, we propose a novel deep learning approach for reliably estimating SSM at a high spatial resolution of 1 km over broad regions. To achieve this objective, the proposed framework employs a Convolutional Neural Network that can capture both multi-modal and spatial correlations. Introducing a novel loss function, the proposed scheme can leverage limited in situ observations while also generating estimates consistent with physical models. This is achieved through the utilization of coarse-resolution data assimilation estimates. For training and assessing the performance of the proposed framework, a novel dataset is generated by combining information from remote sensing, in situ measurements, and data assimilation estimates. Experimental analysis demonstrates that the proposed approach can provide accurate retrieval of SSM, significantly outperforming existing products.

The goal of this work is to propose a simple yet efficient way to dynamically transform a video stream according to the functional properties of the visual system. To achieve this goal, we extend to video sequences the Retina-Inspired Filter (RIF), which we have recently proposed for still images. Under the assumption that the input signal remains constant for a given time, the RIF decomposition was proven to be in-vertible, meaning that the image could be perfectly recovered. In this paper, we relax this assumption into a piece-wise constant input and analytically prove that the RIF can be applied to a group of pictures (GOP). We experimentally show that the size of GOP is important when motion appears, as some artifacts are generated. However, in the absence of motion among the GOP frames we can still perfectly reconstruct the video frames reducing the computational cost of the whole process.

The Age of Incorrect Information (AoII) is studied within the context of remote monitoring a Markov source using variable-length stop-feedback (VLSF) coding. Leveraging recent results on the non-asymptotic channel coding rate, we consider sources with small cardinality, where feedback is non-instantaneous as the transmitted information and feedback message have comparable lengths. We focus on the feedback sequence, i.e. the times of feedback transmissions, and derive AoII-optimal and delay-optimal feedback sequences. Our results showcase the impact of the feedback sequence on the AoII, revealing that a lower average delay does not necessarily correspond to a lower average AoII. We discuss the implications of our findings and suggest directions for coding scheme design.

Predicting the structural class of a protein from its amino acid sequence is among the most significant problems in bioinformatics, especially for proteins with a low sequence similarity. While current methods using recurrent neural networks achieve a notable accuracy in this task, their approach relies on extracting a large quantity of features, impacting the efficiency and reliability of the prediction. In this work, we introduce an efficient and accurate classification scheme based on chaos game representation and recurrent neural networks. The proposed scheme achieves comparable results with state-of-the-art methods, while using a significantly lower-dimensional representation of the feature space.

Due to the importance of the lower bounding distances and the attractiveness of symbolic representations, the family of symbolic aggregate approximations (SAX) has been used extensively for encoding time series data. However, typical SAX-based methods rely on two restrictive assumptions; the Gaussian distribution and equiprobable symbols. This paper proposes two novel data-driven SAX-based symbolic representations, distinguished by their discretization steps. The first representation, oriented for general data compaction and indexing scenarios, is based on the combination of kernel density estimation and Lloyd-Max quantization to minimize the information loss and mean squared error in the discretization step. The second method, oriented for high-level mining tasks, employs the Mean-Shift clustering method and is shown to enhance anomaly detection in the lower-dimensional space. Besides, we verify on a theoretical basis a previously observed phenomenon of the intrinsic process that results in a lower than the expected variance of the intermediate piecewise aggregate approximation. This phenomenon causes an additional information loss but can be avoided with a simple modification. The proposed representations possess all the attractive properties of the conventional SAX method. Furthermore, experimental evaluation on real-world datasets demonstrates their superiority compared to the traditional SAX and an alternative data-driven SAX variant.

Multi-temporal remotely sensed observations acquired by multi-spectral sensors contain a wealth of information related to the Earth’s state. Deep learning methods have demonstrated a great potential in analyzing such observations. Traditional 2D and 3D approaches are unable to effectively extract valuable information encoded across all available dimensions. In this work, we propose the extension of current fully-convolutional models for multi-temporal remote sensing data classification to their high-dimensional analogs, which can naturally capture multi-dimensional dependencies and correlations. Experimental analysis on recently released observations from Landsat-8 reveals that our proposed high-dimensional fully-convolutional approach exhibits a clear classification performance improvement over state of the art low-order fully-convolutional models.

The goal of this work is to propose a simple yet efficient way to dynamically transform a sequence of images according to the functional properties of the visual system. To achieve this goal, we extend to video sequences the Retina-Inspired Filter (RIF), which we have recently proposed for still images. Under the assumption that the input signal remains constant for a given time, the RIF decomposition was proven to be invertible, meaning that the image could be perfectly recovered. In this paper, we relax this assumption into a piece-wise constant input and we prove that RIF can be applied to a Group Of Pictures (GOP). Under the condition that a GOP consists of frames without strong pixel motion, we mathematically prove and experimentally show that when RIF is applied to GOP, whatever the size of the GOP is, we are still able to perfectly recover the video frames and at the same time simplify the complexity of the whole process. In addition, we show that while the GOP size increases, the memory cost required to store this amount of frames is sufficiently reduced.

This work focuses on the problem of surface soil moisture estimation from multi-modal remote sensing observations. We focus on the scenario where both passive radiometer observations from NASA SMAP satellite, as well as active radar measurements from ESA Sentinel 1 are available. We formulate the problem as multi-source observation fusion and develop a deep learning model for SM estimation. To train and validate the performance of the proposed scheme, we consider observations from in-situ SM sensor networks over the continental USA. Experimental results demonstrate that the proposed model achieves high quality SM estimation, surpassing the performance of available products.

MOTIVATION: Protein structural class prediction is one of the most significant problems in bioinformatics, as it has a prominent role in understanding the function and evolution of proteins. Designing a computationally efficient but at the same time accurate prediction method remains a pressing issue, especially for sequences that we cannot obtain a sufficient amount of homologous information from existing protein sequence databases. Several studies demonstrate the potential of utilizing chaos game representation along with time series analysis tools such as recurrence quantification analysis, complex networks, horizontal visibility graphs (HVG) and others. However, the majority of existing works involve a large amount of features and they require an exhaustive, time consuming search of the optimal parameters. To address the aforementioned problems, this work adopts the generalized multidimensional recurrence quantification analysis (GmdRQA) as an efficient tool that enables to process concurrently a multidimensional time series and reduce the number of features. In addition, two data-driven algorithms, namely average mutual information and false nearest neighbors, are utilized to define in a fast yet precise manner the optimal GmdRQA parameters. RESULTS: The classification accuracy is improved by the combination of GmdRQA with the HVG. Experimental evaluation on a real benchmark dataset demonstrates that our methods achieve similar performance with the state-of-the-art but with a smaller computational cost. AVAILABILITY AND IMPLEMENTATION: The code to reproduce all the results is available at https://github.com/aretiz/protein_structure_classification/tree/main. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

This paper introduces a novel coding/decoding mechanism that mimics one of the most important properties of the human visual system: its ability to enhance the visual perception quality in time. In other words, the brain takes advantage of time to process and clarify the details of the visual scene. This characteristic is yet to be considered by the state-of-the-art quantization mechanisms that process the visual information regardless the duration of time it appears in the visual scene. We propose a compression architecture built of neuroscience models; it first uses the leaky integrate-and-fire (LIF) model to transform the visual stimulus into a spike train and then it combines two different kinds of spike interpretation mechanisms (SIM), the time-SIM and the rate-SIM for the encoding of the spike train. The time-SIM allows a high quality interpretation of the neural code and the rate-SIM allows a simple decoding mechanism by counting the spikes. For that reason, the proposed mechanisms is called Dual-SIM quantizer (Dual-SIMQ). We show that (i) the time-dependency of Dual-SIMQ automatically controls the reconstruction accuracy of the visual stimulus, (ii) the numerical comparison of Dual-SIMQ to the state-of-the-art shows that the performance of the proposed algorithm is similar to the uniform quantization schema while it approximates the optimal behavior of the non-uniform quantization schema and (iii) from the perceptual point of view the reconstruction quality using the Dual-SIMQ is higher than the state-of-the-art.

As the field of brain monitoring is evolving rapidly, there is an increasing demand of finding innovative ways to handle relevant signals. Especially electroencephalogram (EEG) signals provide a non-invasive way of diagnostic inference of brain's functionality. Nevertheless, EEG signals are often corrupted by impulsive noise, thus prior denoising is required for accurate analysis and decision making. On the other hand, EEG signals admit naturally a representation in the form of graphs, with the electrodes corresponding to the nodes of the graph and the edges expressing the connectivity strength. To this end, graph signal processing (GSP) is a versatile tool, which enables the representation and analysis of graph-structured signals, whose interdependencies are encoded in the form of an appropriate adjacency matrix. To address the denoising of graph-structured signals, under impulsive noise conditions, this work introduces a regularized graph filtering scheme based on fractional lower order moments, coupled with distinct adjacency matrices inspired both by statistical approaches and visibility graphs that are better capable of capturing the topological and functional connectivity between the distinct nodes. The experimental evaluation on real EEG signals recorded in epileptic and non-epileptic seizures, reveals the effects of the adjacency matrix choice on the denoising performance.

Nowadays, multidimensional data structures, known as tensors, are widely used in many applications like earth observation from remote sensing image sequences. However, the increasing spatial, spectral and temporal resolution of the acquired images, introduces considerable challenges in terms of data storage and transfer, making critical the necessity of an efficient compression system for high dimensional data. In this paper, we propose a tensor-based compression algorithm that retains the structure of the data and achieves a high compression ratio. Specifically, our method learns a dictionary of specially structured tensors using the Alternating Direction Method of Multipliers, as well as a symbol encoding dictionary. During run-time, a quantized and encoded sparse vector of coefficients is transmitted, instead of the whole multidimensional signal. Experimental results on real satellite image sequences demonstrate the efficacy of our method compared to a state-of-the-art compression method.

Image compression still remains one of the most challenging scientific fields as it is widely used in almost every kind of application. To design ground-breaking architectures the image processing community concentrated at first on visual perception. However, understanding how the visual system senses the semantics of the visual scene is still a big issue. Recently, a neuro-inspired compression (NICE) system was introduced, trying to encode the visual information by mimicking the firing mechanisms of neurons. The performance of this system was promising but inefficient when opposed to the JPEG standard. In this work, we aim at improving the architecture design of the NICE system. Thus, we propose 3 different architectures and we show that the quality of the reconstruction is substantially enhanced. In addition, the NICE system outperforms the JPEG standard according to the assessment of two full-reference metrics; the peak signal-to-noise ratio (PSNR) and the structure similarity index measure (SSIM), except for very low bits per pixel values.

High-dimensional data structures, known as tensors, are fundamental in many applications, including multispectral imaging and color video processing. Compression of such huge amount of multidimensional data collected over time is of paramount importance, necessitating the process of quantization of measurements into discrete values. Furthermore, noise and issues related to the acquisition and transmission of signals frequently lead to unobserved, lost or corrupted measurements. In this paper, we introduce a tensor robust principal component analysis algorithm in order to recover a tensor with real-valued entries from a partly observed set of quantized and sparsely corrupted entries. We formulate the problem as a constrained maximum likelihood estimation of the sum of a low-rank tensor and a sparse tensor, through matricizations in each mode, in combination with a quantization and statistical measurement model. Experimental results on satellite derived land surface time-series demonstrate that directly operating with the quantized measurements, rather than treating them as real values, results in a low recovery error, while the proposed method is also capable of detecting temperature anomalies (e.g., forest fires).

In this work, we propose a novel Convolutional Neural Network architecture for increasing the low spatial resolution SMAP radiometer based soil moisture estimations from 36 km to 3 km resolution by using time-series of observations from both SMAP's radiometer and Sentinel-1 radar. By simultaneously extracting features from both current low-resolution input and residuals between high and low resolution at previous time instances, the proposed network is capable of accurately estimating soil moisture using coarse resolution observations. Experimental results on three different locations demonstrate that the proposed scheme is able to estimate soil moisture with accuracy in the range of the requirements set by the SMAP science team.

In this paper, we address the problem of spectroscopic redshift estimation in Astronomy. Due to the expansion of the Universe, galaxies recede from each other on average. This movement causes the emitted electromagnetic waves to shift from the blue part of the spectrum to the red part, due to the Doppler effect. Redshift is one of the most important observables in Astronomy, allowing the measurement of galaxy distances. Several sources of noise render the estimation process far from trivial, especially in the low signal-to-noise regime of many astrophysical observations. In recent years, new approaches for a reliable and automated estimation methodology have been sought out, in order to minimize our reliance on currently popular techniques that heavily involve human intervention. The fulfilment of this task has evolved into a grave necessity, in conjunction with the insatiable generation of immense amounts of astronomical data. In our work, we introduce a novel approach based on Deep Convolutional Neural Networks. The proposed methodology is extensively evaluated on a spectroscopic dataset of full spectral energy galaxy distributions, modelled after the upcoming Euclid satellite galaxy survey. Experimental analysis on observations of idealistic and realistic conditions demonstrate the potent capabilities of the proposed scheme.

Bracketed High Dynamic Range (HDR) imaging architectures acquire a sequence of Low Dynamic Range (LDR) images in order to either produce a HDR image or an “optimally” exposed LDR image, achieving impressive results under static camera and scene conditions. However, in real world conditions, ghost-like artifacts and noise effects limit the quality of HDR reconstruction. We address these limitations by introducing a post-acquisition snapshot HDR enhancement scheme that generates a bracketed sequence from a small set of LDR images, and in the extreme case, directly from a single exposure. We achieve this goal via a sparse-based approach where transformations between differently exposed images are encoded through a dictionary learning process, while we learn appropriate features by employing a stacked sparse autoencoder (SSAE) based framework. Via experiments with real images, we demonstrate the improved performance of our method over the state-of-the-art, while our single-shot based HDR formulation provides a novel paradigm for the enhancement of LDR imaging and video sequences.

This paper presents a method for reducing the Diffusion Weighted Magnetic Resonance Imaging (DW-MRI) examination time based on the mathematical framework of sparse representations. The aim is to undersample the b-values used for DW-MRI image acquisition which reflect the strength and timing of the gradients used to generate the DW-MRI images since their number defines the examination time. To test our method we investigate whether the undersampled DW-MRI data preserve the same accuracy in terms of extracted imaging biomarkers. The main procedure is based on the use of the k-Singular Value Decomposition (k-SVD) and the Orthogonal Matching Pursuit (OMP) algorithms, which are appropriate for the sparse representations computation. The presented results confirm the hypothesis of our study as the imaging biomarkers extracted from the sparsely reconstructed data have statistically close values to those extracted from the original data. Moreover, our method achieves a low reconstruction error and an image quality close to the original.

Dyslexia is a neurodevelopmental learning disorder that affects the acceleration and precision of word recognition, therefore obstructing the reading fluency, as well as text comprehension. Although it is not an oculomotor disease, readers with dyslexia have shown different eye movements than typically developing subjects during text reading. The majority of existing screening techniques for dyslexia's detection employ features associated with the aberrant visual scanning of reading text seen in dyslexia, whilst ignoring completely the behavior of the underlying data generating dynamical system. To address this problem, this work proposes a novel self-tuned architecture for feature extraction by modeling directly the inherent dynamics of wearable sensor data in higher-dimensional phase spaces via multidimensional recurrence quantification analysis (RQA) based on state matrices. Experimental evaluation on real data demonstrates the improved recognition accuracy of our method when compared against its state-of-the-art vector-based RQA counterparts.

High resolution astronomical imagery plays a critical role in multiple remote sensing applications. In this work, we introduce a novel post-acquisition computational technique aiming to recover the high-quality versions of blurry and degraded astronomical observations. Additionally, the proposed scheme is able to retrieve significant information regarding the characteristic properties of the blurring kernel, i.e point spread function (PSF). In order to accomplish this goal, we exploit the mathematical frameworks of Sparse Reprentations, and the Alternating Direction Method of Multipliers (ADMM). Experimental results demonstrate the ability of the proposed approach to synthesize high-quality astronomical imagery.

Convolutional Neural Networks (CNNs) represent the cutting edge in signal analysis tasks like classification and regression. Realization of such architectures in hardware capable of performing high throughput computations, with minimal energy consumption, is a key enabling factor towards the proliferation of analysis immediately after acquisition. Our driving problem is a satellite-based remote sensing platform in which onboard signal processing and classification tasks must take place, given strict bandwidth and energy limitations. In this work, we exploit the implementation of a CNN on Field Programmable Gate Array (FPGA) platforms and explore different ways to minimize the impact of different hardware restrictions to performance. We compare our results against competing technologies such as Graphics Processing Units (GPU) in terms of throughput, latency and energy consumption. In actual experimental runs we demonstrate competitive latency and throughput of the FPGA platform vs. GPU technology at an order-of-magnitude energy savings, which is especially important for space-borne computing.

Land cover classification is a flourishing research topic in the field of remote sensing. Conventional methodologies mainly focus either on the simplified single-label case or on the pixel-based approaches that cannot efficiently handle high-resolution images. On the other hand, the problem of multilabel land cover scene categorization remains, to this day, fairly unexplored. While deep learning and convolutional neural networks have demonstrated an astounding capacity at handling challenging machine learning tasks, such as image classification, they exhibit an underwhelming performance when trained with a limited amount of annotated examples. To overcome this issue, this paper proposes a data augmentation technique that can drastically increase the size of a smaller data set to copious amounts. Our experiments on a multilabel variation of the UC Merced Land Use data set demonstrate the potential of the proposed methodology, which outperforms the current state of the art by more than 6% in terms of the F-score metric.

High-spectral resolution imaging systems play a critical role in the identification and characterization of objects in a scene of interest. Unfortunately, multiple factors impair spectral resolution, as in the case of modern snapshot spectral imagers that associate each hyperpixel with a specific spectral band. In this paper, we introduce a novel postacquisition computational technique aiming to enhance the spectral dimensionality of imaging systems by exploiting the mathematical frameworks of sparse representations and dictionary learning. We propose a coupled dictionary learning model which considers joint feature spaces, composed of low- and high-spectral resolution hypercubes, in order to achieve spectral superresolution performance. We formulate our spectral coupled dictionary learning optimization problem within the context of the alternating direction method of multipliers, and we manage to update the involved quantities via closed-form expressions. In addition, we consider a realistic spectral subsampling scenario, taking into account the spectral response functions of different satellites. Moreover, we apply our spectral superresolution algorithm on real satellite data acquired by Landsat-8 and Sentinel-2 sensors. Finally, we have investigated the problem of hyperspectral image unmixing using the recovered high-spectral resolution data cube, and we are able to demonstrate that the proposed scheme provides significant value in hyperspectral image understanding techniques. Experimental results demonstrate the ability of the proposed approach to synthesize high-spectral-resolution 3-D hypercubes, achieving better performance compared to state-of-the-art resolution enhancement methods.

Human Activity Recognition (HAR) must currently face up to the challenge of rethinking analytics from the perspective of real-time operation, wherein biophysical sensing streams are efficiently intertwined at close vicinity to the point of sensing. As such, feature selection techniques, traditionally employed for off-line data processing, should be evaluated with respect to their ability to filter out redundant information in real-time. In this work, we propose an online architecture for implementing feature selection on mobile devices, and we evaluate popular feature selection methods against constantly alternating activity labels. We perform a qualitative analysis to determine the dominant sensing modality that dictates the activities of a certain time duration. The results indicate that online feature selection performance changes among consecutive data partitions, leading to the conclusion that the type of available activity influences significantly the feature selection procedure.

Sensor-based activity recognition is encountered in innumerable applications of the arena of pervasive healthcare and plays a crucial role in biomedical research. Nonetheless, the frequent situation of unobserved measurements impairs the ability of machine learning algorithms to efficiently extract context from raw streams of data. In this paper, we study the problem of accurate estimation of missing multimodal inertial data and we propose a classification framework that considers the reconstruction of subsampled data during the test phase. We introduce the concept of forming the available data streams into low-rank two-dimensional (2-D) and 3-D Hankel structures, and we exploit data redundancies using sophisticated imputation techniques, namely matrix and tensor completion. Moreover, we examine the impact of reconstruction on the classification performance by experimenting with several state-of-the-art classifiers. The system is evaluated with respect to different data structuring scenarios, the volume of data available for reconstruction, and various levels of missing values per device. Finally, the tradeoff between subsampling accuracy and energy conservation in wearable platforms is examined. Our analysis relies on two public datasets containing inertial data, which extend to numerous activities, multiple sensing parameters, and body locations. The results highlight that robust classification accuracy can be achieved through recovery, even for extremely subsampled data streams.

Obtaining an up-to-date high-resolution description of land cover is a challenging task due to the high cost and labor-intensive process of human annotation through field studies. This work introduces a radically novel approach for achieving this goal by exploiting the proliferation of remote sensing satellite imagery, allowing for the up-to-date generation of global-scale land cover maps. We propose the application of multilabel classification, a powerful framework in machine learning, for inferring the complex relationships between the acquired satellite images and the spectral profiles of different types of surface materials. Introducing a drastically different approach compared to unsupervised spectral unmixing, we employ contemporary ground-collected data from the European Environment Agency to generate the label set and multispectral images from the MODIS sensor to generate the spectral features, under a supervised classification framework. To validate the merits of our approach, we present results using several state-of-the-art multilabel learning classifiers and evaluate their predictive performance with respect to the number of annotated training examples, as well as their capability to exploit examples from neighboring regions or different time instances. We also demonstrate the application of our method on hyperspectral data from the Hyperion sensor for the urban land cover estimation of New York City. Experimental results suggest that the proposed framework can achieve excellent prediction accuracy, even from a limited number of diverse training examples, surpassing state-of-the-art spectral unmixing methods.

Fingerprint-based location sensing technologies play an increasingly important role in pervasive computing applications due to their accuracy and minimal hardware requirements. However, typical fingerprint-based schemes implicitly assume that communication occurs over the same channel (frequency) during the training and the runtime phases. When this assumption is violated, the mismatches between training and runtime fingerprints can significantly deteriorate the localization performance. Additionally, the exhaustive calibration procedure required during training limits the scalability of this class of methods. In this work, we propose a novel, scalable, multi-channel fingerprint-based indoor localization system that employs modern mathematical concepts based on the Sparse Representations and Matrix Completion theories. The contribution of our work is threefold. First, we investigate the impact of channel changes on the fingerprint characteristics and the effects of channel mismatch on state-of-the-art localization schemes. Second, we propose a novel fingerprint collection technique that significantly reduces the calibration time, by formulating the map construction as an instance of the Matrix Completion problem. Third, we propose the use of sparse Bayesian learning to achieve accurate location estimation. Experimental evaluation on real data highlights the superior performance of the proposed framework in terms of reconstruction error and localization accuracy.

This paper addresses the problem of efficient sampling and reconstruction of sparse spike signals, which have been convolved with low-pass filters. A modified compressed sensing (CS) framework is proposed, termed dictionary-based deconvolution CS (DDCS) to achieve this goal. DDCS builds on the assumption that a low-pass filter can be represented sparsely in a dictionary of blurring atoms. Identification of both the sparse spike signal and the sparsely parameterized blurring function is performed by an alternating scheme that minimizes each variable independently, while keeping the other constant. Simulation results reveal that the proposed DDSS scheme achieves an improved reconstruction performance when compared to traditional CS recovery.

This paper introduces a new technique for compressive sampling reconstruction of biomedical ultrasound images that exploits two types of prior information. On the one hand, our proposed approach is based on the observation that ultrasound RF echoes are best characterised statistically using alpha-stable distributions. On the other hand, through knowledge of the acquisition process, the support of the RF echoes in the Fourier domain can be easily inferred. Together, these two facts inform an iteratively reweighted least squares (IRLS) algorithm, which is shown to outperform previously proposed reconstruction techniques, both visually and in terms of two objective evaluation measures.

Dynamic recognition of gestures from video sequences is a challenging task due to the high variability in the characteristics of each gesture with respect to different individuals. In this work, we propose a novel representation of gestures as linear combinations of the elements of an overcomplete dictionary, based on the emerging theory of sparse representations. We evaluate our approach on a publicly available gesture dataset of Palm Grafti Digits and compare it with other state-of-the-art methods, such as Hidden Markov Models, Dynamic Time Warping and the recently proposed distance metric termed Move-Split-Merge. Our experimental results suggest that the proposed recognition scheme offers high recognition accuracy in isolated gesture recognition and a satisfying robustness to noisy data, thus indicating that sparse representations can be successfully applied in the field of gesture recognition.

In this paper, we address the problem of video classification from a set of compressed features. In particular, the properties of linear random projections in the framework of compressive sensing are exploited to reduce the task of classifying a given video sequence into a problem of sparse reconstruction, based on feature vectors consisting of measurements lying in a low-dimensional compressed domain. This can be of great importance in decision systems with limited power, processing, and bandwidth resources, since the classification is performed without handling the original high-resolution video data, but working directly with the set of compressed measurements. The experimental evaluation verifies the efficiency of the proposed scheme and illustrates that the compressed measurements in conjunction with an appropriate decision rule result in an effective video classification scheme, which meets the constraints of systems with limited resources.

The broadcast nature of wireless networks has been widely exploited by adversaries in order to cause severe denial-of-service attacks. Several algorithms are proposed in the literature for the detection and mitigation of such attacks at the physical and medium access layers. In this work, we combine recent advances in compressed sensing theory, along with a cumulative-sum anomaly-based algorithm, for the detection of physical-layer attacks. The algorithm considers a metric based on the Signal-to-Interference-plus-Noise-Ratio (SINR). Compressed sensing makes feasible the use of far fewer SINR measurements for effective intrusion detection. The performance evaluation based on real experimental data shows that attacks are detected with high accuracy using a small number of measurements.

Compressed sensing (CS) samples signals at a much lower rate than the Nyquist rate if they are sparse in some basis. In this paper, the CS methodology is applied to sinusoidally modeled audio signals. As this model is sparse by definition in the frequency domain (being equal to the sum of a small number of sinusoids), we investigate whether CS can be used to encode audio signals at low bitrates. In contrast to encoding the sinusoidal parameters (amplitude, frequency, phase) as current state-of-the-art methods do, we propose encoding few randomly selected samples of the time-domain description of the sinusoidal component (per signal segment). The potential of applying compressed sensing both to single-channel and multi-channel audio coding is examined. The listening test results are encouraging, indicating that the proposed approach can achieve comparable performance to that of state-of-the-art methods. Given that CS can lead to novel coding systems where the sampling and compression operations are combined into one low-complexity step, the proposed methodology can be considered as an important step towards applying the CS framework to audio coding applications.

Traditional bearing estimation techniques perform Nyquist-rate sampling of the received sensor array signals and as a result they require high storage and transmission bandwidth resources. Compressed sensing (CS) theory provides a new paradigm for simultaneously sensing and compressing a signal using a small subset of random incoherent projection coefficients, enabling a potentially significant reduction in the sampling and computation costs. In this paper, we develop a Bayesian CS (BCS) approach for estimating target bearings based on multiple noisy CS measurement vectors, where each vector results by projecting the received source signal on distinct over-complete dictionaries. In addition, the prior belief that the vector of projection coefficients should be sparse is enforced by fitting directly the prior probability distribution with a Gaussian Scale Mixture (GSM) model. The experimental results show that our proposed method, when compared with norm-based constrained optimization CS algorithms, as well as with single-measurement BCS methods, improves the reconstruction performance in terms of the detection error, while resulting in an increased sparsity.

The ease of image storage and transmission in modern applications would be unfeasible without compression, which converts high-resolution images into a relatively small set of significant transform coefficients. Due to the specific content of many real-world images they are highly sparse in an appropriate orthonormal basis. The inherent property of compressed sensing (CS) theory working simultaneously as a sensing and compression protocol, using a small subset of random incoherent projection coefficients, enables a potentially significant reduction in the sampling and computation costs of images favoring its use in real-time applications which do not require an excellent reconstruction performance. In this paper, we develop a Bayesian CS (BCS) approach for obtaining highly sparse representations of images based on a set of noisy CS measurements, where the prior belief that the vector of projection coefficients should be sparse is enforced by fitting directly its prior probability distribution by means of a Gaussian Scale Mixture (GSM). The experimental results show that our proposed method, when compared with norm-based constrained optimization algorithms, maintains the reconstruction performance, in terms of the reconstruction error and the PSNR, while achieving an increased sparsity using much less basis functions.

This paper proposes a novel localization technique based on a multivariate Gaussian modeling of the signal strength measurements collected from several access points (APs) at different locations. It considers a discretized grid-like form of the environment and computes a signature at each cell of the grid. At run time the system compares the signature at the unknown position with the signature of each cell using the Kullback-Leibler Divergence estimation (KLD) between their corresponding probability densities. The paper evaluates the performance of the proposed technique and compares it with other statistical fingerprint-based localization systems. The performance analysis studies were conducted at the premises of a research laboratory and an aquarium under various conditions. Furthermore, the paper evaluates the impact of the number of APs and the size of the measurement datasets.

In this paper, the compressed sensing (CS) methodology is applied to the harmonic part of sinusoidally-modeled audio signals. As this part of the model is sparse by definition in the frequency domain, we investigate how CS can be used to encode this signal at low bitrates, instead of encoding the sinusoidal parameters (amplitude, frequency, phase) as current state-of-the-art methods do. We extend our previous work by considering an improved system model, by comparing our model to other schemes, and exploring the effect of incorrectly reconstructed frames. We show that encouraging results can be obtained by our approach, although inferior at this point compared to state-of-the-art. Good performance is obtained using 24 bits per sinusoid as indicated by our listening tests.

In this paper, a multichannel version of the sinusoids plus noise model (also known as deterministic plus stochastic decomposition) is proposed and applied to spot microphone signals of a music recording. These are the recordings captured by the various microphones placed in a venue, before the mixing process produces the final multichannel audio mix. Coding these microphone signals makes them available to the decoder, allowing for interactive audio reproduction which is a necessary component in immersive audio applications. The proposed model uses a single reference audio signal in order to derive a noise signal per spot microphone. This noise signal can significantly enhance the sinusoidal representation of the corresponding spot signal. The reference can be one of the spot signals or a downmix, depending on the application. Thus, for a collection of multiple spot signals, only the reference is fully encoded (e.g., as an MP3 monophonic signal). For the remaining spot signals, their sinusoidal parameters and corresponding noise spectral envelopes are retained and coded, resulting in bitrates for this side information in the order of 15 kb/s for perceptual performance above the 4.0 grade on the mean opinion score (MOS) scale.

MicroRNAs (miRNAs) are small single stranded RNAs, on average 22nt long, generated from endogenous hairpin-shaped transcripts with post-transcriptional activity. Although many computational methods are currently available for identifying miRNA genes in the genomes of various species, very few algorithms can accurately predict the functional part of the miRNA gene, namely the mature miRNA. We introduce a computational method that uses a Naive Bayes classifier to identify mature miRNA candidates based on sequence and secondary structure information of the miRNA precursor. Specifically, for each mature miRNA, we generate a set of negative examples of equal length on the respective precursor(s). The true and negative sets are then used to estimate probability distributions for sequence composition and secondary structure on each position along the RNA. The distance between these distributions is estimated using the symmetric Kullback-Leibler metric. The positions at which the two distributions differ significantly and consistently over a 10-fold cross-validation procedure are used as features for training the Naive Bayes classifier. A total of 15 classifiers were trained with true positive and negative examples from human and mouse. A performance of 76% sensitivity and 65% specificity was achieved using a consensus averaging over a 10-fold cross-validation procedure. Our findings suggest that position specific sequence and structure information combined with a simple Bayes classifier achieve a good performance on the challenging task of mature miRNA identification.

A novel context enhancement technique is presented to automatically combine images of the same scene captured at different times or seasons. A unique characteristic of the algorithm is its ability to extract and maintain the meaningful information in the enhanced image while recovering the surrounding scene information by fusing the background image. The input images are first decomposed into multiresolution representations using the Dual-Tree Complex Wavelet Transform (DT-CWT) with the subband coefficients modelled as Cauchy random variables. Then, the convolution of Cauchy distributions is applied as a probabilistic prior to model the fused coefficients, and the weights used to combine the source images are optimised via Maximum Likelihood (ML) estimation. Finally, the importance map is produced to construct the composite approximation image. Experiments show that this new model significantly improves the reliability of the feature selection and enhances fusion process.

This paper addresses the construction of a novel efficient rotation-invariant texture retrieval method that is based on the alignment in angle of signatures obtained via a steerable sub-Gaussian model. In our proposed scheme, we first construct a steerable multivariate sub-Gaussian model, where the fractional lower-order moments of a given image are associated with those of its rotated versions. The feature extraction step consists of estimating the so-called covariations between the orientation subbands of the corresponding steerable pyramid at the same or at adjacent decomposition levels and building an appropriate signature that can be rotated directly without the need of rotating the image and recalculating the signature. The similarity measurement between two images is performed using a matrix-based norm that includes a signature alignment in angle between the images being compared, achieving in this way the desired rotation-invariance property. Our experimental results show how this retrieval scheme achieves a lower average retrieval error, as compared to previously proposed methods having a similar computational complexity, while at the same time being competitive with the best currently known state-of-the-art retrieval system. In conclusion, our retrieval method provides the best compromise between complexity and average retrieval performance.

DNA microarrays have demonstrated an excellent potential in correlating specific gene expression profiles to specific conditions. However, they are affected by inherent noise. This paper presents a two-stage approach for noise removal that processes the additive and the multiplicative noise component. The proposed approach first decomposes the signal by a multiresolution transform and then accounts for both the multiscale correlation of the subband decompositions and their heavy-tailed statistics. Real microarray images have been processed by the proposed method and its improved performance is shown through quantitative measures and qualitative visual evaluation.

Network traffic load in an IEEE802.11 infrastructure arises from the superposition of traffic accessed by wireless clients associated with access points (APs). An accurate characterization of these data can be beneficial in modelling network traffic and addressing a variety of problems including coverage planning, resource reservation and network monitoring for anomaly detection. This study focuses on the statistical analysis of the traffic load measured in a campus-wide IEEE802.11 infrastructure at each AP.

In this paper, a novel method for single-channel speech enhancement is proposed, which is based on a spectral conversion feature denoising approach. Spectral conversion has been applied previously in the context of voice conversion, and has been shown to successfully transform spectral features with particular statistical properties into spectral features that best fit (with the constraint of a piecewise linear transformation) different target statistics. This spectral transformation is applied as an initialization step to two well-known single channel enhancement methods, namely the iterative Wiener filter (IWF) and a particular iterative implementation of the Kalman filter. In both cases, spectral conversion is shown here to provide a significant improvement as opposed to initializations using the spectral features directly from the noisy speech. In essence, the proposed approach allows for applying these two algorithms in a user-centric manner, when “clean” speech training data are available from a particular speaker. The extra step of spectral conversion is shown to offer significant advantages regarding output signal-to-noise ratio (SNR) improvement over the conventional initializations, which can reach 2 dB for the IWF and 6 dB for the Kalman filtering algorithm, for low input SNRs and for white and colored noise, respectively.

This paper describes a novel methodology for automatic musical genre classification based on a feature extraction/statistical similarity measurement approach. First, we perform a 1-D wavelet decomposition of the music signal and we model the resulting subband coefficients using the generalized Gaussian density (GGD) and the alpha-stable distribution. Subsequently, the GGD and alpha-stable distribution parameters are estimated during the feature extraction step, while the similarity between two music signals is measured by employing the Kullback-Leibler divergence (KLD) between their corresponding estimated wavelet distributions. We evaluate the performance of the proposed methodology by using a dataset consisting of six different musical genre sets

This paper presents a novel rotation-invariant image retrieval scheme based on a transformation of the texture information via a steerable pyramid. First, we fit the distribution of the subband coefficients using a joint alpha-stable sub-Gaussian model to capture their non-Gaussian behavior. Then, we apply a normalization process in order to Gaussianize the coefficients. As a result, the feature extraction step consists of estimating the covariances between the normalized pyramid coefficients. The similarity between two distinct texture images is measured by minimizing a rotation-invariant version of the Kullback-Leibler Divergence between their corresponding multivariate Gaussian distributions, where the minimization is performed over a set of rotation angles.

This paper presents a novel rotation-invariant image retrieval scheme based on steerable pyramid transforms. First, we model the subband coefficients as sub-Gaussian random vectors to capture their non-Gaussian behavior. Then, we apply a normalization process in order to Gaussianize the coefficients. As a result, the feature extraction step consists of estimating the covariances between the normalized pyramid coefficients. The similarity of two distinct images is measured by minimizing the Kullback-Leibler divergence (KLD) between their corresponding multivariate Gaussian distributions, where the minimization is performed over a set of rotation angles. We provide analytical expressions for the minimum KLD and we demonstrate the effectiveness of our proposed method using a set of real texture images.

In this paper we demonstrate that spectral conversion can be successfully applied to the speech enhancement problem as a feature denoising method. The enhanced spectral features can be used in the context of the Kalman filter for estimating the clean speech signal. In essence, instead of estimating the clean speech features and the clean speech signal using the iterative Kalman filter, we show that is more efficient to initially estimate the clean speech features from the noisy speech features using spectral conversion (using a training speech corpus) and then apply the standard Kalman filter. Our results show an average improvement compared to the iterative Kalman filter that can reach 6 dB in the average segmental output Signal-to-Noise Ratio (SNR), in low input SNR's.

This paper addresses issues that arise in copyright protection systems of digital images, which employ blind watermark verification structures in the discrete cosine transform (DCT) domain. First, we observe that statistical distributions with heavy algebraic tails, such as the alpha-stable family, are in many cases more accurate modeling tools for the DCT coefficients of JPEG-analyzed images than families with exponential tails such as the generalized Gaussian. Motivated by our modeling results, we then design a new processor for blind watermark detection using the Cauchy member of the alpha-stable family. The Cauchy distribution is chosen because it is the only non-Gaussian symmetric alpha-stable distribution that exists in closed form and also because it leads to the design of a nearly optimum detector with robust detection performance. We analyze the performance of the new detector in terms of the associated probabilities of detection and false alarm and we compare it to the performance of the generalized Gaussian detector by performing experiments with various test images.

Synthetic aperture radar (SAR) images are inherently affected by multiplicative speckle noise, which is due to the coherent nature of the scattering phenomenon. This paper proposes a novel Bayesian-based algorithm within the framework of wavelet analysis, which reduces speckle in SAR images while preserving the structural features and textural information of the scene. First, we show that the subband decompositions of logarithmically transformed SAR images are accurately modeled by alpha-stable distributions, a family of heavy-tailed densities. Consequently, we exploit this a priori information by designing a maximum a posteriori (MAP) estimator. We use the alpha-stable model to develop a blind speckle-suppression processor that performs a nonlinear operation on the data and we relate this nonlinearity to the degree of non-Gaussianity of the data. Finally, we compare our proposed method to current state-of-the-art soft thresholding techniques applied on real SAR imagery and we quantify the achieved performance improvement.

Ultrasonic images are generally affected by multiplicative speckle noise, which is due to the coherent nature of the scattering phenomenon. Speckle filtering is thus a critical pre-processing step in medical ultrasound imagery, provided that the features of interest for diagnosis are not lost. We present a novel speckle removal algorithm within the framework of wavelet analysis. First, we show that the subband decompositions of logarithmically transformed ultrasound images are best described by alpha-stable distributions, a family of heavy-tailed densities. Consequently, we design a Bayesian estimator that exploits this a priori information. Using the alpha-stable model we develop a noise-removal processor that performs a nonlinear operation on the data. Finally, we compare our proposed technique to current state-of-the-art speckle reduction methods. Our algorithm effectively reduces speckle, it preserves step edges, and it enhances fine signal details, better than existing methods.

A novel speckle suppression method for medical ultrasound images is presented. First, the logarithmic transform of the original image is analyzed into the multiscale wavelet domain. We show that the subband decompositions of ultrasound images have significantly non-Gaussian statistics that are best described by families of heavy-tailed distributions such as the alpha-stable. Then, we design a Bayesian estimator that exploits these statistics. We use the alpha-stable model to develop a blind noise-removal processor that performs a nonlinear operation on the data. Finally, we compare our technique with current state-of-the-art soft and hard thresholding methods applied on actual ultrasound medical images and we quantify the achieved performance improvement.

In this paper we address the problem of sound source localization in the presence of impulsive noise for application in immersive telepresence and teleconferencing. Traditional Gaussian modeling of noise signals fails when the signals exhibit impulsive behaviour. A new model is used, namely the symmetric /spl alpha/-stable (S/spl alpha/S), which can better account for the outliers that exist in real-world signals. Real data is used to compare the performance of both the Gaussian and the /spl alpha/-stable models. We demonstrate that the astable model gives a much better approximation to the noise signal than the Gaussian model. Furthermore, we study the problem of time delay estimation (TDE) and we demonstrate the shortcomings of TDE techniques based on second-order statistics when the noise is of S/spl alpha/S nature. We propose an alternative to second-order based methods, based on fractional lower-order statistics, and demonstrate the achieved improvement via simulation experiments.

This paper addresses the problem of blind equalization for digital communications using an array of sensors at the receiver to copy constant modulus signals in the presence of heavy-tailed additive channel noise. First, we demonstrate the negative effects of channel noise to the original constant modulus algorithm (CMA) cost function in terms of convergence. Then, we introduce a new CMA criterion based on the fractional lower-order statistics (FLOS) of the received data. The proposed criterion is able to mitigate impulsive noise at the receiver and at the same time restores the constant modulus character of the transmitted communication signal. We perform an analytical study of the properties of the new cost function and we illustrate its convergence behavior through computer simulations.

A new representation of audio noise signals is proposed, based on symmetric /spl alpha/-stable (S/spl alpha/S) distributions in order to better model the outliers that exist in real signals. This representation addresses a shortcoming of the Gaussian model, namely, the fact that it is not well suited for describing signals with impulsive behavior. The /spl alpha/-stable and Gaussian methods are used to model measured noise signals. It is demonstrated that the /spl alpha/-stable distribution, which has heavier tails than the Gaussian distribution, gives a much better approximation to real-world audio signals. The significance of these results is shown by considering the time delay estimation (TDE) problem for source localization in teleimmersion applications. In order to achieve robust sound source localization, a novel time delay estimation approach is proposed. It is based on fractional lower order statistics (FLOS), which mitigate the effects of heavy-tailed noise. An improvement in TDE performance is demonstrated using FLOS that is up to a factor of four better than what can be achieved with second-order statistics.

This paper studies methods for space-time adaptive processing-based (STAP-based) parameter estimation in the presence of impulsive noise backgrounds. Towards this goal, the theory of alpha-stable random processes provides an elegant and mathematically tractable framework for the solution of the detection and parameter estimation problems in the presence of impulsive radar clutter. Our goal is to develop joint target angle and Doppler, maximum likelihood-based estimation techniques from radar measurements retrieved in the presence of severe clutter modeled as an alpha-stable, complex random process. We derive the Cramer-Rao bounds for the additive Cauchy interference scenario to assess the best-case estimation accuracy which can be achieved. The results are of great importance in the study of space-time adaptive processing (STAP) for airborne pulse Doppler radar arrays operating in impulsive interference environments.

We develop an new adaptive beamforming technique based on fractional lower-order moment theory. The proposed adaptive beamformer adjusts the array response to a desired signal while discriminating against impulsive interference modeled as a stable process. Simulation results show that the new technique performs better in localizing a target both in space and Doppler, and thus offers the potential for improved airborne radar performance in space-time adaptive processing (STAP) applications.

There exist real world applications in which impulsive channels tend to produce large amplitude interferences more frequently than Gaussian channels. The stable law has been shown to successfully model noise over certain impulsive channels. The authors propose subspace-based methods for the direction-of-arrival estimation problem in impulsive noise environments. They define the covariation matrix of the array sensor outputs and show that eigendecomposition-based methods, such as the MUSIC algorithm, can be applied to the sample covariation matrix to extract the bearing information from the measurements.

Multivariate data ordering and its use in color image filtering are presented. Several of the filters presented are extensions of the single-channel filters based on order statistics. The statistical analysis of the marginal order statistics is presented for the p-dimensional case. A family of multichannel filters based on multivariate data ordering, such as the marginal median, the vector median, the marginal alpha -trimmed mean, and the multichannel modified trimmed mean filter, is described in detail. The performance of the marginal median and the vector median filters in impulsive noise filtering is investigated. Simulation examples of the filters under study are described.>