Cooperative inference in distributed sensor networks is challenged by limited communication bandwidth and the risk of node failures. This paper introduces Compressed Feature Diffusion for Decentralized Classification (CFD-DC), a novel framework that addresses these challenges. Each node performs local inference using its own features and compressed feature representations received from other nodes. Our approach relies on two key components: first, a trainable feature compressor at each node that learns compact representations, reducing communication while preserving critical discriminative information; second, an adaptive node weighting mechanism that dynamically adjusts the influence of local and remote features, providing robustness to unreliable or failed nodes. Experiments on multi-view image classification and a simulated multi-node underwater acoustic target classification task demonstrate the effectiveness of the framework. The results show competitive performance compared to centralized and state-of-the-art multi-view methods, reduced communication costs, and superior robustness in scenarios with node failures.

Active noise control (ANC) is widely recognized as an effective and efficient solution for attenuating urban noise. Least mean square (LMS)-based adaptive algorithms, particularly the filtered-reference LMS (FxLMS) algorithm, play a central role in adaptive ANC systems due to their computational efficiency and optional steady-state performance. However, their effectiveness heavily depends on appropriate step-size selection. An unsuitable step size can severely degrade convergence speed and stability. Traditional step-size strategies, such as variable step-size approaches, often involve high computational complexity and are limited to specific noise types. To address this, this letter proposes a data-driven step-size selection method for the FxLMS algorithm based on modified model-agnostic meta-learning (MAML), incorporating a forgetting factor to mitigate the filter's initial zero effect. Compared to conventional methods, the proposed approach can determine an optimal step size across various noise types without requiring additional computations during control, making it highly suitable for practical deployment. Numerical simulations using real-world paths and noise further verify its effectiveness.

Weakly-Supervised Anomaly Detection (WSAD) has garnered increasing research interest in recent years, as it enables superior detection performance while demanding only a small fraction of labeled data. However, existing WSAD methods face two major limitations. From the data aspect, they struggle to detect anomalies between normal clusters or collective anomalies due to overlooking the multi-distribution and complex manifolds of real-world data. From the label aspect, they fall short of detecting unknown anomalies because of the label-insufficiency and anomaly contamination. To address these issues, we propose MMM, a unified WSAD framework for multi-distributional data. The framework consists of three components: a Multi-distribution data modeler captures latent representations of complex data distributions, followed by a Multiform feature extractor that extracts multiple underlying features from the modeler, highlighting the characteristics of potential anomalies. Finally, a Multi-strategy anomaly score estimator converts these features into anomaly scores, with the aid of a novel training approach with three strategies that maximize the utility of both data and labels. Experimental results showed that MMM achieved superior performance and robustness compared to state-of-the-art WSAD methods, while providing interpretable results that facilitate practical anomaly analysis.

Self-supervised monocular depth estimation (MDE) has achieved performance levels comparable to supervised methods in well-lit environments. However, current methods struggle particularly with challenging nighttime scenes. Existing all-day self-supervised MDE methods often rely on specialized nighttime datasets, which require extensive data collection and annotation, adding complexity and resource demands to the training process. To overcome this limitation, we propose ADDepth, a novel lightweight all-day self-supervised MDE network. ADDepth leverages CoMoGAN to transform daytime images into nighttime scenes, thereby circumventing the need for a separate nighttime dataset. Additionally, we introduce a low-scale consistency loss to enhance depth map quality by mitigating the issue of blurred depth predictions, a common challenge caused by the reduced number of convolutional kernels in decoder layers. Our approach retains the network’s lightweight design while significantly improving its generalization across different lighting conditions. Experimental results on public benchmarks validate the superiority of the proposed ADDepth. The source code is available at https://github.com/zjdzhou/ADDepth.

Modeling the relationships that may connect optimal parameter vectors is essential for the performance of parameter estimation methods in distributed networks. In this paper, we consider a low-rank relationship and introduce matrix factorization to promote this low-rank property. To devise a distributed algorithm that does not require any prior knowledge about the low-rank space, we first formulate local optimization problems at each node, which are subsequently addressed using the Alternating Direction Method of Multipliers (ADMM). Three subproblems naturally arise from ADMM, each resolved in an online manner with low computational costs. Specifically, the first one is solved using stochastic gradient descent (SGD), while the other two are handled using the conjugate gradient descent method to avoid matrix inversion operations. To further enhance performance, a variance reduction algorithm is incorporated into the SGD. Simulation results validate the effectiveness of the proposed algorithm.

Due to its ability to handle strict constraints on feasible domains, distributed optimization over a Riemannian manifold offers an attractive solution for many practical applications. To develop such an algorithm for scenarios where the explicit expression of the cost function is unavailable, we introduce the zeroth-order (ZO) Riemannian stochastic gradient into distributed optimization on a Riemannian manifold. Specifically, an intermediate estimate is first obtained through a local update step using the ZO Riemannian stochastic gradient, which is ap proximated based on two function evaluations. Subsequently, an improved estimate is derived by minimizing the weighted Fr´echet mean over the manifold using information from neighboring nodes. To further enhance performance, a mini-batch strategy is incorporated into the gradient estimation process. Finally, simulation results are presented to validate the effectiveness of the proposed algorithm.

Data generated by network-structured applications, such as sensor networks or communication networks, typically reside on complex and irregular structures. These data necessitate specific graph signal processing tools to harness their characteristics. Detecting anomalous events in graph signals is significant in enhancing reliability of systems, where anomalies often activate localized groups of vertices. In this paper, we introduce a novel approach, the Joint Graph Wavelet Canonical Correlation Analysis, for detecting anomalies in graph signals through cooperative filtering while identifying their locations. This approach conducts canonical correlation analysis on graph signals to achieve data fusion within the wavelet domain while accounting for the graph topology. Subsequently, we devise an optimization algorithm specifically tailored for anomaly detection in graph signals. Finally, we illustrate its effectiveness through numerical simulations on synthetic data and by presenting test results from a multi-microphone network.

Compared with the functional link neural network (FLNN) algorithm, the delayless multi-sampled multiband-structured subband FLNN (DMSFLNN) algorithm provides fast convergence when encountering highly auto-correlated input signals, but there is a compromise between convergence and steady-state performances. Therefore, in order to overcome this flaw, we develop an optimal DMSFLNN (ODMSFLNN) algorithm by minimizing the mean square deviation of the weight vector with respect to the subband gain vectors. Interestingly, a vectorized version is also proposed for the ODMSFLNN algorithm, which aims at reducing computational complexity. Additionally, this paper also presents a stability analysis of this algorithm. Then, considering the impulsive noise environment, we develop two robust variants of ODMSFLNN that are the R-ODMSFLNN-I and R-ODMSFLNN-II algorithms, which are based on the specified robust function and the energy constraint of the weight update increment, respectively. Finally, to resolve that the DMSFLNN algorithm may not exploit cross-terms of input samples in nonlinear active noise control scenarios, we further propose the subband second-order Volterra filter (SSOVF) framework in an analogy way and apply the R-ODMSFLNN-II learning principle to obtain the robust optimal SSOVF algorithm. Simulations in several nonlinear scenarios have shown that the proposed algorithms perform better than their competitors.

Deep learning (DL) has recently garnered substantial interest in hyperspectral unmixing (HU) due to its exceptional learning capabilities. In particular, unsupervised unmixing methods based on autoencoders have become a research hotspot, with many existing networks focusing on the fusion of spatial and spectral information. However, the diversity of fusion structures makes it challenging to select appropriate modules that meet unmixing requirements, while the issue of endmember variability is often neglected. In this article, we propose a novel spatial-spectral adaptive fusion network (SSAF-Net) that accounts for endmember variability. The network consists of two cascaded encoders and a deep generative model (DGM) based on a variational autoencoder (VAE). The encoders perform local spatial-spectral information fusion through channel and spatial attention mechanisms, respectively, while self-perception loss facilitates global information fusion during the cascading process. In addition, we address endmember variability using a proportional perturbation model (PPM), learning the necessary endmember parameters through an elaborately designed DGM. Our SSAF-Net learns both endmember variability and the corresponding abundances in an unsupervised manner. Experimental results on a synthetic dataset and real-world datasets validate the significant superiority of SSAF-Net compared to other methods. The code for this work is available athttps://github.com/yjysimply/SSAF-Net.

Hyperspectral unmixing (HU) aims to decompose mixed pixels in remote sensing imagery into material-specific spectra and their respective abundance fractions. Recently, autoencoders have made significant advances in HU due to their strong representational capabilities and ease of implementation. However, relying exclusively on feature extraction from a single-modality hyperspectral image can fail to fully utilize both spatial and spectral information, thereby limiting the ability to distinguish objects in complex scenes. To address these limitations, we propose a multimodal spectral-spatial feature fusion network (MSSF-Net) for enhanced HU. MSSF-Net adopts a dual-stream architecture to extract feature representations from complementary input modalities. Specifically, the hyperspectral subnetwork leverages a convolutional neural network (CNN) to capture spatial information, while the light detection and ranging (LiDAR) subnetwork incorporates an enhanced channel attention mechanism (ECAM) to capture the dynamic changes in spatial information across different channels. Furthermore, we introduce a cross-modal fusion (CMF) module that integrates spectral and spatial information across modalities, leading to more robust feature representations. Experimental results indicate that MSSF-Net significantly outperforms existing traditional and deep learning-based methods in terms of unmixing accuracy. The code is available at https://github.com/Octopus-Squidward/MSSF-Net.

Deep learning-based unmixing methods have received great attention in recent years and achieved remarkable performance. These methods employ a data-driven approach to extract structure features from hyperspectral images; however, they tend to be less physically interpretable. Conventional unmixing methods have much more interpretability, whereas they require manually designing regularization and choosing penalty parameters. To overcome these limitations, we propose a novel unmixing method by unrolling the plug-and-play unmixing algorithm to conduct the deep architecture. Our method integrates both inner and outer priors. The carefully designed unfolding deep architecture is used to learn the spectral and spatial information from the hyperspectral image, which we refer to as inner priors. Additionally, our approach incorporates deep denoisers that have been pretrained on a large volume of image data to leverage the outer priors. Second, we design a dynamic convolution to model the multiscale information. Different scales are fused with an attention module. Experimental results of both synthetic and real datasets demonstrate that our method outperforms compared methods.

The present work sought to instil metrology in existing hyperspectral texture feature extraction methods. Specifically, we propose distance-based expressions of graylevel cooccurrence matrix (GLCM), local binary pattern (LBP), and Gabor filtering directly computable for hyperspectral images without any pre- or post-processing. At the core of our proposition is Radical of Extended Mean Information for Discrimination (REID), a novel spectral distance with information-theoretic roots. Respecting the physics of spectrum as continuous function of wavelengths, REID is mathematically decomposable into spectral direction and spectral magnitude distances. The resulted feature calculations are fullband (utilizing all wavelengths), yet lightweight and fully interpretable. A similarity measure based on information theory is also justified. Their efficiency is demonstrated in the context of texture classification, content-based image retrieval, and cancer detection in which they consistently outperform existing computations based on dimensionally reduced space using PCA, ICA, and NMF. The propositions could be potentially integrated into machine/deep learning systems towards explainable AI (XAI).

Hyperspectral unmixing aims to decompose the mixed pixels into pure spectra and calculate their corresponding fractional abundances. It holds a critical position in hyperspectral image processing. Traditional model-based unmixing methods use convex optimization to iteratively solve the unmixing problem with hand-crafted regularizers. While their performance is limited by these manually designed constraints, which may not fully capture the structural information of the data. Recently, deep learning-based unmixing methods have shown remarkable capability for this task. However, they have limited generalizability and lack interpretability. In this paper, we propose a novel hyperspectral unmixing method regularized by a diffusion model (URDM) to overcome these shortcomings. Our method leverages the advantages of both conventional optimization algorithms and deep generative models. Specifically, we formulate the unmixing objective function from a variational perspective and integrate it into a diffusion sampling process to introduce generative priors from a denoising diffusion probabilistic model (DDPM). Since the original objective function is challenging to optimize, we introduce a splitting-based strategy to decouple it into simpler subproblems. Extensive experiment results conducted on both synthetic and real datasets demonstrate the efficiency and superior performance of our proposed method.

As an adaptive beamformer, the Minimum Variance Distortionless Response (MVDR) method has proven its efficiency in separating target speech from background noise and interference. Conventionally, MVDR relies on physical information regarding signal angles and covariance matrices, however, ignores that the beamformer output can potentially benefit from the prior structures of speech signals. Motivated by the recent advance in integrating physics-based and data-driven approaches, this paper introduces a novel speech separation framework. Our approach enhances MVDR by incorporating Plug-and-Play (PnP) techniques to capture speech priors, specifically employing the Regularization by Denoising (RED) method to integrate prior speech information obtained from data into the optimization process. Experimental results validate the effectiveness of the proposed approach.

During the last decade, many methods have been proposed to enhance the performance of hyperspectral unmixing (HU) for linear mixing problems. However, most methods typically do not take into account the effects of spectral variability, limiting their ability to improve unmixing performance. Therefore, we propose a proportional perturbation model (PPM) for HU accounting for endmember variability. The PPM can characterize both the proportional variations of endmembers and the local fluctuations in real-world scenarios by incorporating scaling factors and a perturbation term. In addition, we design an unmixing network based on PPM, so-called PPM-Net. The PPM-Net can learn more accurate endmember parameters from the latent representation of input pixels and estimate abundance simultaneously. Specifically, we constrain the abundance through a traditional method during the pretraining phase to further enhance its robustness. The experimental results on synthetic and real data indicate that the proposed PPM-Net can outperform the state-of-the-art unmixing methods, particularly improving over 5.9% in terms of average root-mean-square error ($\text {aRMSE}_{A}$) over the second best method. The source code is available athttps://github.com/yjysimply/PPM-Net.

Incremental speech enhancement (ISE), with the ability to incrementally adapt to new noise domains, represents a critical yet comparatively under-investigated topic. While the regularization-based method has been proposed to solve the ISE task, it usually suffers from the dilemma wherein the gain of one domain directly entails the loss of another. To solve this issue, we propose an effective paradigm, termed Learning Noise Adapters (LNA), which significantly mitigates the catastrophic domain forgetting phenomenon in the ISE task. In our methodology, we employ a frozen pre-trained model to train and retain a domain-specific adapter for each newly encountered domain, enabling the capture of variations in feature distributions within these domains. Subsequently, our approach involves the development of an unsupervised, training-free noise selector for the inference stage, which is responsible for identifying the domains of test speech samples. A comprehensive experimental validation has substantiated the effectiveness of our approach.

Speech dereverberation aims to alleviate the detrimental effects of late-reverberant components. While the weighted prediction error (WPE) method has shown superior performance in dereverberation, there is still room for further improvement in terms of performance and robustness in complex and noisy environments. Recent research has highlighted the effectiveness of integrating physics-based and data-driven methods, enhancing the performance of various signal processing tasks while maintaining interpretability. Motivated by these advancements, this paper presents a novel dereverberation framework for the single-source case, which incorporates data-driven methods for capturing speech priors within the WPE framework. The plug-and-play (PnP) framework, specifically the regularization by denoising (RED) strategy, is utilized to incorporate speech prior information learnt from data during the optimization problem solving iterations. Experimental results validate the effectiveness of the proposed approach.

Unsupervised spectral unmixing consists of representing each observed pixel as a combination of several pure materials known as endmembers, along with their corresponding abundance fractions. Beyond the linear assumption, various nonlinear unmixing models have been proposed, with the associated optimization problems solved either by traditional optimization algorithms or deep learning techniques. Current deep learning-based nonlinear unmixing mainly focuses on additive, bilinear-based formulations. The multilinear mixing model (MLM) offers a unique perspective by interpreting the reflection process by discrete Markov chains, allowing it to account for interactions between endmembers up to infinite order. However, explicitly simulating the physics of MLM using neural networks has remained a challenging problem. In this paper, we propose a novel autoencoder-based network for unsupervised unmixing based on MLM. Leveraging an elaborate network design, this approach explicitly models the relationships among all model parameters: endmembers, abundances, and transition probability. The network operates in two modes: MLM-1DAE, which considers only pixel-wise spectral information, and MLM-3DAE, which explores spectral-spatial correlations within input patches. Experiments on both the synthetic and real datasets validate the effectiveness of the proposed method, demonstrating competitive performance compared to classic MLM-based solutions. The code is available at https://github.com/ting-Fang09/Hyperspectral-unmixing-MLM-AE.

Spectral unmixing has been extensively studied with a variety of methods and used in many applications. Recently, data-driven techniques with deep learning methods have obtained great attention to spectral unmixing for its superior learning ability to automatically learn the structure information. In particular, autoencoder based architectures are elaborately designed to solve blind unmixing and model complex nonlinear mixtures. Nevertheless, these methods perform unmixing task as black-boxes and lack interpretability. On the other hand, conventional unmixing methods carefully design the regularizer to add explicit information, in which algorithms such as plug-and-play (PnP) strategies utilize off-the-shelf denoisers to plug powerful priors. In this paper, we propose a generic unmixing framework to integrate the autoencoder network with regularization by denoising (RED), named AE-RED. More specially, we decompose the unmixing optimized problem into two subproblems. The first one is solved using deep autoencoders to implicitly regularize the estimates and model the mixture mechanism. The second one leverages the denoiser to bring in the explicit information. In this way, both the characteristics of the deep autoencoder based unmixing methods and priors provided by denoisers are merged into our well-designed framework to enhance the unmixing performance. Experiment results on both synthetic and real data sets show the superiority of our proposed framework compared with state-of-the-art unmixing approaches.

In the application of system identification, not only the output but also the input of the system may be corrupted by noise, which is often characterized by the errors-in-variables (EIV) model. To identify such systems, a gradient-descent total least-squares (GD-TLS) and a maximum total correntropy (MTC) algorithms were proposed. In some scenarios, the weight vector of the unknown system may be sparse, e.g., the echo path in acoustic echo cancelation (AEC). Employing TLS or MTC to estimate such systems may result in slow convergence rate, since they assign the same gain to the update of each weight and therefore cannot make use of the sparsity feature of the system to accelerate convergence. To address the above problem, this article proposes a uniform optimization model for deriving proportionate total adaptive filtering algorithms, and then two proportionate total adaptive filtering algorithms are developed, namely, the proportionate total normalized least mean square (PTNLMS) algorithm for Gaussian noise disturbance and the proportionate MTC (PMTC) algorithm for impulsive noise interference, which are both derived by utilizing the method of Lagrange multipliers. Moreover, this article also makes a steady-state performance analysis of the two proposed algorithms. Simulations are performed to demonstrate the superior performance of the two proposed algorithms and to test the accuracy of the theory on the steady-state performance analysis.

Detecting change points in streaming time series data is a long standing problem in signal processing. A plethora of methods have been proposed to address it, depending on the hypotheses at hand. Non-parametric approaches are particularly interesting as they do not make any assumption on the distribution of data or on the nature of changes. Nevertheless, leveraging recent advances in deep learning to detect change points in time series data is still challenging. In this paper, we propose a change point detection method using an online approach based on neural networks to directly estimate the density-ratio between current and reference windows of the data stream. A variational continual learning framework is employed to train the neural network in an online manner while retaining information learned from past data. This leads to a statistically-principled fully nonparametric framework to detect change points from streaming data. Experimental results with synthetic and real data illustrate the effectiveness of the proposed approach.

This paper proposes a simple and effective algorithm for image enlargement, aims to improve upon the widely-used bi-linear and nearest-neighbour interpolations for monitor displays. The proposed method uses nearest-neighbour interpolation to generate preliminary enlarged images, and selectively modifies only diagonal edges to reduce blocking artifacts. Experiments demonstrate the proposed method produces clear images with fewer artifacts across different enlargement factors and scenes.

Hyperspectral anomaly detection is a widely studied topic that has garnered significant attention in recent years. However, designing effective nonlinear detectors remains a challenge for many traditional methods. To address this issue, we propose the integration of a variational autoencoder (VAE) in this paper. The VAE enables efficient feature extraction from hyperspectral images (HSIs) by mapping inputs to latent variables that follow a Gaussian distribution. The resulting latent representations are subsequently passed to the Reed-Xiaoli (RX) detector to obtain the final detection results. Through extensive testing on three real datasets, the detection results demonstrate the superiority of our proposed method.

Blind hyperspectral unmixing (HU) involves identifying pixel spectra as distinct materials (endmembers) and simultaneously determining their proportions (abundances) at each pixel. In this paper, we present Swin-HU, a novel method based on the Swin Transformer, designed to efficiently tackle blind HU. This method addresses the limitations of existing techniques, such as Convolutional Neural Networks (CNNs) and Vision Transformers (ViT), in capturing global spatial information and spectral sequence attributes. Swin-HU employs Window Multi-head Self-Attention (W-MSA) and Shifted Window Multi-head Self-Attention (SW-MSA) mechanisms to extract global spatial priors while maintaining linear computational complexity. We evaluate Swin-HU against six other unmixing methods on both synthetic and real datasets, demonstrating its superior performance in endmember extraction and abundance estimation. The source code is available at https://github.com/wangyunjeff/Swin-HU.

Current object-detection methods for small-scale objects are often marred by poor performance. Using relatively high-resolution input images can be considered a remedy for this issue, but it usually leads to performance degeneration for large-scale objects. We define this problem as the imbalance of detection performance for multi-scale objects when the resolution of input images varies. In addition, the use of high-resolution images results in significant computational resource consumption and inference-speed impairment. In this paper, we propose a friendly varying-resolution object-detection method for multi-scale objects. We analyze in detail the reasons leading to the performance degradation in the detection of large-scale objects with increasing input-image resolution, and propose a novel lightweight bidirectional feature-flow module to enhance the performance of multi-scale object detection in high-resolution images, especially for large-scale objects. The proposed approach can also ease the problems of computational resource consumption and inference-speed impairment caused by high-resolution images. Additionally, a decoupled detection head is designed to further improve performance by separating classification and regression sub-tasks, and an adaptive feature-fusion module is designed to better fuse different feature levels. The proposed scheme alleviates the negative effects of using high-resolution input images and achieves an excellent balance between inference speed and precision. Experiments on the MS COCO dataset show that the scheme achieves 44.6 AP at 42.6 FPS and 47 AP at 26.7 FPS, showing significant advantages over the methods to which it is compared.

Deconvolution is a widely used strategy to mitigate the blurring and noisy degradation of hyperspectral images (HSI) generated by the acquisition devices. This issue is usually addressed by solving an ill-posed inverse problem. While investigating proper image priors can enhance the deconvolution performance, it is not trivial to handcraft a powerful regularizer and to set the regularization parameters. To address these issues, in this paper we introduce a tuning-free Plug-and-Play (PnP) algorithm for HSI deconvolution. Specifically, we use the alternating direction method of multipliers (ADMM) to decompose the optimization problem into two iterative sub-problems. A flexible blind 3D denoising network (B3DDN) is designed to learn deep priors and to solve the denoising sub-problem with different noise levels. A measure of 3D residual whiteness is then investigated to adjust the penalty parameters when solving the quadratic sub-problems, as well as a stopping criterion. Experimental results on both simulated and real-world data with ground-truth demonstrate the superiority of the proposed method.

Hyperspectral image (HI) and multispectral image (MI) fusion allows us to overcome the hardware limitations of hyperspectral imaging systems inherent to their lower spatial resolution. Nevertheless, existing algorithms usually fail to consider realistic image acquisition conditions. This article presents a general imaging model that considers inter-image variability of data from heterogeneous sources and flexible image priors. The fusion problem is stated as an optimization problem in the maximum a posteriori framework. We introduce an original image fusion method that, on one hand, solves the optimization problem accounting for inter-image variability with an iteratively reweighted scheme and, on the other hand, that leverages lightweight convolutional neural network (CNN)-based networks to learn realistic image priors from data. In addition, we propose a zero-shot strategy to directly learn the image-specific prior of the latent images in an unsupervised manner. The performance of the algorithm is illustrated with real data subject to inter-image variability.

When adopting a model-based formulation, solving inverse problems encountered in multiband imaging requires to define spatial and spectral regularizations. In most of the works of the literature, spectral information is extracted from the observations directly to derive data-driven spectral priors. Conversely, the choice of the spatial regularization often boils down to the use of conventional penalizations (e.g., total variation) promoting expected features of the reconstructed image (e.g., piece-wise constant). In this work, we propose a generic framework able to capitalize on an auxiliary acquisition of high spatial resolution to derive tailored data-driven spatial regularizations. This approach leverages on the ability of deep learning to extract high level features. More precisely, the regularization is conceived as a deep generative network able to encode spatial semantic features contained in this auxiliary image of high spatial resolution. To illustrate the versatility of this approach, it is instantiated to conduct two particular tasks, namely multiband image fusion and multiband image inpainting. Experimental results obtained on these two tasks demonstrate the benefit of this class of informed regularizations when compared to more conventional ones.

In this paper, we propose a novel clustered multitask diffusion RLS (MT-DRLS) algorithm over network to further improve the performance of its counterpart, the multitask diffusion LMS (MT-DLMS) algorithm. Its transient behavior is investigated, in the mean and mean-square error sense. Simulation results illustrate the significant improvement of the MT-DRLS over the MT-DLMS in terms of convergence rate and steady-state error, as well as the accuracy of the theoretical findings.

To overcome inherent hardware limitations of hyperspectral imaging systems with respect to their spatial resolution, fusion-based hyper-spectral image (HSI) super-resolution is attracting increasing attention. This technique aims to fuse a low-resolution (LR) HSI and a conventional high-resolution (HR) RGB image in order to obtain an HR HSI. Recently, deep learning architectures have been used to address the HSI super-resolution problem and have achieved remarkable performance. However, they ignore the degradation model even though this model has a clear physical interpretation and may contribute to improving the performance. We address this problem by proposing a method that, on the one hand, makes use of the linear degradation model in the data-fidelity term of the objective function and, on the other hand, utilizes the output of a convolutional neural network for designing a deep prior regularizer in spectral and spatial gradient domains. Experiments show the performance improvement achieved with this strategy.

The inherent spectral variability in hyperspectral images, the noise, and other factors bring difficulties to traditional detectors to separate the target and background by using linear decision boundaries. In this paper, by generalizing the classical constrained energy minimization (CEM) method, and considering the feature auto-extraction ability of deep neural networks (DNN), we propose a nonlinear detector based on semi-supervised learning (named deepCEM). This approach designs a deep neural network structure to provide a specific form of the nonlinear detector and trains the DNN model with knowledge of target spectra and unlabeled samples. Experiments performed on several hyperspectral data sets show that the proposed method performs better than other state-of-the-art methods.

Properly exploiting image properties is crucial for boosting the hyperspectral unmixing performance. Recent advanced image processing methods use deep architectures to learn image priors. However, these deep priors take effect in an implicit manner and it is nontrivial to characterize their properties. Introducing extra regularization terms is an explicit way of encoding image priors, and the plug-and-play technique enables to construct priors from data by denoisers. In this work, we propose a new unmixing framework to combine both the deep image priors (DIP) and plug-and-play (PnP) priors to further enhance the unmixing performance. The alter-nating direction method of multipliers (ADMM) framework is used to separate the optimization problem into two subproblems. The first one is solved using a U-net training step to obtain DIP, and a proximal denoising step is then used to solve the second subproblem to add denoiser priors. Experiment results demonstrate the effectiveness of our proposed method.

Jointly exploiting spectral information and spatial information, rather than working on individual pixels, is important for hyperspectral target detection. In this letter, we propose a hyperspectral target detection method relying on superpixel structures of the input image. Multiscale superpixels are generated to capture textures of the image, and each superpixel is summarized to its representative, which is the average of all its pixels. The SparseCEM detector is then applied to these representatives. Finally, the detection results from all scales are fused to achieve the final output. Our experiment results show that the multiscale-superpixel-based SparseCEM detector (MSSD) outperforms the compared typical detection methods.

Nowadays, nonnegative matrix factorization (NMF)-based methods have been widely applied to blind spectral unmixing. Introducing proper regularizers to NMF is crucial for mathematically constraining the solutions and physically exploiting spectral and spatial properties of images. Generally, properly handcrafted regularizers and solving the associated complex optimization problem are nontrivial tasks. In our work, we propose an NMF-based unmixing framework which jointly uses a learned regularizer from data and a handcrafted regularizer. To be specific, we plug learned priors of abundances where the associated subproblem can be addressed using various image denoisers, and we consider an$\ell _{2,1}$-norm as an example to illustrate the way of integrating handcrafted regularizers. The proposed framework is flexible and extendable. Both synthetic data and real airborne data are conducted to confirm the effectiveness of our method.

Recently, the use of a deep autoencoder-based method in blind spectral unmixing has attracted great attention as the method can achieve superior performance. However, most autoencoder-based unmixing methods use non-structured reconstruction loss to train networks, leading to the ignorance of band-to-band-dependent characteristics and fine-grained information. To cope with this issue, we propose a general perceptual loss-constrained adversarial autoencoder network for hyperspectral unmixing. Specifically, the adversarial training process is used to update our framework. The discriminate network is found to be efficient in discovering the discrepancy between the reconstructed pixels and their corresponding ground truth. Moreover, the general perceptual loss is combined with the adversarial loss to further improve the consistency of high-level representations. Ablation studies verify the effectiveness of the proposed components of our framework, and experiments with both synthetic and real data illustrate the superiority of our framework when compared with other competing methods.

International audience

Hyperspectral image (HSI) super-resolution is commonly used to overcome the hardware limitations of existing hyperspectral imaging systems on spatial resolution. It fuses a low-resolution (LR) HSI and a high-resolution (HR) conventional image of the same scene to obtain an HR HSI. In this work, we propose a method that integrates a physical model and deep prior information. Specifically, a novel, yet effective two-stream fusion network is designed to serve as a regularizer for the fusion problem. This fusion problem is formulated as an optimization problem whose solution can be obtained by solving a Sylvester equation. Furthermore, the regularization parameter is simultaneously estimated to automatically adjust contribution of the physical model and the learned prior to reconstruct the final HR HSI. Experimental results on both simulated and real data demonstrate the superiority of the proposed method over other state-of-the-art methods on both quantitative and qualitative comparisons.

This article presents a novel Bayesian approach for hyperspectral image unmixing. The observed pixels are modeled by a linear combination of material signatures weighted by their corresponding abundances. A spike-and-slab abundance prior is adopted to promote sparse mixtures and an Ising prior model is used to capture spatial correlation of the mixture support across pixels. We approximate the posterior distribution of the abundances using the expectation-propagation (EP) method. We show that it can significantly reduce the computational complexity of the unmixing stage and meanwhile provide uncertainty measures, compared to expensive Monte Carlo strategies traditionally considered for uncertainty quantification. Moreover, many variational parameters within each EP factor can be updated in a parallel manner, which enables mapping of efficient algorithmic architectures based on graphics processing units (GPUs). Under the same approximate Bayesian framework, we then extend the proposed algorithm to semi-supervised unmixing, whereby the abundances are viewed as latent variables and the expectation-maximization (EM) algorithm is used to refine the endmember matrix. Experimental results on synthetic data and real hyperspectral data illustrate the benefits of the proposed framework over state-of-art linear unmixing methods.

Spectral unmixing is an effective tool to mine information at the subpixel level from complex hyperspectral images. To consider the spatially correlated materials distributions in the scene, many algorithms unmix the data in a spatial–spectral fashion; however, existing models are usually unable to model spectral variability simultaneously. In this article, we present a variational autoencoder-based deep generative model for spatial–spectral unmixing (DGMSSU) with endmember variability, by linking the generated endmembers to the probability distributions of endmember bundles extracted from the hyperspectral imagery via discriminators. Besides the convolutional autoencoder-like architecture that can only model the spatial information within the regular patch inputs, DGMSSU is able to alternatively choose graph convolutional networks or self-attention mechanism modules to handle the irregular but more flexible data—superpixel. Experimental results on a simulated dataset, as well as two well-known real hyperspectral images, show the superiority of our proposed approach in comparison with other state-of-the-art spatial–spectral unmixing methods. Compared to the conventional unmixing methods that consider the endmember variability, our proposed model generates more accurate endmembers on each subimage by the adversarial training process. The codes of this work will be available athttps://github.com/shuaikaishi/DGMSSUfor the sake of reproducibility.

The complex nature of hyperspectral images makes the analysis of spectral signatures a challenging task in remote sensing. For quantitative analysis, spectral unmixing is a well-established and effective tool to analyze the spectra and spatial distribution of substances in the scene. The classical unmixing algorithms usually fail to tackle spectral variability caused by variations in environmental conditions. Many variants based on the linear mixing process have been proposed to tackle this problem; however, the spectral variability modeling capacity of these algorithms is usually insufficient. In this article, we present a probabilistic generative model to address endmember variability and provide more accurate abundance and endmember estimates. The proposed model simultaneously extracts the endmembers and estimates abundances in an unsupervised manner. In particular, it allows fitting arbitrary endmember distributions through the nonlinear modeling capability of neural networks compared to other methods that use parametric endmember variability models. The performance of the proposed approach is evaluated on both synthetic and real datasets. Experimental results show its superiority in comparison with other state-of-the-art methods. The code of this work is available athttps://github.com/shuaikaishi/PGMSUfor the sake of reproducibility.

Hyperspectral unmixing plays an important role in hyperspectral image processing and analysis. It aims to decompose mixed pixels into pure spectral signatures and their associated abundances. The hyperspectral image contains spatial information in neighborhood regions, and spectral signatures existing in the region also have a high correlation. However, most autoencoder (AE)-based unmixing methods are pixel-to-pixel methods and ignore these priors. It is helpful to add spectral–spatial information into unmixing methods. A recent trend to deal with this problem is to use convolutional neural networks (CNNs). Our proposed framework uses 3-D-CNN-based networks to jointly learn spectral–spatial priors. Moreover, previous AE-based unmixing methods use fixed spectral signatures for each pure material. In our work, we use a carefully designed decoder to cope with the endmember variability issue, and variational inference strategy is applied to add uncertainty property into endmembers. To avoid overfitting, we use structured sparsity regularizers to the encoder networks, and$\ell _{2,1}$-loss is added to the estimated abundances to guarantee the sparseness. Experimental results on both simulated and real data demonstrate the effectiveness of our proposed method.

Spectral unmixing is a widely used technique in hyperspectral image processing and analysis. It aims to separate mixed pixels into the component materials and their corresponding abundances. Early solutions to spectral unmixing are performed independently on each pixel. Nowadays, investigating proper priors into the unmixing problem has been popular as it can significantly enhance the unmixing performance. However, it is nontrivial to handcraft a powerful regularizer, and complex regularizers may introduce extra difficulties in solving optimization problems in which they are involved. To address this issue, we present a plug-and-play (PnP) priors framework for hyperspectral unmixing. More specifically, we use the alternating direction method of multipliers (ADMM) to decompose the optimization problem into two iterative subproblems. One is a regular optimization problem depending on the forward model, and the other is a proximity operator related to the prior model and can be regarded as an image denoising problem. Our framework is flexible and extendable which allows a wide range of denoisers to replace prior models and avoids handcrafting regularizers. Experiments conducted on both synthetic data and real airborne data illustrate the superiority of the proposed strategy compared with other state-of-the-art hyperspectral unmixing methods.

Spectral unmixing is an important task in hyperspectral image processing for separating the mixed spectral data pertaining to various materials observed aiming at analyzing the material components in observed pixels. Recently, nonlinear spectral unmixing has received particular attention in hyperspectral image processing, as there are many situations in which the linear mixture model may not be appropriate and could be advantageously replaced by a nonlinear one. Existing nonlinear unmixing approaches are often based on specific assumptions on the nonlinearity and can be less effective when used for scenes with unknown nonlinearity. This article presents an unsupervised nonlinear spectral unmixing method that addresses a general model that consists of a linear mixture part and an additive nonlinear mixture part. The structure of a deep autoencoder network, which has a clear physical interpretation, is specifically designed to achieve this purpose. Moreover, a convolutional neural network (CNN) is used to capture the spectral-spatial priors from hyperspectral data. Extensive experiments with synthetic and real data illustrate the generality and effectiveness of this scheme compared with state-of-the-art methods.

Spectral signatures are usually affected by variations in environmental conditions. The spectral variability is thus one of the most important and challenging problems to be addressed in hyperspectral unmixing. Generally, it is a non-trivial task to model the endmember variability, and existing spectral unmixing methods that address the spectral variability have different limitations. This paper presents a variational autoencoder (VAE) framework for hyperspectral unmixing accounting for the endmember variability. The endmembers are generated using the posterior distributions of the latent variables to describe their variability in the image. Compared with other existing distribution based methods, the proposed method is able to fit an arbitrary distribution of endmembers for each material through the representation capacity of deep neural networks. Evaluated with both synthetic and real datasets, the proposed method shows superior unmixing results compared with other state-of-the-art unmixing methods.

Removing shadows that are often present in remotely sensed hyperspectral images is important for both enhancing the interpretability of the data and further target analysis. Shadow removal approaches based on spectral unmixing have been proposed in the literature using the linear mixture model. However, objects that produce shadows may also introduce light scattering, and the higher order interactions of photons can cause nonlinearity. This letter integrates the nonlinear hyperperspectral unmixing into the unmixing-based shadow removal, and the effects of applying typical nonlinear algorithms within the approach are investigated. The usefulness of nonlinear unmixing in hyperspectral shadow removal is verified based on the results of applications to both laboratory-created real data and actual airborne data.

The variable span linear filters (VSLFs) constitute a unified framework of conventional subspace and linear filtering techniques for noise reduction. The construction of VSLFs, however, relies on the generalized eigendecomposition (GEVD) methods, which are computationally expensive. This in turn stymies the employment of such filters in practical online processing problems. To address this issue, we first propose in this paper a fixed-point iteration technique to extract the generalized eigenvectors. It is based on maximizing the pre-whitened generalized Rayleigh quotient (GRQ). We then integrate this technique with online statistic estimation to construct VSLFs. Our proposed method is computationally efficient and can also harness parallel architectures. To show its effectiveness, we consider a speech enhancement application and compare the results with those of several existing methods.

Convergence of the diffusion RLS (DRLS) algorithm to steady-state has been extensively studied in the literature, whereas no analysis of its transient convergence behavior has been reported yet. In this letter, we conduct a theoretical analysis of the transient behavior of the DRLS algorithm for cyclostationary colored inputs, in the mean and mean-square error sense. The resulting analytical models allows us to thoroughly investigate the convergence behavior of the algorithm over adaptive networks in such complex scenarios. Simulation results support the accuracy and correctness of the theoretical findings.

High spectral resolution of hyperspectral images allows the detection and classification of materials in the observed images. However, existing research on hyperspectral detection mainly focuses on pixel-level study, partially due to the low spatial resolution in typical earth observation applications. With the development of imaging techniques, high-spatial-resolution hyperspectral data can be obtained and object-level detection is necessary for many applications. In this work, the object-based hyperspectral detection problem is formulated, and a convolutional neural network is then designed based on the specific characteristics of this problem. Moreover, a hyperspectral dataset with over 400 high-quality images for object-level target detection is created. Experimental results validate the proposed framework and show its superior performance.

In this paper, the design of multichannel noise reduction filters in the short-time-Fourier-transform (STFT) domain is addressed. By investigating the structure of the linear filter, a set of multichannel iterative noise reduction filters are developed based on the Kronecker product decomposition. Instead of computing a long noise reduction filter, we compute three much shorter sub-filters which are separately applied in the spatial, temporal and frequency dimensions. Consequently, compared with the traditional multichannel STFT-domain noise reduction filters, the proposed approaches have two advantages: 1) significantly lower computational complexity; 2) less past observations are needed to construct the iterative filters, which leads to better tracking ability for the temporal/spatial signal nonstationarity. Experimental results demonstrate the advantages of the developed iterative filters over the traditional ones.

Conducting the spatial active noise control (ANC) in wave-domain has been shown advantageous over conventional point-based methods. In the existing schemes, signals at all error microphones are collected and processed in a centralized manner to update the secondary source driving signals. The high computational complexity of this centralized strategy represents one of the major challenges for applying multi-channel ANC systems in large-scale applications. In order to address this issue, this work presents a distributed wave-domain ANC scheme by resorting to distributed optimization techniques. The global ANC problem is formulated as a sum of local costs, and the diffusion adaptation strategy is subsequently utilized to provide a distributed solution that only requires local information exchanges. Simulation results show that the proposed algorithm achieves sufficiently good performance compared to its centralized counterpart.

Modeling relations between local optimum parameter vectors in multitask networks has attracted much attention over the last years. This work considers a distributed optimization problem for parameter vectors with a jointly sparse structure among nodes, that is, the parameter vectors share the same support set. By introducing an ℓ∞,1-norm penalty at each node, and using a proximal gradient method to minimize the regularized cost, we devise a proximal multitask diffusion LMS algorithm which promotes the joint-sparsity to enhance the estimation performance. Analyses are provided to ensure the stability. Simulation results are presented to highlight the performance.

Given a matrix A with numerical rank k, the two-sided orthogonal decomposition (TSOD) computes a factorization A = UDVT, where U and V are unitary, and D is (upper/lower) triangular. TSOD is rank-revealing as the middle factor D reveals the rank of A. The computation of TSOD, however, is demanding, especially when a low-rank representation of the input matrix is desired. To treat such a case efficiently, in this paper we present an algorithm called randomized pivoted TSOD (RP-TSOD) that constructs a highly accurate approximation to the TSOD decomposition. Key in our work is the exploitation of randomization, and we furnish (i) upper bounds on the error of the low-rank approximation, and (ii) bounds for the canonical angles between the approximate and the exact singular subspaces, which take into account the randomness. Our bounds describe the characteristics and behavior of our proposed algorithm. We validate the effectiveness of our proposed algorithm and devised bounds with synthetic data as well as real data of image reconstruction problem.

Hyperspectral image (HSI) deconvolution is an ill-posed problem aiming at recovering sharp images with tens or hundreds of spectral channels from blurred and noisy observations. In order to successfully conduct the deconvolution, proper priors are required to regularize the optimization problem. However, handcrafting a good regularizer may not be trivial and complex regularizers lead to difficulties in solving the optimization problem. In this paper, we use the alternating direction method of multipliers (ADMM) to decompose the optimization problem into iterative subproblems where the prior only appears in a denoising subproblem. Then a 3D denoising convolutional neural network (3DDnCNN) is designed and trained with data for solving this problem. In this way, the hyperspectral image deconvolution is then solved with a framework that integrates the optimization techniques and deep learning. Experimental results demonstrate the superiority of the proposed method with several blurring settings in both quantitative and qualitative comparisons.

Nonlinear spectral unmixing is a challenging and important task in hyperspectral image analysis. The kernel-based bi-objective non-negative matrix factorization (Bi-NMF) has shown its usefulness in nonlinear unmixing; However, it suffers several issues that prohibit its practical application. In this work, we propose an unsupervised nonlinear unmixing method that overcomes these weaknesses. Specifically, the new method introduces into each pixel a parameter that adjusts the nonlinearity therein. These parameters are jointly optimized with endmembers and abundances, using a carefully designed objective function by multiplicative update rules. Experiments on synthetic and real datasets confirm the effectiveness of the proposed method.

Hyperspectral unmixing aims at separating a mixed pixel into a set of pure spectral signatures and their corresponding fractional abundances. Investigating prior spatial and spectral information to regularize the unmixing problem can effectively improve the estimation performance. However, handcrafting a powerful regularizer is a non-trivial task and complex regularizers introduce extra difficulties in solving the optimization problem. In this paper, we present a flexible spectral unmixing method using plug-and-play priors. This method benefits from the alternating direction method of multipliers (ADMM) to decompose the optimization problem into iterative subproblems and incorporates the image denoisers as prior models in a subproblem. In this form, we can plug in various image denoising operations to bypass handcrafting regularizers. We demonstrate the superiority of the proposed unmixing method comparing with other state-of-the-art methods both on synthetic data and real airborne data.

The aim of spectral unmixing of hyperspectral images is to determine the component materials and their associated abundances from mixed pixels. In this paper, we present sparse linear unmixing via an Expectation-Propagation method based on the classical linear mixing model and a spike-and-slab prior promoting abundance sparsity. The proposed method, which allows approximate uncertainty quantification (UQ), is compared to existing sparse unmixing methods, including Monte Carlo strategies traditionally considered for UQ. Experimental results on synthetic data and real hyperspectral data illustrate the benefits of the proposed algorithm over state-of-art linear unmixing methods.

Near-infrared (NIR) images provide spectral information beyond the visible light spectrum and thus are useful in many applications. However, single-channel NIR images contain less information per pixel than RGB images and lack visibility for human perception. Transforming NIR images to RGB images is necessary for performing further analysis and computer vision tasks. In this work, we propose a novel NIR-to-RGB translation method. It contains two sub-networks and a fusion operator. Specifically, a U-net based neural network is used to learn the texture information while a CycleGAN based neural network is adopted to excavate the color information. Finally, a guided filter based fusion strategy is applied to fuse the outputs of these two neural networks. Experiment results show that our proposed method achieves superior NIR-to-RGB translation performance.

A 4-dimensional (4D) image can be viewed as a stack of volumetric images over channels of observation depth or temporal frames. This data contains rich information at the cost of high demands for storage and transmission resources due to its large volume. In this paper, we present a lossless 4D image compression algorithm by extending CCSDS-123.0-B-1 standard. Instead of separately compressing the volumetric image at each channel of 4D images, the proposed algorithm efficiently exploits redundancy across the fourth dimension of data. Experiments conducted on two types of 4D images demonstrate the effectiveness of the proposed lossless compression method.

Recently, face recognition systems have received significant attention, and there have been many works focused on presentation attacks (PAs). However, the generalization capacity of PAs is still challenging in real scenarios, as the attack samples in the training database may not cover all possible PAs. In this paper, we propose to perform the face presentation attack detection (PAD) with multi-channel images using the convolutional neural network based anomaly detection. Multi-channel images endow us with rich information to distinguish between different mode of attacks, and the anomaly detection based technique ensures the generalization performance. We evaluate the performance of our methods using the wide multi-channel presentation attack (WMCA) dataset.

Exploiting the latent low-rankness of tensors is crucial in tensor denoising. Classically, many methods use the Tucker model to find the low-rank structure of a tensor. Recently, the tensor train (TT) model has drawn wide attention owing to its powerful representation ability, and well-balanced matricization scheme for a tensor, and it has been successfully applied to various problems in signal processing, and machine learning applications. In this letter, we propose a tensor denoising method using the TT singular value decomposition, and information criteria, where we leverage the minimum description length to automatically estimate the TT rank. Furthermore, we establish the relationship between Tucker decomposition, and TT decomposition. In specific, the low Tucker rank of a tensor is the sufficient but unnecessary condition to the low TT rank. It unveils in theory the potential advantages of the TT model in characterizing the latent low-rankness of tensor. Denoising experiments on both synthetic data, and real HSI dataset demonstrate its superiority against Tucker-based methods.

Conducting the spatial active noise control (ANC) in wave-domain has been shown advantageous over conventional point-based methods. In the existing schemes, signals at all error microphones are collected and processed in a centralized manner to update the secondary source driving signals. The high computational complexity of this centralized strategy represents one of the major challenges for applying multi-channel ANC systems in large-scale applications. In order to address this issue, this work presents a distributed wave-domain ANC scheme by resorting to distributed optimization techniques. The global ANC problem is formulated as a sum of local costs, and the diffusion adaptation strategies are subsequently utilized to provide a distributed solution that only requires local information exchanges. Simulation results show that the proposed algorithms achieve sufficiently good performance compared to its centralized counterpart.

This work introduces a novel matrix decomposition method termed Intermingled Randomized Singular Value Decomposition (InR-SVD), along with an InR-SVD variant powered by the power iteration scheme. InR-SVD computes a low-rank approximation to an input matrix by means of random sampling techniques. Given a large and dense m × n matrix, InR-SVD constructs a low-rank approximation with a few passes over the data in O(mnk) floating-point operations, where k is much smaller than m and n. Furthermore, InR-SVD can exploit modern computational platforms and thereby being optimized for maximum efficiency. InR-SVD is applied to synthetic data as well as real data in image reconstruction and robust principal component analysis problems. Simulations show that InR-SVD outperforms existing approaches.

Multichannel active noise control (MC-ANC) aims to cancel low-frequency noise in an enclosure. If noise sources are distributed sparsely in space, adding an ℓ1-norm constraint to the standard MC-ANC helps to reduce the complexity of the system and accelerate the convergence rate. However, the convergence performance of ℓ1-norm constrained MC-ANC (cℓ1-MC-ANC) degrades significantly in reverberant environments. In this paper, we analyze the necessity of using sparsity-inducing algorithms with distinct zero-attracting strengths over loudspeakers, and then derive three algorithms of this kind in the complex domain. Simulation results show that, compared to cℓ1-MC-ANC, the proposed algorithms exhibit faster convergence or higher noise reduction at steady state in both free field and reverberant environments.

Nonlinear spectral unmixing is an important and challenging problem in hyperspectral image processing. Classical nonlinear algorithms are usually derived based on specific assumptions on the nonlinearity. In recent years, deep learning shows its advantage in addressing general nonlinear problems. However, existing ways of using deep neural networks for unmixing are limited and restrictive. In this letter, we develop a novel blind hyperspectral unmixing scheme based on a deep autoencoder network. Both encoder and decoder of the network are carefully designed so that we can conveniently extract estimated endmembers and abundances simultaneously from the nonlinearly mixed data. Because an autoencoder is essentially an unsupervised algorithm, this scheme only relies on the current data and, therefore, does not require additional training. Experimental results validate the proposed scheme and show its superior performance over several existing algorithms.

This paper proposes a hyperspectral image deconvolution algorithm for the online restoration of hyperspectral images as provided by wiskbroom and pushbroom scanning systems. We introduce a least-mean-squares (LMS)-based framework accounting for the convolution kernel noncausality and including nonquadratic (zero attracting and piecewise constant) regularization terms. This results in the so-called sliding block regularized LMS (SBR-LMS), which maintains a linear complexity compatible with real-time processing in industrial applications. A model for the algorithm mean and mean-squares transient behavior is derived and the stability condition is studied. Experiments are conducted to assess the role of each hyper-parameter. A key feature of the proposed SBR-LMS is that it outperforms standard approaches in low SNR scenarios such as ultra-fast scanning.

We study the problem of parametric modeling of network-structured signals with graph filters. To benefit from the properties of several graph shift operators simultaneously, and to enhance interpretability, we investigate combinations of parallel graph filters with different shift operators. Due to their extra degrees of freedom, these models might suffer from over-fitting. We address this problem through a weighted ℓ2-norm regularization formulation to perform model selection by encouraging group sparsity. What makes this formulation interesting is that it is actually a smooth convex optimization problem. Experiments on real-world data structured by undirected and directed graphs show the effectiveness of this method.

In this letter, we first present a rank-revealing matrix factorization algorithm by using randomization called randomized truncated pivoted QLP (RTp-QLP) to approximate an input matrix. For a dense and large n1× n2matrix with numerical rank k, RTp-QLP needs only a few passes over the matrix (regardless of k) and O(n1n2d) floating-point operations, where d is much smaller than both n1and n2. Next, we develop a robust principal component analysis (RPCA) method by utilizing RTp-QLP. In addition, we propose a rank estimation technique that efficiently solves the RPCA task. RTp-QLP is highly accurate and numerically stable. Our proposed RTp-QLP-based RPCA method yields the optimal solution, and it is faster than existing methods. Our simulation results support our claims.

The design of variable span linear filters for noise reduction involves a generalized eigenvalue decomposition problem that is of high computational complexity. In order to address this issue, this work proposes a recursive algorithm that computes the filter weights with streaming signal data. Specifically, the inverse square root of the noise covariance matrix is recursively computed with a rank-one update strategy, and the generalized eigenvalues and eigenvectors are approached with the projection approximation subspace tracking method. Numerical simulations show that the proposed recursive method is able to achieve satisfactory performance with significantly lower complexity as compared to the batch algorithm.

In this paper, we design and analyze a new zeroth-order online algorithm, namely, the zeroth-order online alternating direction method of multipliers (ZOO-ADMM), which enjoys dual advantages of being gradient-free operation and employing the ADMM to accommodate complex structured regularizers. Compared to the first-order gradient-based online algorithm, we show that ZOO-ADMM requires $\sqrt{m}$ times more iterations, leading to a convergence rate of $O(\sqrt{m}/\sqrt{T})$, where $m$ is the number of optimization variables, and $T$ is the number of iterations. To accelerate ZOO-ADMM, we propose two minibatch strategies: gradient sample averaging and observation averaging, resulting in an improved convergence rate of $O(\sqrt{1+q^{-1}m}/\sqrt{T})$, where $q$ is the minibatch size. In addition to convergence analysis, we also demonstrate ZOO-ADMM to applications in signal processing, statistics, and machine learning.

Diffusion adaptation is an efficient strategy to perform distributed estimation over networks with streaming data. Existing diffusion-based estimation algorithms require the knowledge of analytical forms of the cost functions or their gradients associated with agents. This setting can be restrictive for practical applications where gradient calculation is difficult or systems operate in a black-box manner. Motivated by the advance of the zeroth-order (gradient-free) optimization, in this work we propose the zeroth-order (ZO) diffusion strategy using randomized gradient estimates. We also examine the stability conditions of the proposed ZO-diffusion strategy. Simulations are performed to examine properties of the algorithm and to compare it with its non-cooperative and stochastic gradient counterparts.

International audience

Most works on graph signal processing assume static graph signals, which is a limitation even in comparison to traditional DSP techniques where signals are modeled as sequences that evolve over time. For broader applicability, it is necessary to develop techniques that are able to process dynamic or streaming data. Many earlier works on adaptive networks have addressed problems related to this challenge by developing effective strategies that are particularly well-suited to data streaming into graphs. We are thus faced with two paradigms: one where signals are modeled as static and sitting on the graph nodes, and another where signals are modeled as dynamic and streaming into the graph nodes. The objective of this work is to blend these concepts and propose diffusion strategies for adaptively learning from streaming graph signals.

Constrained energy minimization (CEM) is a popular method for target detection in hyperspectral images. Its variant SparseCEM uses a sparsity regularization term to promote the sparsity of the detection output. However, these approaches do not consider the spatial correlation of hyperspectral pixels, and target detection can further benefit from exploiting the spatial information. In this paper, we propose a novel constrained detection algorithm, referred to as Spatial-SparseCEM, to simultaneously force the sparsity of the output and piecewise continuity via proper regularizations. The formulated problem is solved efficiently by using alternating direction method of multipliers (ADMM). We illustrate the enhanced performance of the Spatial-SparseCEM algorithm via both synthetic and real hyperspectral data.

Spectral unmixing is one of the most important issues of hyperspectral data processing. However, the lack of publicly available dataset with ground-truth makes it difficult to evaluate and compare the performance of unmixing algorithms. In this work, we create several experimental scenes in our laboratory with controlled settings where the pure material spectra and material compositions are known. Lab-made hyperspectral datasets with these scenes are then provided, and mutually validated with typical linear and nonlinear unmixing algorithms.

This letter proposes a novel kernel least mean p-power (KLMP) algorithm for nonlinear system identification in the presence of additive non-Gaussian impulsive noises, modeled by a symmetric α-stable distribution with heavy tail. The KLMP algorithm based on the fractional lower order statistics error criterion can effectively scale down the dynamic recursive weight coefficients affected by the impulsive estimation error to avoid the significant performance degradation. Simulation results demonstrate that the proposed algorithm has favorable convergence properties than the classical kernel least-mean-square algorithm using a conventional error criterion in the non-Gaussian impulsive environment.

Considering groups of variables, rather than variables individually, can be beneficial for estimation accuracy if structural relationships between variables exist (e.g., spatial, hierarchical or related to the physics of the problem). Group-sparsity inducing estimators are typical examples that benefit from such type of prior knowledge. Building on this principle, we show that the diffusion LMS algorithm for distributed inference over networks can be extended to deal with structured criteria built upon groups of variables, leading to a flexible framework that can encode various structures in the parameters to estimate. We also propose an unsupervised online strategy to differentially promote or inhibit collaborations between nodes depending on the group of variables at hand.

Diffusion LMS is an efficient strategy for solving distributed optimization problems with cooperating agents. In some applications, the optimum parameter vectors may not be the same for all agents. Moreover, agents usually exchange information through noisy communication links. In this work, we analyze the theoretical performance of the single-task diffusion LMS when it is run, intentionally or unintentionally, in a multitask environment in the presence of noisy links. To reduce the impact of these nuisance factors, we introduce an improved strategy that allows the agents to promote or reduce exchanges of information with their neighbors.

Hyperspectral data unmixing has attracted considerable attention in recent years. Hyperspectral data may however suffer from varying levels of signal-to-noise ratio over spectral bands. In this paper, we investigate a robust approach for nonlinear hyperspectral data unmixing. Each observed pixel is modeled as a linear mixing of endmember spectra with nonlinear fluctuations embedded in a reproducing kernel Hilbert space. Welsch M-estimator is considered for reducing the sensitivity of the unmixing process. Experimental results, with both synthetic and real data, illustrate the effectiveness of the proposed scheme.

MOTIVATION: Topological domains have been proposed as the backbone of interphase chromosome structure. They are regions of high local contact frequency separated by sharp boundaries. Genes within a domain often have correlated transcription. In this paper, we present a computational efficient spectral algorithm to identify topological domains from chromosome conformation data (Hi-C data). We consider the genome as a weighted graph with vertices defined by loci on a chromosome and the edge weights given by interaction frequency between two loci. Laplacian-based graph segmentation is then applied iteratively to obtain the domains at the given compactness level. Comparison with algorithms in the literature shows the advantage of the proposed strategy. RESULTS: An efficient algorithm is presented to identify topological domains from the Hi-C matrix. AVAILABILITY AND IMPLEMENTATION: The Matlab source code and illustrative examples are available at http://bionetworks.ccmb.med.umich.edu/ CONTACT: : [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Zero-attracting least-mean-square (ZA-LMS) algorithm has been widely used for online sparse system identification. It combines the LMS framework and l1-norm regularization to promote sparsity, and relies on subgradient iterations. Despite the significant interest in ZA-LMS, few works analyzed its transient behavior. The main difficulty lies in the nonlinearity of the update rule. In this study, a detailed analysis in the mean and mean-square sense is carried out in order to examine the behavior of the algorithm. Simulation results illustrate the accuracy of the model and highlight its performance through comparisons with an existing model.

Complex kernel-based adaptive algorithms have been recently introduced for complex-valued nonlinear system identification. These algorithms are built upon the same framework as complex linear adaptive filtering techniques and Wirtinger's calculus in complex reproducing kernel Hilbert spaces. In this paper, we study the convergence behavior of the augmented complex Gaussian KLMS algorithm. Simulation results illustrate the accuracy of the analysis.

The kernel least-mean-square (KLMS) algorithm is an appealing tool for online identification of nonlinear systems due to its simplicity and robustness. In addition to choosing a reproducing kernel and setting filter parameters, designing a KLMS adaptive filter requires to select a so-called dictionary in order to get a finite-order model. This dictionary has a significant impact on performance, and requires careful consideration. Theoretical analysis of KLMS as a function of dictionary setting has rarely, if ever, been addressed in the literature. In an analysis previously published by the authors, the dictionary elements were assumed to be governed by the same probability density function of the input data. In this paper, we modify this study by considering the dictionary as part of the filter parameters to be set. This theoretical analysis paves the way for future investigations on KLMS dictionary design.

Incorporating spatial information into hyperspectral unmixing procedures has been shown to have positive effects, due to the inherent spatial-spectral duality in hyperspectral scenes. Current research works that consider spatial information are mainly focused on the linear mixing model. In this paper, we investigate a variational approach to incorporating spatial correlation into a nonlinear unmixing procedure. A nonlinear algorithm operating in reproducing kernel Hilbert spaces, associated with an ℓ1local variation norm as the spatial regularizer, is derived. Experimental results, with both synthetic and real data, illustrate the effectiveness of the proposed scheme.

Recent research works on distributed adaptive networks have intensively studied the case where the nodes estimate a common parameter vector collaboratively. However, there are many applications that are multitask-oriented in the sense that there are multiple parameter vectors that need to be inferred simultaneously. In this paper, we employ diffusion strategies to develop distributed algorithms that address clustered multitask problems by minimizing an appropriate mean-square error criterion with ℓ2-regularization. Some results on the mean-square stability and convergence of the algorithm are also provided. Simulations are conducted to illustrate the theoretical findings.

The Non-Negative Least-Mean-Square (NNLMS) algorithm and its variants have been proposed for online estimation under non-negativity constraints. The transient behavior of the NNLMS, Normalized NNLMS, Exponential NNLMS and Sign-Sign NNLMS algorithms have been studied in the literature. In this letter, we derive closed-form expressions for the steady-state excess mean-square error (EMSE) for the four algorithms. Simulation results illustrate the accuracy of the theoretical results. This work complements the understanding of the behavior of these algorithms.

Integrating spatial information into hyperspectral unmixing procedures has been shown to have a positive effect on the estimation of fractional abundances due to the inherent spatial-spectral duality in hyperspectral scenes. However, current research works that take spatial information into account are mainly focused on the linear mixing model. In this paper, we investigate how to incorporate spatial correlation into a nonlinear abundance estimation process. A nonlinear unmixing algorithm operating in reproducing kernel Hilbert spaces, coupled with a l1-type spatial regularization, is derived. Experiment results, with both synthetic and real hyperspectral images, illustrate the effectiveness of the proposed scheme.

In recent years, nonlinear unmixing of hyperspectral data has become an attractive topic in hyperspectral image analysis, because nonlinear models appear as more appropriate to represent photon interactions in real scenes. For this challenging problem, nonlinear methods operating in reproducing kernel Hilbert spaces have shown particular advantages. In this paper, we derive an efficient nonlinear unmixing algorithm based on a recently proposed linear mixture/ nonlinear fluctuation model. A multi-kernel learning support vector regressor is established to determine material abundances and nonlinear fluctuations. Moreover, a low complexity locally-spatial regularizer is incorporated to enhance the unmixing performance. Experiments with synthetic and real data illustrate the effectiveness of the proposed method.

Nonlinear adaptive filtering with kernels has become a topic of high interest over the last decade. A characteristics of kernel-based techniques is that they deal with kernel expansions whose number of terms is equal to the number of input data, making them unsuitable for online applications. Kernel-based adaptive filtering algorithms generally rely on a two-stage process at each iteration: a model order control stage that limits the increase in the number of terms by including only valuable kernels into the so-called dictionary, and a filter parameter update stage. It is surprising to note that most existing strategies for dictionary update can only incorporate new elements into the dictionary. This unfortunately means that they cannot discard obsolete kernel functions, within the context of a time-varying environment in particular. Recently, to remedy this drawback, it has been proposed to associate an ℓ1-norm regularization criterion with the mean-square error criterion. The aim of this paper is to provide theoretical results on the convergence of this approach.

Spectral model is simple and efficient for modeling the rain attenuation which occurs in satellite communication channels. The prediction of this attenuation series is a vital step for adaptive coding or adaptive power control, which can improve the efficiency of a communication system. In simulation tasks, the discretized spectral model is usually used for generating the attenuation sequence. Due to this reason, in this paper we derive the conditional probability distribution of the predicted attenuation based on the discretized spectral model. This predictor can be used as a bound for others linear or nonlinear predictor of this model.