The sparse unmixing of hyperspectral data has attracted much attention in recent years because it does not need to estimate the number of endmembers nor consider the lack of pure pixels in a given hyperspectral scene. However, the high mutual coherence of spectral libraries strongly affects the practicality of sparse unmixing. The collaborative sparse unmixing via variable splitting and augmented Lagrangian (CLSUnSAL) algorithm is a classic sparse unmixing algorithm that performs better than other sparse unmixing methods. In this paper, we propose a CLSUnSAL-based hyperspectral unmixing method based on dictionary pruning and reweighted sparse regression. First, the algorithm identifies a subset of the original library elements using a dictionary pruning strategy. Second, we present a weighted sparse regression algorithm based on CLSUnSAL to further enhance the sparsity of endmember spectra in a given library. Third, we apply the weighted sparse regression algorithm on the pruned spectral library. The effectiveness of the proposed algorithm is demonstrated on both simulated and real hyperspectral datasets. For simulated data cubes (DC1, DC2 and DC3), the number of the pruned spectral library elements is reduced by at least 94% and the runtime of the proposed algorithm is less than 10% of that of CLSUnSAL. For simulated DC4 and DC5, the runtime of the proposed algorithm is less than 15% of that of CLSUnSAL. For the real hyperspectral datasets, the pruned spectral library successfully reduces the original dictionary size by 76% and the runtime of the proposed algorithm is 11.21% of that of CLSUnSAL. These experimental results show that our proposed algorithm not only substantially improves the accuracy of unmixing solutions but is also much faster than some other state-of-the-art sparse unmixing algorithms.

FPGA-based switches are appealing nowadays due to the balance between hardware performance and software flexibility. Packet parser, as the foundational component of FPGA-based switches, is to identify and extract specific fields used in forwarding decisions, e.g., destination IP address. However, traditional parsers are too rigid to accommodate new protocols. In addition, FPGAs usually have a much lower clock frequency and fewer hardware resources, compared to ASICs. In this paper, we present PLANET, a programmable packet-level parallel parsing architecture for FPGA-based switches, to overcome these two limitations. First, PLANET has flexible programmability of updating parsing algorithms at run-time. Second, PLANET highly exploits parallelism inside packet parsing to compensate FPGA's low clock frequency and reduces resource consumption with one-block recycling design. We implemented PLANET on an FPGA-based switch prototype with well-integrated datacenter protocols. Evaluation results show that our design can parse packets at up to 100 Gbps, as well as maintain a relative low parsing latency and fewer hardware resources than existing proposals.

Image quality depends on the randomness of the k-space signal under-sampling in compressed sensing MRI (CS-MRI), especially for two-dimensional image acquisition. We investigate the feasibility of non-random signal under-sampling CS-MRI to stabilize the quality of reconstructed images and avoid arbitrariness in sampling point selection. Regular signal under-sampling for the phase-encoding direction is adopted, in which sampling points are chosen at equal intervals for the phase-encoding direction while varying the sampling density. Curvelet transform was adopted to remove the aliasing artifacts due to regular signal under-sampling. To increase the incoherence between the measurement matrix and the sparsifying transform function, the scale of the curvelet transform was varied in each iterative image reconstruction step. We evaluated the obtained images by the peak-signal-to-noise ratio and root mean squared error in localized 3×3 pixel regions. Simulation studies and experiments showed that the signal-to-noise ratio and the structural similarity index of reconstructed images were comparable to standard random under-sampling CS. This study demonstrated the feasibility of non-random under-sampling based CS by using the multi-scale curvelet transform as a sparsifying transform function. The technique may help to stabilize the obtained image quality in CS-MRI.

In compressive sensing theory (CS), the restricted isometry property (RIP) is commonly used for the measurement matrix to guarantee the reliable recovery of sparse signals from linear measurements. Although many works have indicated that random matrices with excellent recovery performance satisfy the RIP with high probability, Toeplitz-structured matrices arise naturally in real scenarios, such as applications of linear time-invariant systems. Thus, the corresponding measurement matrix can be modeled as a Toeplitz (partial) structured matrix instead of a completely random matrix. The structure characteristics introduce coherence and cause the performance degradation of the measurement matrix. To enhance the recovery performance of the Toeplitz structured measurement matrix in multichannel convolution source separation, an efficient construction of measurement matrix is presented, referred to as sparse random block-banded Toeplitz matrix (SRBT). The sparse signal is pre-randomized by locally scrambling its sample locations. Then, the signal is subsampled using the sparse random banded matrix. Finally, the mixing measurements are obtained. Based on the analysis of eigenvalues, the theoretical results indicate that the SRBT matrix satisfies the RIP with high probability. Simulation results show that the SRBT matrix almost matches the recovery performance of random matrices. Compared with the existing banded block Toeplitz matrix, SRBT significantly improves the probability of successful recovery. Additionally, SRBT has the advantages of low storage requirements and fast computation in reconstruction.

Cell phones with GPS function as well as GPS loggers are widely used and we can easily obtain users' geographic information. Now classifying the measured GPS positions into indoor/outdoor positions is one of the major challenges. In this letter, we propose a robust indoor/outdoor detection method based on sparse GPS measured positions utilizing machine learning. Given a set of clusters of measured positions whose center position shows the user's estimated stayed position, we calculate the feature values composed of: positioning accuracy, spatial features, and temporal feature of measured positions included in every cluster. Then a random forest classifier learns these feature values of the known data set. Finally, we classify the unknown clusters of measured positions into indoor/outdoor clusters using the learned random forest classifier. The experiments demonstrate that our proposed method realizes the maximum F1 measure of 1.000, which classifies measured positions into indoor/outdoor ones with almost no errors.

This paper proposes an enhanced affinity graph (EA-graph) for image segmentation. Firstly, the original image is over-segmented to obtain several sets of superpixels with different scales, and the color and texture features of the superpixels are extracted. Then, the similarity relationship between neighborhood superpixels is used to construct the local affinity graph. Meanwhile, the global affinity graph is obtained by sparse reconstruction among all superpixels. The local affinity graph and global affinity graph are superimposed to obtain an enhanced affinity graph for eliminating the influences of noise and isolated regions in the image. Finally, a bipartite graph is introduced to express the affiliation between pixels and superpixels, and segmentation is performed using a spectral clustering algorithm. Experimental results on the Berkeley segmentation database demonstrate that our method achieves significantly better performance compared to state-of-the-art algorithms.

In sparsity-based optimization problems for two dimensional (2-D) direction-of-arrival (DOA) estimation using L-shaped nested arrays, one of the major issues is computational complexity. A 2-D DOA estimation algorithm is proposed based on reconsitution sparse Bayesian learning (RSBL) and cross covariance matrix decomposition. A single measurement vector (SMV) model is obtained by the difference coarray corresponding to one-dimensional nested array. Through spatial smoothing, the signal measurement vector is transformed into a multiple measurement vector (MMV) matrix. The signal matrix is separated by singular values decomposition (SVD) of the matrix. Using this method, the dimensionality of the sensing matrix and data size can be reduced. The sparse Bayesian learning algorithm is used to estimate one-dimensional angles. By using the one-dimensional angle estimations, the steering vector matrix is reconstructed. The cross covariance matrix of two dimensions is decomposed and transformed. Then the closed expression of the steering vector matrix of another dimension is derived, and the angles are estimated. Automatic pairing can be achieved in two dimensions. Through the proposed algorithm, the 2-D search problem is transformed into a one-dimensional search problem and a matrix transformation problem. Simulations show that the proposed algorithm has better angle estimation accuracy than the traditional two-dimensional direction finding algorithm at low signal-to-noise ratio and few samples.

We formulate the design of an optimal quantizer as an optimization problem that finds the quantization indices that minimize quantization error. As a solution of the optimization problem, an approach based on dynamic programming, which is called DP quantization, is proposed. It is observed that quantized signals do not always contain all kinds of signal values which can be represented with given bit-depth. This property is called amplitude sparseness. Because quantization is the amplitude discretization of signal value, amplitude sparseness is closely related to quantizer design. Signal values with zero frequency do not impact quantization error, so there is the potential to reduce the complexity of the optimal quantizer by not computing signal values that have zero frequency. However, conventional methods for DP quantization were not designed to consider amplitude sparseness, and so fail to reduce complexity. The proposed algorithm offers a reduced complexity optimal quantizer that minimizes quantization error while addressing amplitude sparseness. Experimental results show that the proposed algorithm can achieve complexity reduction over conventional DP quantization by 82.9 to 84.2% on average.

Sparse representation has been successfully applied to visual tracking. Recent progresses in sparse tracking are mainly made within the particle filter framework. However, most sparse trackers need to extract complex feature representations for each particle in the limited sample space, leading to expensive computation cost and yielding inferior tracking performance. To deal with the above issues, we propose a novel sparse tracking method based on the circulant reverse lasso model. Benefiting from the properties of circulant matrices, densely sampled target candidates are implicitly generated by cyclically shifting the base feature descriptors, and then embedded into a reverse sparse reconstruction model as a dictionary to encode a robust appearance template. The alternating direction method of multipliers is employed for solving the reverse sparse model and the optimization process can be efficiently solved in the frequency domain, which enables the proposed tracker to run in real-time. The calculated sparse coefficient map represents the similarity scores between the template and circular shifted samples. Thus the target location can be directly predicted according to the coordinates of the peak coefficient. A scale-aware template updating strategy is combined with the correlation filter template learning to take into account both appearance deformations and scale variations. Both quantitative and qualitative evaluations on two challenging tracking benchmarks demonstrate that the proposed algorithm performs favorably against several state-of-the-art sparse representation based tracking methods.

In order to improve an imaging performance of a sparse array radar system we propose an optimization method to find a new antenna array layout. The method searches for a minimum of the cost function based on a 3D point spread function of the array. We found a solution for the simulated problem in a form of the new layout for the antenna array with more sparse middle-point distribution comparing with initial one.

An autoencoder has the potential ability to compress and decompress information. In this work, we consider the process of generating a stego-image from an original image and watermarks as compression, and the process of recovering the original image and watermarks from the stego-image as decompression. We propose embedder and extractor neural networks based on the autoencoder. The embedder network learns mapping from the DCT coefficients of the original image and a watermark to those of the stego-image. The extractor network learns mapping from the DCT coefficients of the stego-image to the watermark. Once the proposed neural network has been trained, the network can embed and extract the watermark into unlearned test images. We investigated the relation between the number of neurons and network performance by computer simulations and found that the trained neural network could provide high-quality stego-images and watermarks with few errors. We also evaluated the robustness against JPEG compression and found that, when suitable parameters were used, the watermarks were extracted with an average BER lower than 0.01 and image quality over 35 dB when the quality factor Q was over 50. We also investigated how to represent the watermarks in the stego-image by our neural network. There are two possibilities: distributed representation and sparse representation. From the results of investigation into the output of the stego layer (3rd layer), we found that the distributed representation emerged at an early learning step and then sparse representation came out at a later step.

Side scan sonar using low frequency can quickly search a wide range, but the images acquired are of low quality. The image super resolution (SR) method can mitigate this problem. The SR method typically uses sparse coding, but accurately estimating sparse coefficients incurs substantial computational costs. To reduce processing time, we propose a region-selective sparse coding based SR system that emphasizes object regions. In particular, the region that contains interesting objects is detected for side scan sonar based underwater images so that the subsequent sparse coding based SR process can be selectively applied. Effectiveness of the proposed method is verified by the reduced processing time required for image reconstruction yet preserving the same level of visual quality as conventional methods.

We propose an approach to allocate bandwidth for a satellite communications (SATCOM) system that includes the recent high-throughput satellite (HTS) with frequency flexibility. To efficiently operate the system, we manage the limited bandwidth resources available for SATCOM by employing a control method that allows the allocated bandwidths to exceed the communication demand of user terminals per HTS beam. To this end, we consider bandwidth allocation for SATCOM as an optimal control problem. Then, assuming that the model of communication requests is available, we propose an optimal control method by combining model predictive control and sparse optimization. The resulting control method enables the efficient use of the limited bandwidth and reduces the bandwidth loss and number of control actions for the HTS compared to a setup with conventional frequency allocation and no frequency flexibility. Furthermore, the proposed method allows to allocate bandwidth depending on various control objectives and beam priorities by tuning the corresponding weighting matrices. These findings were verified through numerical simulations by using a simple time variation model of the communication requests and predicted aircraft communication demand obtained from the analysis of actual flight tracking data.

Sparse code multiple access (SCMA) based on the message passing algorithm (MPA) for multiuser detection is a competitive non-orthogonal multiple access technique for fifth-generation wireless communication networks Among the existing multiuser detection schemes for uplink (UP) SCMA systems, the serial MPA (S-MPA) scheme, where messages are updated sequentially, generally converges faster than the conventional MPA (C-MPA) scheme, where all messages are updated in a parallel manner. In this paper, the optimization of message scheduling in the S-MPA scheme is proposed. Firstly, some statistical results for the probability density function (PDF) of the received signal are obtained at various signal-to-noise ratios (SNR) by using the Monte Carlo method. Then, based on the non-orthogonal property of SCMA, the data mapping relationship between resource nodes and user nodes is comprehensively analyzed. A partial codeword transmission of S-MPA (PCTS-MPA) with threshold decision scheme of PDF is proposed and verified. Simulations show that the proposed PCTS-MPA not only reduces the complexity of MPA without changing the bit error ratio (BER), but also has a faster convergence than S-MPA, especially at high SNR values.

It is essential to improve intra prediction performance to raise the efficiency of video coding. In video coding standards such as H.265/HEVC, intra prediction is seen as an extension of directional prediction schemes, examples include refinement of directions, planar extension, filtering reference sampling, and so on. From the view point of reducing prediction error, some improvements on intra prediction for standardized schemes have been suggested. However, on the assumption that the correlation between neighboring pixels are static, these conventional methods use pre-defined predictors regardless of the image being encoded. Therefore, these conventional methods cannot reduce prediction error if the images break the assumption made in prediction design. On the other hand, adaptive predictors that change the image being encoded may offer poor coding efficiency due to the overhead of the additional information needed for adaptivity. This paper proposes an adaptive intra prediction scheme that resolves the trade-off between prediction error and adaptivity overhead. The proposed scheme is formulated as a constrained optimization problem that minimizes prediction error under sparsity constraints on the prediction coefficients. In order to solve this problem, a novel solver is introduced as an extension of LARS for multi-class support. Experiments show that the proposed scheme can reduce the amount of encoded bits by 1.21% to 3.24% on average compared to HM16.7.

Binary sparse representation based on arbitrary quality metrics and its applications are presented in this paper. The novelties of the proposed method are twofold. First, the proposed method newly derives sparse representation for which representation coefficients are binary values, and this enables selection of arbitrary image quality metrics. This new sparse representation can generate quality metric-independent subspaces with simplification of the calculation procedures. Second, visual saliency is used in the proposed method for pooling the quality values obtained for all of the parts within target images. This approach enables visually pleasant approximation of the target images more successfully. By introducing the above two novel approaches, successful image approximation considering human perception becomes feasible. Since the proposed method can provide lower-dimensional subspaces that are obtained by better image quality metrics, realization of several image reconstruction tasks can be expected. Experimental results showed high performance of the proposed method in terms of two image reconstruction tasks, image inpainting and super-resolution.

This paper addresses joint estimation of the frequency and the direction-of-arrival (DOA), under the relaxed condition that both snapshots in the temporal domain and sensors in the spacial domain are sparsely spaced. Specifically, a novel coprime sparse array allowing a large range for interelement spacings is employed in the proposed joint scheme, which greatly alleviates the conventional array's half-wavelength constraint. Further, by incorporating small-sized DFT spectrum correction with the closed-form robust Chinese Remainder Theorem (CRT), both spectral aliasing and integer phase ambiguity caused by spatio-temporal under-sampling can be removed in an efficient way. As a result, these two parameters can be efficiently estimated by reusing the observation data collected in parallel at different undersampling rates, which remarkably improves the data utilization. Numerical results demonstrate that the proposed joint scheme is highly accurate.

In recent years, deep learning based approaches have substantially improved the performance of face recognition. Most existing deep learning techniques work well, but neglect effective utilization of face correlation information. The resulting performance loss is noteworthy for personal appearance variations caused by factors such as illumination, pose, occlusion, and misalignment. We believe that face correlation information should be introduced to solve this network performance problem originating from by intra-personal variations. Recently, graph deep learning approaches have emerged for representing structured graph data. A graph is a powerful tool for representing complex information of the face image. In this paper, we survey the recent research related to the graph structure of Convolutional Neural Networks and try to devise a definition of graph structure included in Compressed Sensing and Deep Learning. This paper devoted to the story explain of two properties of our graph - sparse and depth. Sparse can be advantageous since features are more likely to be linearly separable and they are more robust. The depth means that this is a multi-resolution multi-channel learning process. We think that sparse graph based deep neural network can more effectively make similar objects to attract each other, the relative, different objects mutually exclusive, similar to a better sparse multi-resolution clustering. Based on this concept, we propose a sparse graph representation based on the face correlation information that is embedded via the sparse reconstruction and deep learning within an irregular domain. The resulting classification is remarkably robust. The proposed method achieves high recognition rates of 99.61% (94.67%) on the benchmark LFW (YTF) facial evaluation database.

Sparse representation has been widely applied to visual tracking for several years. In the sparse representation framework, tracking problem is transferred into solving an L1 minimization issue. However, during the tracking procedure, the appearance of target was affected by external environment. Therefore, we proposed a robust tracking algorithm based on the traditional sparse representation jointly particle filter framework. First, we obtained the observation image set from particle filter. Furthermore, we introduced a 2D transformation on the observation image set, which enables the tracking target candidates set more robust to handle misalignment problem in complex scene. Moreover, we adopt the occlusion detection mechanism before template updating, reducing the drift problem effectively. Experimental evaluations on five public challenging sequences, which exhibit occlusions, illuminating variations, scale changes, motion blur, and our tracker demonstrate accuracy and robustness in comparisons with the state-of-the-arts.

Multi-task joint sparse representation (MTJSR) is one kind of efficient multi-task learning (MTL) method for solving different problems together using a shared sparse representation. Based on the learning mechanism in human, which is a self-paced learning by gradually training the tasks from easy to difficult, I apply this mechanism into MTJSR, and propose a multi-task joint sparse representation with self-paced learning (MTJSR-SP) algorithm. In MTJSR-SP, the self-paced learning mechanism is considered as a regularizer of optimization function, and an iterative optimization is applied to solve it. Comparing with the traditional MTL methods, MTJSR-SP has more robustness to the noise and outliers. The experimental results on some datasets, i.e. two synthesized datasets, four datasets from UCI machine learning repository, an oxford flower dataset and a Caltech-256 image categorization dataset, are used to validate the efficiency of MTJSR-SP.