Showing papers in "IEEE Signal Processing Letters in 2013"
TL;DR: This work has recently derived a blind IQA model that only makes use of measurable deviations from statistical regularities observed in natural images, without training on human-rated distorted images, and, indeed, without any exposure to distorted images.
Abstract: An important aim of research on the blind image quality assessment (IQA) problem is to devise perceptual models that can predict the quality of distorted images with as little prior knowledge of the images or their distortions as possible. Current state-of-the-art “general purpose” no reference (NR) IQA algorithms require knowledge about anticipated distortions in the form of training examples and corresponding human opinion scores. However we have recently derived a blind IQA model that only makes use of measurable deviations from statistical regularities observed in natural images, without training on human-rated distorted images, and, indeed without any exposure to distorted images. Thus, it is “completely blind.” The new IQA model, which we call the Natural Image Quality Evaluator (NIQE) is based on the construction of a “quality aware” collection of statistical features based on a simple and successful space domain natural scene statistic (NSS) model. These features are derived from a corpus of natural, undistorted images. Experimental results show that the new index delivers performance comparable to top performing NR IQA models that require training on large databases of human opinions of distorted images. A software release is available at http://live.ece.utexas.edu/research/quality/niqe_release.zip.
3,722 citations
TL;DR: A very fast noniterative algorithm is proposed for denoising or smoothing one-dimensional discrete signals, by solving the total variation regularized least-squares problem or the related fused lasso problem.
Abstract: A very fast noniterative algorithm is proposed for denoising or smoothing one-dimensional discrete signals, by solving the total variation regularized least-squares problem or the related fused lasso problem. A C code implementation is available on the web page of the author.
302 citations
TL;DR: A novel bottom-up salient object detection approach by exploiting contrast, center and smoothness priors is presented and the smoothness prior enables the proposed method to uniformly highlight the salient object and simultaneously suppress the background effectively.
Abstract: Object level saliency detection is useful for many content-based computer vision tasks. In this letter, we present a novel bottom-up salient object detection approach by exploiting contrast, center and smoothness priors. First, we compute an initial saliency map using contrast and center priors. Unlike most existing center prior based methods, we apply the convex hull of interest points to estimate the center of the salient object rather than directly use the image center. This strategy makes the saliency result more robust to the location of objects. Second, we refine the initial saliency map through minimizing a continuous pairwise saliency energy function with graph regularization which encourages adjacent pixels or segments to take the similar saliency value (i.e., smoothness prior). The smoothness prior enables the proposed method to uniformly highlight the salient object and simultaneously suppress the background effectively. Extensive experiments on a large dataset demonstrate that the proposed method performs favorably against the state-of-the-art methods in terms of accuracy and efficiency.
249 citations
TL;DR: This letter proposes a joint cooperative beamforming and jamming scheme to enhance the security of a cooperative relay network, where a part of intermediate nodes adopt distributed beamforming while others jam the eavesdropper, simultaneously.
Abstract: Cooperative beamforming and jamming are two efficient schemes to improve the physical-layer security of a wireless relay system in the presence of passive eavesdroppers. However, in most works these two techniques are adopted separately. In this letter, we propose a joint cooperative beamforming and jamming scheme to enhance the security of a cooperative relay network, where a part of intermediate nodes adopt distributed beamforming while others jam the eavesdropper, simultaneously. Since the instantaneous channel state information (CSI) of the eavesdropper may not be known, we propose a cooperative artificial noise transmission based secrecy strategy, subjected to the individual power constraint of each node. The beamformer weights and power allocation can be obtained by solving a second-order convex cone programming (SOCP) together with a linear programming problem. Simulations show the joint scheme greatly improves the security.
193 citations
TL;DR: Experimental results indicate that the proposed SAD scheme is highly effective and provides superior and consistent performance across various noise types and distortion levels.
Abstract: Effective speech activity detection (SAD) is a necessary first step for robust speech applications. In this letter, we propose a robust and unsupervised SAD solution that leverages four different speech voicing measures combined with a perceptual spectral flux feature, for audio-based surveillance and monitoring applications. Effectiveness of the proposed technique is evaluated and compared against several commonly adopted unsupervised SAD methods under simulated and actual harsh acoustic conditions with varying distortion levels. Experimental results indicate that the proposed SAD scheme is highly effective and provides superior and consistent performance across various noise types and distortion levels.
186 citations
TL;DR: This letter studies cooperative secure beamforming for amplify-and-forward (AF) relay networks in the presence of multiple eavesdroppers and proves that this problem can be exactly solved by SDR with one SDP only.
Abstract: This letter studies cooperative secure beamforming for amplify-and-forward (AF) relay networks in the presence of multiple eavesdroppers. Under both total and individual relay power constraints, we propose two schemes, namely secrecy rate maximization (SRM) beamforming and null-space beamforming. In the first scheme, our design problem is based on SRM. Using a suboptimal, but convex, technique-semidefinite relaxation (SDR), we show that this problem can be handled by performing a one-dimensional search which involves solving a sequence of semidefinite programs (SDPs). To reduce the complexity, in the second scheme, we instead maximize the information rate at the destination while completely eliminating the information leakage to all eavesdroppers. We prove that this problem can be exactly solved by SDR with one SDP only. Simulation results demonstrate the performance gains of the two proposed designs.
180 citations
TL;DR: This letter proposes two improvements of the MOD and K-SVD dictionary learning algorithms, by modifying the two main parts of these algorithms-the dictionary update and the sparse coding stages, and suggests to leverage the known representations from the previous sparse-coding in the quest for the updated representations.
Abstract: In this letter, we propose two improvements of the MOD and K-SVD dictionary learning algorithms, by modifying the two main parts of these algorithms-the dictionary update and the sparse coding stages. Our first contribution is a different dictionary-update stage that aims at finding both the dictionary and the representations while keeping the supports intact. The second contribution suggests to leverage the known representations from the previous sparse-coding in the quest for the updated representations. We demonstrate these two ideas in practice and show how they lead to faster training and better quality outcome.
134 citations
TL;DR: This letter proposes an edge-strength-similarity-based image quality metric (ESSIM) that can achieve slightly better performance than the state-of-the-art image quality metrics as evaluated on six subject-rated image databases.
Abstract: The objective image quality assessment aims to model the perceptual fidelity of semantic information between two images. In this letter, we assume that the semantic information of images is fully represented by edge-strength of each pixel and propose an edge-strength-similarity-based image quality metric (ESSIM). Through investigating the characteristics of the edge in images, we define the edge-strength to take both anisotropic regularity and irregularity of the edge into account. The proposed ESSIM is considerably simple, however, it can achieve slightly better performance than the state-of-the-art image quality metrics as evaluated on six subject-rated image databases.
128 citations
TL;DR: In this article, a cooperative network with multiple source-destination pairs and one EH relay is considered and the outage probability experienced by users in this network is characterized by taking the spatial randomness of user locations into consideration.
Abstract: This letter considers a cooperative network with multiple source-destination pairs and one energy harvesting relay. The outage probability experienced by users in this network is characterized by taking the spatial randomness of user locations into consideration. In addition, the cooperation among users is modeled as a canonical coalitional game and the grand coalition is shown to be stable in the addressed scenario. Simulation results are provided to demonstrate the accuracy of the developed analytical results.
127 citations
TL;DR: The bias and mean square error (MSE) analysis of the frequency estimator suggested in is given and an improved version of the estimator, with the removal of estimator bias, is suggested.
Abstract: The bias and mean square error (MSE) analysis of the frequency estimator suggested in is given and an improved version of the estimator, with the removal of estimator bias, is suggested. The signal-to-noise ratio (SNR) threshold above which the bias removal is effective is also determined.
125 citations
TL;DR: It is shown that the eigen-analysis of the precision matrix of the Gaussian Markov random field model is optimal in decorrelating the signal, and an optimal scheme to perform predictive transform coding based on conditional probabilities of a GMRF model is presented.
Abstract: In this letter, we provide a theoretical analysis of optimal predictive transform coding based on the Gaussian Markov random field (GMRF) model. It is shown that the eigen-analysis of the precision matrix of the GMRF model is optimal in decorrelating the signal. The resulting graph transform degenerates to the well-known 2-D discrete cosine transform (DCT) for a particular 2-D first order GMRF, although it is not a unique optimal solution. Furthermore, we present an optimal scheme to perform predictive transform coding based on conditional probabilities of a GMRF model. Such an analysis can be applied to both motion prediction and intra-frame predictive coding, and may lead to improvements in coding efficiency in the future.
TL;DR: The proposed approach can handle arbitrary peak-to-average-power ratio constraints on the transmit sequence, and can be used for large dimension designs (with ~ 103 variables) even on an ordinary PC.
Abstract: Due to its long-standing importance, the problem of designing the receive filter and transmit sequence for clutter/interference rejection in active sensing has been studied widely in the last decades. In this letter, we propose a cyclic optimization of the transmit sequence and the receive filter. The proposed approach can handle arbitrary peak-to-average-power ratio (PAR) constraints on the transmit sequence, and can be used for large dimension designs (with ~ 103 variables) even on an ordinary PC.
TL;DR: This work derives two conjugate gradient algorithms for the computation of the filter coefficients and shows improved audio source separation performance compared to the classical Wiener filter both in oracle and in blind conditions.
Abstract: Wiener filtering is one of the most ubiquitous tools in signal processing, in particular for signal denoising and source separation. In the context of audio, it is typically applied in the time-frequency domain by means of the short-time Fourier transform (STFT). Such processing does generally not take into account the relationship between STFT coefficients in different time-frequency bins due to the redundancy of the STFT, which we refer to as consistency. We propose to enforce this relationship in the design of the Wiener filter, either as a hard constraint or as a soft penalty. We derive two conjugate gradient algorithms for the computation of the filter coefficients and show improved audio source separation performance compared to the classical Wiener filter both in oracle and in blind conditions.
TL;DR: This letter investigates multiple-input multiple-output (MIMO) communications under energy harvesting (EH) constraints and focuses on the beamforming designs with partial CSI, taking minimum mean-square-error (MMSE) and mutual information as performance metrics.
Abstract: In this letter, we investigate multiple-input multiple-output (MIMO) communications under energy harvesting (EH) constraints. In our considered EH system, there is one information transmitting (ITx) node, one traditional information receiving (IRx) node and multiple EH nodes. EH nodes can transform the received electromagnetic waves into energy to enlarge the network operation life. When the ITx node sends signals to the destination, it should also optimize the beamforming/precoder matrix to charge the EH nodes efficiently simultaneously. Additionally, the charged energy should be larger than a predefined threshold. Under the EH constraints, in our work both minimum mean-square-error (MMSE) and mutual information are taken as the performance metrics for the beamforming designs at the ITx node. In order to make the proposed algorithms suitable for practical implementation and have affordable overhead, our work focuses on the beamforming designs with partial CSI and this is the distinct contribution of our work. Finally, numerical results are given to show the performance advantages of the proposed algorithms.
TL;DR: Experimental results demonstrate the superiority of the proposed descriptor as compared to other methods considered in this letter, and suggest a novel approach to rotation and scale invariant texture classification.
Abstract: This letter introduces a novel approach to rotation and scale invariant texture classification. The proposed approach is based on Gabor filters that have the capability to collapse the filter responses according to the scale and orientation of the textures. These characteristics are exploited to first calculate the homogeneous texture of images followed by the rearrangement of features as a two-dimensional matrix (scale and orientation), where scaling and rotation of images correspond to shifting in this matrix. The shift invariance property of discrete fourier transform is used to propose rotation and scale invariant image features. The performance of the proposed feature set is evaluated on Brodatz texture album. Experimental results demonstrate the superiority of the proposed descriptor as compared to other methods considered in this letter.
TL;DR: This letter investigates dictionary training algorithms for sparse representation like K-SVD, MOD, and their variation from that viewpoint and proposes an alternative to MOD; a sequential generalization of K-means (SGK), which shows MOD and SGK to be faster under a dimensionality condition.
Abstract: Recent dictionary training algorithms for sparse representation like K-SVD, MOD, and their variation are reminiscent of K -means clustering, and this letter investigates such algorithms from that viewpoint. It shows: though K -SVD is sequential like K -means, it fails to simplify to K-means by destroying the structure in the sparse coefficients. In contrast, MOD can be viewed as a parallel generalization of K-means, which simplifies to K -means without perturbing the sparse coefficients. Keeping memory usage in mind, we propose an alternative to MOD; a sequential generalization of K-means (SGK). While experiments suggest a comparable training performances across the algorithms, complexity analysis shows MOD and SGK to be faster under a dimensionality condition.
TL;DR: It is shown that under some conditions on RIP and the minimum magnitude of the nonzero elements of the sparse signal, OMP with proper stopping rules can recover the support of the signal exactly from the noisy observation.
Abstract: Orthogonal matching pursuit (OMP) algorithm is a classical greedy algorithm in Compressed Sensing. In this letter, we study the performance of OMP in recovering the support of a sparse signal from a few noisy linear measurements. We consider two types of bounded noise and our analysis is in the framework of restricted isometry property (RIP). It is shown that under some conditions on RIP and the minimum magnitude of the nonzero elements of the sparse signal, OMP with proper stopping rules can recover the support of the signal exactly from the noisy observation. We also discuss the case of Gaussian noise. Our conditions on RIP improve some existing results.
TL;DR: A comparison with state-of-the-art metrics shows that the proposed Perceptual Sharpness Index correlates highly with human perception and exhibits low computational complexity.
Abstract: In this letter, a no-reference perceptual sharpness metric based on a statistical analysis of local edge gradients is presented. The method takes properties of the human visual system into account. Based on perceptual properties, a relationship between the extracted statistical features and the metric score is established to form a Perceptual Sharpness Index (PSI). A comparison with state-of-the-art metrics shows that the proposed method correlates highly with human perception and exhibits low computational complexity. In contrast to existing metrics, the PSI performs well for a wide range of blurriness and shows a high degree of invariance for different image contents.
TL;DR: Markov based features are adopted to detect double compression artifacts, which imply that the original video may have been interpolated, and demonstrate that the scheme outperforms most existing methods.
Abstract: With the spread of powerful and easy-to-use video editing software, digital videos are exposed to various forms of tampering Nowadays, a considerable proportion of surveillance systems and video cameras have built-in MPEG-4 codec Therefore, the detection of double compression in MPEG-4 videos as a first step in video forensics research is of significance In this paper, Markov based features are adopted to detect double compression artifacts, which imply that the original video may have been interpolated The advantages and limitations of double MPEG-4 compression detection are analyzed Experimental results have demonstrated that our scheme outperforms most existing methods
TL;DR: Simulation results on recovery of a known dictionary and dictionary learning for natural image patches show that the new problem considerably improves performance with a little additional computational load.
Abstract: A dictionary learning problem is a matrix factorization in which the goal is to factorize a training data matrix, Y, as the product of a dictionary, D, and a sparse coefficient matrix, X, as follows, Y ≃ DX. Current dictionary learning algorithms minimize the representation error subject to a constraint on D (usually having unit column-norms) and sparseness of X. The resulting problem is not convex with respect to the pair (D,X). In this letter, we derive a first order series expansion formula for the factorization, DX. The resulting objective function is jointly convex with respect to D and X. We simply solve the resulting problem using alternating minimization and apply some of the previously suggested algorithms onto our new problem. Simulation results on recovery of a known dictionary and dictionary learning for natural image patches show that our new problem considerably improves performance with a little additional computational load.
TL;DR: A novel sparsity prior for compressive imaging in the context of the theory of compressed sensing with coherent redundant dictionaries is discussed, based on the observation that natural images exhibit strong average sparsity over multiple coherent frames.
Abstract: We discuss a novel sparsity prior for compressive imaging in the context of the theory of compressed sensing with coherent redundant dictionaries, based on the observation that natural images exhibit strong average sparsity over multiple coherent frames. We test our prior and the associated algorithm, based on an analysis reweighted $\ell _{1}$ formulation, through extensive numerical simulations on natural images for spread spectrum and random Gaussian acquisition schemes. Our results show that average sparsity outperforms state-of-the-art priors that promote sparsity in a single orthonormal basis or redundant frame, or that promote gradient sparsity. Code and test data are available at https://github.com/basp-group/sopt .
TL;DR: This letter presents a new mixed-order MUSIC algorithm for far-field and near-field sources localization using a sparse symmetric array that has moderate computation complexity, and provides higher resolution, and also improves the parameters estimation accuracy.
Abstract: This letter presents a new mixed-order MUSIC algorithm for far-field and near-field sources localization using a sparse symmetric array. By exploiting the special array geometry, the proposed algorithm constructs a cumulant matrix to estimate the directions of arrival (DOAs) of both far-field and near-field sources using the conventional MUSIC method. With the estimated DOAs and the covariance matrix of the sparse array, the far-field and near-field sources are identified and the range parameters of near-field sources are also obtained by defining the range spectrum. Compared with the traditional algorithms, the proposed algorithm has moderate computation complexity, and provides higher resolution, and also improves the parameters estimation accuracy. Simulation results are provided to demonstrate the performance improvement of the proposed method.
TL;DR: The L-estimate transforms and time-frequency representations are presented within the framework of compressive sensing to recover signal or local auto-correlation function samples corrupted by impulse noise.
Abstract: The L-estimate transforms and time-frequency representations are presented within the framework of compressive sensing. The goal is to recover signal or local auto-correlation function samples corrupted by impulse noise. The signal is assumed to be sparse in a transform domain or in a joint-variable representation. Unlike the standard L-statistics approach, which suffers from degraded spectral characteristics due to the omission of samples, the compressive sensing in combination with the L-estimate permits signal reconstruction that closely approximates the noise free signal representation.
TL;DR: Experiments show large intelligibility improvements with the proposed method over the unprocessed noisy speech and better performance than one state-of-the art method.
Abstract: In this letter the focus is on linear filtering of speech before degradation due to additive background noise. The goal is to design the filter such that the speech intelligibility index (SII) is maximized when the speech is played back in a known noisy environment. Moreover, a power constraint is taken into account to prevent uncomfortable playback levels and deal with loudspeaker constraints. Previous methods use linear approximations of the SII in order to find a closed-form solution. However, as we show, these linear approximations introduce errors in low SNR regions and are therefore suboptimal. In this work we propose a nonlinear approximation of the SII which is accurate for all SNRs. Experiments show large intelligibility improvements with the proposed method over the unprocessed noisy speech and better performance than one state-of-the art method.
TL;DR: The experimental results showed that compared to existing CPW solutions, the LJSCPW is more robust and effective under various noise levels and attains higher means with smaller variances in terms of the peak signal and noise ratio (PSNR) and structural similarity (SSIM).
Abstract: Non-Local Means (NLM) and its variants have proven to be effective and robust in many image denoising tasks. In this letter, we study approaches to selecting center pixel weights (CPW) in NLM. Our key contributions are 1) we give a novel formulation of the CPW problem from a statistical shrinkage perspective; 2) we construct the James-Stein shrinkage estimator in the CPW context; and 3) we propose a new local James-Stein type CPW (LJSCPW) that is locally tuned for each image pixel. Our experimental results showed that compared to existing CPW solutions, the LJSCPW is more robust and effective under various noise levels. In particular, the NLM with the LJSCPW attains higher means with smaller variances in terms of the peak signal and noise ratio (PSNR) and structural similarity (SSIM), implying it improves the NLM denoising performance and makes the denoising less sensitive to parameter changes.
TL;DR: A minimum mean squared error (MMSE) optimal estimator for clean speech spectral amplitudes is derived, which is applied in single channel speech enhancement and it is shown that the phase contains additional information that can be exploited to distinguish outliers in the noise from the target signal.
Abstract: In this letter, we derive a minimum mean squared error (MMSE) optimal estimator for clean speech spectral amplitudes, which we apply in single channel speech enhancement. As opposed to state-of-the-art estimators, the optimal estimator is derived for a given clean speech spectral phase. We show that the phase contains additional information that can be exploited to distinguish outliers in the noise from the target signal. With the proposed technique, incorporating the phase can potentially improve the PESQ-MOS by 0.5 in babble noise as compared to state-of-the-art amplitude estimators. In a blind setup we achieve a PESQ improvement of around 0.25 in voiced speech.
TL;DR: This work proposes a novel framework to extract compact and discriminative features from Electrocardiogram (ECG) signals for human identification based on sparse representation of local segments that achieves an 99.48% accuracy on a 100 subjects dataset constructed from a publicly available database.
Abstract: This work proposes a novel framework to extract compact and discriminative features from Electrocardiogram (ECG) signals for human identification based on sparse representation of local segments Specifically, local segments extracted from an ECG signal are projected to a small number of basic elements in a dictionary, which is learned from training data A final representation is extracted by performing a max pooling procedure over all the sparse coefficient vectors in the ECG signal Unlike most of existing methods for human identification from ECG signals which require segmentation of individual heartbeats or extraction of fiducial points, the proposed method does not need to segment individual heartbeats or detect any fiducial points The method achieves an 9948% accuracy on a 100 subjects dataset constructed from a publicly available database, which demonstrates that both local and global structural information are well captured to characterize the ECG signals
TL;DR: This paper proposes a linear discriminant regression classification algorithm (LDRC) that performs better than the related regression based algorithms and shows a promising ability for face recognition.
Abstract: To improve the robustness of the linear regression classification (LRC) algorithm, in this paper, we propose a linear discriminant regression classification (LDRC) algorithm to boost the effectiveness of the LRC for face recognition We embed the Fisher criterion into the LRC as a novel discriminant regression analysis method The LDRC attempts to maximize the ratio of the between-class reconstruction error (BCRE) over the within-class reconstruction error (WCRE) to find an optimal projection matrix for the LRC such that the LRC on that subspace can achieve a high discrimination for classification Then, the projected coefficients are executed by the LRC for face recognition Extensive experiments carried out on the FERET and AR face databases show that the LDRC performs better than the related regression based algorithms and shows a promising ability for face recognition
TL;DR: This paper proposes a closed-loop single-channel speech enhancement approach to estimate both amplitude and phase spectra of the speech signal by combining a group-delay based phase estimator with a phase-aware amplitude estimator in a closed loop design.
Abstract: Many short-time Fourier transform (STFT) based single-channel speech enhancement algorithms are focused on estimating the clean speech spectral amplitude from the noisy observed signal in order to suppress the additive noise. To this end, they utilize the noisy amplitude information and the corresponding a priori and a posteriori SNRs while they employ the observed noisy phase when reconstructing enhanced speech signal. This paper presents two contributions: i) reconsidering the relation between the phase group delay deviation and phase deviation, and ii) proposing a closed-loop single-channel speech enhancement approach to estimate both amplitude and phase spectra of the speech signal. To this end, we combine a group-delay based phase estimator with a phase-aware amplitude estimator in a closed loop design. Our experimental results on various noise scenarios show considerable improvement in the objective perceived signal quality obtained by the proposed iterative phase-aware approach compared to conventional Wiener filtering which uses the noisy phase in signal reconstruction.
TL;DR: This work proves analytically that the Chernoff distance amounts to calculate an equivalent but simpler Bregman divergence defined on the distribution parameters, and proposes three novel information-theoretic symmetric distances and middle distributions, from which two of them admit always closed-form expressions.
Abstract: The Chernoff information was originally introduced for bounding the probability of error of the Bayesian decision rule in binary hypothesis testing. Nowadays, it is often used as a notion of symmetric distance in statistical signal processing or as a way to define a middle distribution in information fusion. Computing the Chernoff information requires to solve an optimization problem that is numerically approximated in practice. We consider the Chernoff distance for distributions belonging to the same exponential family including the Gaussian and multinomial families. By considering the geometry of the underlying statistical manifold, we define exactly the solution of the optimization problem as the unique intersection of a geodesic with a dual hyperplane. Furthermore, we prove analytically that the Chernoff distance amounts to calculate an equivalent but simpler Bregman divergence defined on the distribution parameters. It follows a closed-form formula for the singly-parametric distributions, or an efficient geodesic bisection search for multiparametric distributions. Finally, based on this information-geometric characterization, we propose three novel information-theoretic symmetric distances and middle distributions, from which two of them admit always closed-form expressions.