Showing papers in "IEEE Signal Processing Letters in 2005"
TL;DR: Improvements to the nonlocal means image denoising method introduced by Buades et al. are presented and filters that eliminate unrelated neighborhoods from the weighted average are introduced.
Abstract: In this letter, improvements to the nonlocal means image denoising method introduced by Buades et al. are presented. The original nonlocal means method replaces a noisy pixel by the weighted average of pixels with related surrounding neighborhoods. While producing state-of-the-art denoising results, this method is computationally impractical. In order to accelerate the algorithm, we introduce filters that eliminate unrelated neighborhoods from the weighted average. These filters are based on local average gray values and gradients, preclassifying neighborhoods and thereby reducing the original quadratic complexity to a linear one and reducing the influence of less-related areas in the denoising of a given pixel. We present the underlying framework and experimental results for gray level and color images as well as for video.
562 citations
TL;DR: Two novel ways of applying the histogram characteristic function (HCF), introduced by Harmsen for the detection of steganography in color images but ineffective on grayscale images, are introduced: calibrating the output using a downsampled image and computing the adjacency histogram instead of the usual histogram.
Abstract: We consider the problem of detecting spatial domain least significant bit (LSB) matching steganography in grayscale images, which has proved much harder than for its counterpart, LSB replacement. We use the histogram characteristic function (HCF), introduced by Harmsen for the detection of steganography in color images but ineffective on grayscale images. Two novel ways of applying the HCF are introduced: calibrating the output using a downsampled image and computing the adjacency histogram instead of the usual histogram. Extensive experimental results show that the new detectors are reliable, vastly more so than those previously known.
544 citations
TL;DR: This letter describes a simple but powerful algorithm that searches the exponentially large space of partitions of N data points in time O(N/sup 2/), which is guaranteed to find the exact global optimum.
Abstract: Many signal processing problems can be solved by maximizing the fitness of a segmented model over all possible partitions of the data interval. This letter describes a simple but powerful algorithm that searches the exponentially large space of partitions of N data points in time O(N/sup 2/). The algorithm is guaranteed to find the exact global optimum, automatically determines the model order (the number of segments), has a convenient real-time mode, can be extended to higher dimensional data spaces, and solves a surprising variety of problems in signal detection and characterization, density estimation, cluster analysis, and classification.
377 citations
TL;DR: In this letter, a precise relationship between RDWT-domain and original-signal-domain distortion for additive white noise in the RDWT domain is derived.
Abstract: The behavior under additive noise of the redundant discrete wavelet transform (RDWT), which is a frame expansion that is essentially an undecimated discrete wavelet transform, is studied. Known prior results in the form of inequalities bound distortion energy in the original signal domain from additive noise in frame-expansion coefficients. In this letter, a precise relationship between RDWT-domain and original-signal-domain distortion for additive white noise in the RDWT domain is derived.
331 citations
TL;DR: A new method for bidimensional empirical mode decomposition (EMD) based on Delaunay triangulation and on piecewise cubic polynomial interpolation is described, which shows its efficiency in terms of computational cost and the decomposition of Gaussian white noises leads to bidimensional selective filter banks.
Abstract: In this letter, we describe a new method for bidimensional empirical mode decomposition (EMD). This decomposition is based on Delaunay triangulation and on piecewise cubic polynomial interpolation. Particular attention is devoted to boundary conditions that are crucial for the feasibility of the bidimensional EMD. The study of the behavior of the decomposition on a different kind of image shows its efficiency in terms of computational cost, and the decomposition of Gaussian white noises leads to bidimensional selective filter banks.
269 citations
TL;DR: This letter analyzes the conventional clipping and filtering using a parabolic approximation of the clipping pulse to get a new clip and filtering technique that obtains the same PAR reduction as that of the existing iterative techniques with 2K+1 FFT/IFFT operations, where K represents the number of iterations.
Abstract: The existing iterative clipping and filtering techniques require several iterations to mitigate the peak regrowth. In this letter, we analyze the conventional clipping and filtering using a parabolic approximation of the clipping pulse. We show that the clipping noise obtained after several clipping and filtering iterations is approximately proportional to that generated in the first iteration. Therefore, we scale the clipping noise generated in the first iteration to get a new clipping and filtering technique that, with three fast Fourier transform/inverse fast Fourier transform (FFT/IFFT) operations, obtains the same PAR reduction as that of the existing iterative techniques with 2K+1 FFT/IFFT operations, where K represents the number of iterations.
240 citations
TL;DR: Two new demosaicking algorithms based on the frequency-domain representation of color images sampled with the Bayer color filter array are described and shown to give excellent results.
Abstract: This letter presents a new and simplified derivation of the frequency-domain representation of color images sampled with the Bayer color filter array. Two new demosaicking algorithms based on the frequency-domain representation are described and shown to give excellent results.
215 citations
TL;DR: Results from simulations and experiments with real video data show the improved performance of the proposed algorithm when compared with that of the standard particle filter.
Abstract: A new particle filter-the Kernel Particle Filter (KPF)-is proposed for visual tracking in image sequences. The KPF invokes kernels to form a continuous estimate of the posterior density function. Particles are allocated based on the gradient information estimated from the kernel density estimate of the posterior. Results from simulations and experiments with real video data show the improved performance of the proposed algorithm when compared with that of the standard particle filter. The superior performance is evident in scenarios of small system noise or weak dynamic models where the standard particle filter usually fails.
206 citations
TL;DR: A new spectral search-based direction-of-arrival (DOA) estimation method is proposed that extends the idea of the conventional ESPRIT DOA estimator to a much more general class of array geometries than assumed by the conventional EspRIT technique.
Abstract: A new spectral search-based direction-of-arrival (DOA) estimation method is proposed that extends the idea of the conventional ESPRIT DOA estimator to a much more general class of array geometries than assumed by the conventional ESPRIT technique. A computationally efficient polynomial rooting-based search-free implementation of the proposed algorithm is also developed.
204 citations
TL;DR: The authors introduce a new selected mapping (SLM) orthogonal frequency division multiplexing (OFDM) scheme with low computational complexity, while it shows almost the same performance of PAPR reduction as that of the conventional SLM OFDM scheme.
Abstract: The authors introduce a new selected mapping (SLM) orthogonal frequency division multiplexing (OFDM) scheme with low computational complexity. The proposed SLM scheme transforms an input symbol sequence into a set of OFDM signals by multiplying the phase sequences to the signal after a certain intermediate stage of inverse fast Fourier transform (IFFT). Then, the OFDM signal with the lowest peak-to-average power ratio (PAPR) is selected for transmission. The new SLM OFDM scheme reduces the computational complexity, while it shows almost the same performance of PAPR reduction as that of the conventional SLM OFDM scheme.
201 citations
TL;DR: The coefficient adaptation process of the steepest descent algorithm is analyzed and how to calculate the optimal proportionate step size is derived in order to achieve the fastest convergence.
Abstract: A proportionate normalized least mean square (PNLMS) algorithm has been proposed for sparse impulse response identification. It provides fast initial convergence, but it begins to slow down dramatically after the initial period. In this letter, we analyze the coefficient adaptation process of the steepest descent algorithm and derive how to calculate the optimal proportionate step size in order to achieve the fastest convergence. The results bring forward a novel view of the concept of proportion. We propose a /spl mu/-law PNLMS (MPNLMS) algorithm using an approximation of the optimal proportionate step size. Line segment approximation and partial update techniques are discussed to bring down the computational complexity.
TL;DR: This letter presents a new voice activity detector (VAD) for improving speech detection robustness in noisy environments and the performance of speech recognition systems using an optimum likelihood ratio test (LRT) involving multiple and independent observations.
Abstract: Currently, there are technology barriers inhibiting speech processing systems that work in extremely noisy conditions from meeting the demands of modern applications. This letter presents a new voice activity detector (VAD) for improving speech detection robustness in noisy environments and the performance of speech recognition systems. The algorithm defines an optimum likelihood ratio test (LRT) involving multiple and independent observations. The so-defined decision rule reports significant improvements in speech/nonspeech discrimination accuracy over existing VAD methods that are defined on a single observation and need empirically tuned hangover mechanisms. The algorithm has an inherent delay that, for several applications, including robust speech recognition, does not represent a serious implementation obstacle. An analysis of the overlap between the distributions of the decision variable shows the improved robustness of the proposed approach by means of a clear reduction of the classification error as the number of observations is increased. The proposed strategy is also compared to different VAD methods, including the G.729, AMR, and AFE standards, as well as recently reported algorithms showing a sustained advantage in speech/nonspeech detection accuracy and speech recognition performance.
TL;DR: This letter considers the robust finite-horizon filtering problem for a class of discrete time-varying systems with missing measurements and norm-bounded parameter uncertainties, and shows that the desired filter can be obtained in terms of the solutions to two discrete Riccati difference equations.
Abstract: In this letter, we consider the robust finite-horizon filtering problem for a class of discrete time-varying systems with missing measurements and norm-bounded parameter uncertainties. The missing measurements are described by a binary switching sequence satisfying a conditional probability distribution. An upper bound for the state estimation error variance is first derived for all possible missing observations and all admissible parameter uncertainties. Then, a robust filter is designed, guaranteeing that the variance of the state estimation error is not more than the prescribed upper bound. It is shown that the desired filter can be obtained in terms of the solutions to two discrete Riccati difference equations, which are of a form suitable for recursive computation in online applications. A simulation example is presented to show the effectiveness of the proposed approach by comparing to the traditional Kalman filtering method.
TL;DR: The performance of the proposed approach is shown to significantly outperform existing methods based on superimposed training (ST) in frequency-selective channel estimation and symbol detection.
Abstract: We address the problem of frequency-selective channel estimation and symbol detection using superimposed training. The superimposed training consists of the sum of a known sequence and a data-dependent sequence that is unknown to the receiver. The data-dependent sequence cancels the effects of the unknown data on channel estimation. The performance of the proposed approach is shown to significantly outperform existing methods based on superimposed training (ST).
TL;DR: The interpolated minimum mean squared error (MMSE) solution is described and the normalized least mean squares (NLMS) and affine-projection (AP) algorithms for both the filter and the interpolator are proposed.
Abstract: In this letter, we propose a broadly applicable reduced-rank filtering approach with adaptive interpolated finite impulse response (FIR) filters in which the interpolator is rendered adaptive. We describe the interpolated minimum mean squared error (MMSE) solution and propose normalized least mean squares (NLMS) and affine-projection (AP) algorithms for both the filter and the interpolator. The resulting filtering structures are considered for equalization and echo cancellation applications. Simulation results showing significant improvements are presented for different scenarios.
TL;DR: Simulations show that with proper design, cooperative transmission can enhance energy efficiency and prolong sensor network lifetime.
Abstract: The efficiency of space-time block code-encoded (STBC) cooperative transmission is studied within low-energy adaptive clustering hierarchy (LEACH), which is a typical networking/communication protocol for wireless sensor networks. Cooperation protocol with low overhead is proposed, and synchronization requirements among cooperating sensors are discussed. Energy efficiency is analyzed as a tradeoff between the reduced transmission energy consumption and the increased electronic and overhead energy consumption. Simulations show that with proper design, cooperative transmission can enhance energy efficiency and prolong sensor network lifetime.
TL;DR: This letter proposes a novel steganographic scheme that employs human vision sensitivity to hide a large amount of secret bits into a still image with a high imperceptibility by using a series of symbols in a notation system with multiple bases.
Abstract: This letter proposes a novel steganographic scheme that employs human vision sensitivity to hide a large amount of secret bits into a still image with a high imperceptibility. In this method, data to be embedded are converted into a series of symbols in a notation system with multiple bases. The specific bases used are determined by the degree of local variation of the pixel magnitudes in the host image so that pixels in busy areas can potentially carry more hidden data. Experimental results are given to show the advantage of this adaptive technique.
TL;DR: A bivariate maximum a posteriori estimator is designed, which relies on the family of isotropic α-stable distributions, to better capture the heavy-tailed nature of the data as well as the interscale dependencies of wavelet coefficients.
Abstract: Recently, the dual-tree complex wavelet transform has been proposed as an analysis tool featuring near shift-invariance and improved directional selectivity compared to the standard wavelet transform. Within this framework, we describe a novel technique for removing noise from digital images. We design a bivariate maximum a posteriori estimator, which relies on the family of isotropic α-stable distributions. Using this relatively new statistical model we are able to better capture the heavy-tailed nature of the data as well as the interscale dependencies of wavelet coefficients. We test our algorithm for the Cauchy case, in comparison with several recently published methods. The simulation results show that our proposed technique achieves state-of-the-art performance in terms of root mean squared (RMS) error.
TL;DR: The present letter proposes an alternate procedure that can be effectively employed to replace the essentially algorithmic sifting process in Huang's empirical mode decomposition (EMD) method by a parabolic partial differential equation (PDE)-based approach.
Abstract: The present letter proposes an alternate procedure that can be effectively employed to replace the essentially algorithmic sifting process in Huang's empirical mode decomposition (EMD) method. Recent works have demonstrated that EMD acts essentially as a dyadic filter bank that can be compared to wavelet decompositions. However, the origin of EMD is algorithmic in nature and, hence, lacks a solid theoretical framework. The present letter proposes to resolve the major problem in the EMD method-the mean envelope detection of a signal-by a parabolic partial differential equation (PDE)-based approach. The proposed approach is validated by employing several numerical studies where the PDE-based sifting process is applied to some synthetic composite signals.
TL;DR: A computationally efficient approximation of the maximum likelihood detector for 16 quadrature amplitude modulation in multiple-input multiple-output (MIMO) systems based on a convex relaxation of the ML problem is developed.
Abstract: We develop a computationally efficient approximation of the maximum likelihood (ML) detector for 16 quadrature amplitude modulation (16-QAM) in multiple-input multiple-output (MIMO) systems. The detector is based on a convex relaxation of the ML problem. The resulting optimization is a semidefinite program that can be solved in polynomial time with respect to the number of inputs in the system. Simulation results in a random MIMO system show that the proposed algorithm outperforms the conventional decorrelator detector by about 2.5 dB at high signal-to-noise ratios.
TL;DR: This work uses the second-order blind identification (SOBI) algorithm to separate the EEG into statistically independent sources and SVMs to identify the artifact components and thereby to remove such signals.
Abstract: Artifacts such as eye blinks and heart rhythm (ECG) cause the main interfering signals within electroencephalogram (EEG) measurements. Therefore, we propose a method for artifact removal based on exploitation of certain carefully chosen statistical features of independent components extracted from the EEGs, by fusing support vector machines (SVMs) and blind source separation (BSS). We use the second-order blind identification (SOBI) algorithm to separate the EEG into statistically independent sources and SVMs to identify the artifact components and thereby to remove such signals. The remaining independent components are remixed to reproduce the artifact-free EEGs. Objective and subjective assessment of the simulation results shows that the algorithm is successful in mitigating the interference within EEGs.
TL;DR: The proposed modified LS with sparse channel-estimation algorithm has a 5-dB lower mean square error in channel estimation when compared to the conventional approach, which translates to approximately 0.5 dB improvement in signal-to-noise ratio at the receiver.
Abstract: We describe an algorithm for sparse channel estimation applicable to orthogonal frequency division multiplexing systems. The proposed algorithm uses a least squares (LS) technique for channel estimation and a generalized Akaike information criterion to estimate the channel length and tap positions. This effectively reduces the signal space of the LS estimator, and hence improves the estimation performance as demonstrated using computer simulations. For example, the proposed modified LS with sparse channel-estimation algorithm has a 5-dB lower mean square error in channel estimation when compared to the conventional approach , which translates to approximately 0.5 dB improvement in signal-to-noise ratio at the receiver.
TL;DR: This letter provides an exact BER formulation and its simple bounds for a case of optimum-combining cooperation system with differential M-ary phase-shift keying (DMPSK) signals and shows that the proposed differential cooperative transmission scheme together with the optimum power allocation yields comparable performance to the optimum- combines scheme.
Abstract: In this letter, we propose a differential amplify-and-forward (AF) transmission scheme for a two-user cooperative communications system. By efficiently combining signals from both direct and relay links, the proposed scheme provides superior performance compared to those of direct transmissions with either differential detection or coherent detection. While the exact bit-error-rate (BER) formulation of the proposed scheme is not available currently, we provide, as a performance benchmark, an exact BER formulation and its simple bounds for a case of optimum-combining cooperation system with differential M-ary phase-shift keying (DMPSK) signals. The optimum power allocation is also determined based on the provided BER formulations. We show that the proposed differential cooperative transmission scheme together with the optimum power allocation yields comparable performance to the optimum-combining scheme. Simulation results show that the proposed differential scheme with optimum power allocation yields significant performance improvement over that with an equal power allocation scheme.
TL;DR: A simple heuristic to estimate the performance of the linear predictor from a pixel spatial context and a context modeling mechanism with one-band look-ahead capability that improves the overall compression with marginal usage of additional memory are introduced.
Abstract: We present a new low-complexity algorithm for hyperspectral image compression that uses linear prediction in the spectral domain. We introduce a simple heuristic to estimate the performance of the linear predictor from a pixel spatial context and a context modeling mechanism with one-band look-ahead capability, which improves the overall compression with marginal usage of additional memory. The proposed method is suitable to spacecraft on-board implementation, where limited hardware and low power consumption are key requirements. Finally, we present a least-squares optimized linear prediction technique that achieves better compression on data cubes acquired by the NASA JPL Airborne Visible/Infrared Imaging Spectrometer (AVIRIS).
TL;DR: The spherical harmonic framework is employed to compare the well-known delay-and-sum to the phase-mode processing for spherical arrays to show similar performance at frequencies where the upper spherical harmonic order equals the product of the wave number and sphere radius.
Abstract: Phase-mode spherical microphone array processing, also known as spherical harmonic array processing, has been recently studied for various applications. The spherical array configuration provides desired three-dimensional symmetry, while the phase modes provide frequency-independent spatial processing. This letter employs the spherical harmonic framework to compare the well-known delay-and-sum to the phase-mode processing for spherical arrays. The two approaches show similar performance at frequencies where the upper spherical harmonic order equals the product of the wave number and sphere radius. However, at lower frequencies, phase-mode processing maintains the same directivity, limited by signal-to-noise ratio, while for delay-and-sum, spatial resolution deteriorates.
TL;DR: Under the condition of n+/spl kappa/=const, the basic difference between the unscented Kalman filtering for the nonlinear dynamic systems with additive process and measurement noises is that the augmented UKF draws a sigma set only once within a filtering recursion, while the nonaugmented UKF has to redraw a new set of sigma points to incorporate the effect of additive process noise.
Abstract: This paper concerns the unscented Kalman filtering (UKF) for the nonlinear dynamic systems with additive process and measurement noises. It is widely accepted for such a case that the system state needs not to be augmented with noise vectors and the resultant nonaugmented UKF yields similar, if not the same, results to the augmented UKF. In this letter, we find that under the condition of n+/spl kappa/=const, the basic difference between them is that the augmented UKF draws a sigma set only once within a filtering recursion, while the nonaugmented UKF has to redraw a new set of sigma points to incorporate the effect of additive process noise. This difference generally favors the augmented UKF in that the odd-order moment information is partly captured by the nonlinearly transformed sigma points and propagated throughout the recursion. The simulation results agree well with the analyses.
TL;DR: G/spl Gamma/D can model the distribution of the real speech signal more accurately than the conventional Gaussian, Laplacian, Gamma, or generalized Gaussian distribution (GGD).
Abstract: In this letter, we propose a new statistical model, two-sided generalized gamma distribution (G/spl Gamma/D) for an efficient parametric characterization of speech spectra. G/spl Gamma/D forms a generalized class of parametric distributions, including the Gaussian, Laplacian, and Gamma probability density functions (pdfs) as special cases. We also propose a computationally inexpensive online maximum likelihood (ML) parameter estimation algorithm for G/spl Gamma/D. Likelihoods, coefficients of variation (CVs), and Kolmogorov-Smirnov (KS) tests show that G/spl Gamma/D can model the distribution of the real speech signal more accurately than the conventional Gaussian, Laplacian, Gamma, or generalized Gaussian distribution (GGD).
TL;DR: A general form of the MVDR where any unitary matrix can be used to estimate the spectrum and it is shown that this algorithm gives much more reliable results than the one based on the popular Welch's method.
Abstract: The minimum variance distortionless response (MVDR) approach is very popular in array processing. It is also employed in spectral estimation where the Fourier matrix is used in the optimization process. First, we give a general form of the MVDR where any unitary matrix can be used to estimate the spectrum. Second and most importantly, we show how the MVDR method can be used to estimate the magnitude squared coherence function, which is very useful in so many applications but so few methods exist to estimate it. Simulations show that our algorithm gives much more reliable results than the one based on the popular Welch's method.
TL;DR: The robustness and discriminability of the AM-FM features is investigated in combination with mel cepstrum coefficients (MFCCs), and it is shown that these hybrid features perform well in the presence of noise, both in terms of phoneme-discrimination and speech recognition performance.
Abstract: In this letter, a nonlinear AM-FM speech model is used to extract robust features for speech recognition. The proposed features measure the amount of amplitude and frequency modulation that exists in speech resonances and attempt to model aspects of the speech acoustic information that the commonly used linear source-filter model fails to capture. The robustness and discriminability of the AM-FM features is investigated in combination with mel cepstrum coefficients (MFCCs). It is shown that these hybrid features perform well in the presence of noise, both in terms of phoneme-discrimination (J-measure) and in terms of speech recognition performance in several different tasks. Average relative error rate reduction up to 11% for clean and 46% for mismatched noisy conditions is achieved when AM-FM features are combined with MFCCs.
TL;DR: This letter proposes a multilevel HQAM arrangement with adaptive constellation distances that provides a graceful degradation in the quality of the decoded video without requiring feedback from the receiver.
Abstract: In this letter, hierarchical quadrature amplitude modulation (HQAM) is used to provide unequal error protection (UEP) for layered (data partitioned) H.264 coded video. In a conventional HQAM system, the high-priority (HP) and low-priority (LP) capacities have a constant ratio, whereas in H.264 data partitioning, the corresponding parts do not necessarily have this constant ratio. This letter proposes a multilevel HQAM arrangement with adaptive constellation distances that provides a graceful degradation in the quality of the decoded video without requiring feedback from the receiver. The arrangement improves the quality relative to nonhierarchical transmission through poor signal-to-noise ratio (SNR) channels at the price of a modest quality reduction through good SNR channels.