Showing papers in "Signal Processing in 2001"

TL;DR: This work estimates both the number of sources and the mixing matrix by the maxima of a potential function along the circle of unit length, and obtains the minimal l1 norm representation of each data point by a linear combination of the pair of basis vectors that enclose it.

TL;DR: An optimally-modi#ed log-spectral amplitude (OM-LSA) speech estimator and a minima controlled recursive averaging (MCRA) noise estimation approach for robust speech enhancement are presented.

TL;DR: A criterion that can provide a measure of time–frequency distribution concentration is proposed that does not need normalization in order to behave properly when cross-terms are present and does not discriminate low concentrated components with respect to the highly concentrated ones within the same distribution.

TL;DR: A stochastic formulation of watermarking attacks using an estimation-based concept and a new method of evaluating image quality based on the Watson metric which overcomes the limitations of the PSNR are proposed.

TL;DR: A new direction finding algorithm for non-circular sources based on the polynomial rooting technique that is able to handle more sources than sensors and reduces computation cost and enhances resolution power significantly.

TL;DR: The present bibliography represents a comprehensive list of references on nonlinear system identification and its applications in signal processing, communications, and biomedical engineering.

TL;DR: A novel digital watermarking technique based on neural networks for color images that hides an invisible watermark into a color image, and then cooperates neural networks to learn the characteristics of the embedded watermark related to the watermarked image.

TL;DR: A multiple change-point analysis for which an MCMC sampler plays a fundamental role, and it is shown that different interesting statistics can be derived from the posterior distribution.

TL;DR: Simulation results demonstrate that the proposed iterative receiver offers significant performance gain over traditional non-iterative receiver structures and permits its use in severe multipath channels (10 or more paths).

TL;DR: The role of information theory is reviewed in characterizing the fundamental limits of watermarking systems and in guiding the development of optimal watermark embedding algorithms and optimal attacks.

TL;DR: A general method allowing us to devise maximum-likelihood multiplicative algorithms for inverse problems, and particularly for signal and image restoration with non-negativity constraint, founded on the Kuhn–Tucker first-order optimality conditions is given.

TL;DR: An efficient square wave transform (SWT) is presented, based on applying the signum function operator to the conventional discrete cosine transform (DCT) and is termed the signed DCT (SDCT).

TL;DR: A method for blind separation of instantaneous mixture of colored sources, based on the minimization a Gaussian mutual information criterion, amounts to jointly approximately diagonalizing a set of estimated spectral density matrices.

TL;DR: The foundations upon which modern numerical Bayesian methods are based are set out, give one application to missing data in audio restoration and then give references to application areas that can be addressed.

TL;DR: This paper addresses the MCMC methods from the second group, which allow for generation of samples from probability distributions defined on unions of disjoint spaces of different dimensions and shows why sampling from such distributions is a nontrivial task.

TL;DR: It is shown that the newly proposed detector has the constant false alarm rate property with respect to the covariance matrix of the noise.

TL;DR: It is shown that the contrast functions, previously introduced in the case of one-input/one-output blind deconvolution and then in linear source sepertion, allow one to separate the sources by a deflation approach.

TL;DR: It is shown that the use of superimposed pulses does not produce any benefit when applying them to data hiding and the possibility of using codes at the ‘sample level’ (that is, without repeating every codeword symbol) is introduced and its potential analyzed for both hard- and soft-decision decoding.

TL;DR: The main idea is to structure the watermark embedding as a linear programming problem in which the author wishes to maximize the strength of the watermarks subject to a set of linear constraints on the pixel distortions as determined by a masking function.

TL;DR: A detector of transient acoustic signals that combines two powerful detection tools: a local wavelet analysis and higher-order statistical properties of the signals and shows that the detector performs better than decompositions using other classical splitting criteria.

TL;DR: A new procedure is developed that has better performance for large estimation errors, and when used to initialize the Weiss and Friedlander (1991) MUSIC-based iterative technique, is seen significant improvement over existing techniques for both small and large errors.

TL;DR: Experimental results show that GVQ and GSAVQ need a little longer CPU time than, the maximum decent (MD) algorithm, but they outperform MD by 0.2–0.5 dB in PSNR.

TL;DR: Simulation results show that this architecture may be a suitable candidate for low power/low voltage applications and in speed in terms of latency and throughput.

TL;DR: This work derives the optimum attack, which minimizes the channel capacity for a given attack distortion, and derives the rule-of-thumb for the defense, which is extended to the case of frequency-weighted MSE as a distortion measure.

TL;DR: In this article, uniform scalar quantization of watermarked documents is investigated theoretically, extending results from theory of dithered quantization, and experimentally, the watermark is embedded by an independent additive pseudo-noise sequence.

TL;DR: The probabilistic algorithms of noise reduction presented in the second part of this paper are able to eliminate strong noise, while preserving edges and image texture.

TL;DR: The achieved results have proven that the proposed system makes it possible to obtain reliable people counting in different environmental situations, with an absolute mean error at most equal to 10%.

TL;DR: A backward sequential elimination algorithm wherein, starting from the full dictionary, elements are deleted until a subset of a desired size is obtained, which is able to conclude that the p-norm close to 1 performs best while the system considered is overcomplete.

TL;DR: The M-band wavelet decomposition, which is a direct generalization of the standard 2-band waveshell decomposition is applied to the problem of an unsupervised segmentation of two texture images, and simple K-means clustering is obtained.

TL;DR: A new method to reduce the number of multipliers in the design of sharp FIR filters by frequency-response masking technique is presented, based on a modified frequency- response masking approach where one of the subfilters in the approach is implemented by interpolated finite-impulse response (IFIR) technique.