Mohamed Najim

An efficient algorithm for computation of shape moments from run-length codes or chain codes

Abstract Moments are very useful for shape analysis. Zero- to third-order moments have been used for computer vision applications such as shape recognition and orientation. They can serve for composition of the well-known moment invariants used as desirable features as well as for the detection of the location and the principal axis direction of a shape. The shape is often represented by a binary image and its moments can be obtained by use of fast algorithms considering the shape as a discrete point array. In this paper a new algorithm based on the double-integral formulation is presented. The shape is considered as a continuous region and the contribution of boundary points is used for fast computation of shape moments. This method can be used to calculate moments from either the run-length codes or the chain codes of shape.

Adaptive nonlinear filters for 2D and 3D image enhancement

Abstract Unsharp masking method is a popular approach for image enhancement, in which a highpass version of an image is added to the original one. This method is easy to run, but is very sensitive to noise. Suppressing noise is generally performed with lowpass filters, and leads to edge blurring. So, an approach which is a combination of a nonlinear lowpass and highpass filters is proposed. These filters are based on an adaptive filter mask. We demonstrate that this approach performs noise reduction as well as edge enhancement. It also improves the contrast enhancement in comparison with other methods. These results are illustrated by processing blurred and noisy images. The method is then extended for 3D data processing and used on 3D seismic images.

Image segmentation by a dynamic thresholding using edge detection based on cascaded uniform filters

Abstract A new method for gray level thresholding is presented in this paper. First, the edge pixels can be localized by the difference between two cascaded uniform filters with different scales. Then, considering the gray level value of an edge pixel as the ideal threshold for its neighborhood, a dynamic threshold function will be constructed by a dilation process from these primitive thresholds. The experimental results we provide in this paper show that this technique is very effective in segmenting images taken under non-uniform illuminations.

Audio digital watermarking based on hybrid spread spectrum

Watermarking is a technique used to label digital media, to protect copyright ownership, by hiding information into the signal. Watermarks must be imperceptible and robust against attacks. Besides, watermark robustness against several signal processing techniques can be ensured by exploiting a technique related to spread spectrum communications. In this paper, we propose a procedure based on frequential substitution using a BPSK modulation with an adaptive carrier frequency, the choice of which depends on the original signal features. A secret key, generated during the watermarking process, is used for retrieving hidden information.

2-D high resolution spectral estimation based on multiple regions of support

This paper deals with frequency estimation in the 2-D case when one has only few data points. We propose a method to estimate the frequencies of a sum of exponentials. This method is based on an original set of 2-D linear prediction models with new regions of support derived from the standard quarter plane support region. These models define various spectra which are finally combined by computing their harmonic mean. This method benefits from the subspace decomposition of the covariance matrix to perform well. It is demonstrated that the new regions of support improve the spectrum geometry and the estimation accuracy compared to the classical quarter plane (QP) support regions.

Estimation of autoregressive fading channels based on two cross-coupled H∞ filters

This paper deals with the on-line estimation of time-varying frequency-flat Rayleigh fading channels based on training sequences and using H∞ filtering. When the fading channel is approximated by an autoregressive (AR) process, the AR model parameters must be estimated. As their direct estimations from the available noisy observations at the receiver may yield biased values, the joint estimation of both the channel and its AR parameters must be addressed. Among the existing solutions to this joint estimation issue, Expectation Maximization (EM) algorithm or cross-coupled filter based approaches can be considered. They usually require Kalman filtering which is optimal in the H2 sense provided that the initial state, the driving process and measurement noise are independent, white and Gaussian. However, in real cases, these assumptions may not be satisfied. In addition, the state-space matrices and the noise variances are not necessarily accurately estimated. To take into account the above problem, we propose to use two cross-coupled H∞ filters. This method makes it possible to provide robust estimation of the fading channel and its AR parameters.

Joint wavelet transform and vector quantization for speech coding

A new scheme for speech compression is presented using both discrete wavelet transform (DWT) and vector quantization (VQ). This method named discrete wavelet vector transform quantization (DWVTQ) is a coding system in which the speech signal is transformed into wavelet coefficients corresponding to different frequency bands which are vector quantized separately. The proposed implementation uses product code structure for each frequency band. The bit assignment of vector codes is adapted taking into account both the statistics of the wavelet coefficients and the fact that the human ear is less sensitive in the high-frequency bands. Vector quantization design procedures for discrete wavelet transformed speech and simulation results are presented. These results show the efficiency of DWVTQ over discrete wavelet scalar transform quantization (DWSTQ) of the transformed speech, and improvement of the optimal bit allocation in comparison to uniform bit allocation.<<ETX>>

Texture characterization based on 2-D reflection coefficients

In the framework of model based image processing, we propose a new parametric approach for classifying textured images. The image, considered as a two-dimensional stochastic process, is characterized by a set of reflection coefficients computed using a two-dimensional adaptive lattice filter based on the recursive least squares (RLS) criterion. The corresponding algorithm is named the two-dimensional fast lattice RLS. In order to evaluate this method, classification rates are calculated on a set of 8 different textures from the Brodatz album. We carry out performance comparisons with methods of characterization based on two-dimensional AR coefficients computed with two-dimensional transversal filters or based on statistical features calculated from co-occurrence matrices and neighbouring matrices.

Noisy speech dereverberation as a SIMO system identification issue

This paper deals with the speech dereverberation issue based on a single input multiple output (SIMO) system, when the reverberations are modeled by finite impulse response (FIR) filters. In most of the existing methods, the authors assume either that the white noises have the same variance or that the noise statistics are available. Here, we investigate the blind speech deconvolution using two microphones, when the white noise variances are not equal. For this purpose, we present a modified version of an identification approach previously developed in the framework of control and based on the properties of the definiteness and the positiveness of the autocorrelation matrices of the reverberated versions of the speech and the observations. This makes it possible to estimate both the variances of the additive noises and the FIR. Then, the speech signal is retrieved in the least square (LS) or minimum variance (MV) sense

EOG segmentation using fast algorithms

A new method is presented for segmentation of rapidly time-varying signals based on fast least squares algorithms, due to their low complexity. The detection test used is based on the likelihood variable which appears explicitly in the fast algorithms. In order to show the performances of the method, it is applied to E.O.G. (electrooculograms) signals.<<ETX>>