Driss Aboutajdine

A two-stage gene selection scheme utilizing MRMR filter and GA wrapper

Gene expression data usually contain a large number of genes, but a small number of samples. Feature selection for gene expression data aims at finding a set of genes that best discriminates biological samples of different types. In this paper, we propose a two-stage selection algorithm for genomic data by combining MRMR (Minimum Redundancy–Maximum Relevance) and GA (Genetic Algorithm). In the first stage, MRMR is used to filter noisy and redundant genes in high-dimensional microarray data. In the second stage, the GA uses the classifier accuracy as a fitness function to select the highly discriminating genes. The proposed method is tested for tumor classification on five open datasets: NCI, Lymphoma, Lung, Leukemia and Colon using Support Vector Machine (SVM) and Naïve Bayes (NB) classifiers. The comparison of the MRMR-GA with MRMR filter and GA wrapper shows that our method is able to find the smallest gene subset that gives the most classification accuracy in leave-one-out cross-validation (LOOCV).

Developed Distributed Energy-Efficient Clustering (DDEEC) for heterogeneous wireless sensor networks

Typically, a wireless sensor network contains an important number of inexpensive power constrained sensors, which collect data from the environment and transmit them towards the base station in a cooperative way. Saving energy and therefore, extending the wireless sensor networks lifetime, imposes a great challenge. Clustering techniques are largely used for these purposes. In this paper, we propose and evaluate a clustering technique called a Developed Distributed Energy-Efficient Clustering scheme for heterogeneous wireless sensor networks. This technique is based on changing dynamically and with more efficiency the cluster head election probability. Simulation results show that our protocol performs better than the Stable Election Protocol (SEP) by about 30% and than the Distributed Energy-Efficient Clustering (DEEC) by about 15% in terms of network lifetime and first node dies.

Fast adaptive algorithms for AR parameters estimation using higher order statistics

Time-varying statistics in linear filtering and linear estimation problems necessitate the use of adaptive or time-varying filters in the solution. With the rapid availability of vast and inexpensive computation power, models which are non-Gaussian even nonstationary are being investigated at increasing intensity. Statistical tools used in such investigations usually involve higher order statistics (HOS). The classical instrumental variable (IV) principle has been widely used to develop adaptive algorithms for the estimation of ARMA processes. Despite, the great number of IV methods developed in the literature, the cumulant-based procedures for pure autoregressive (AR) processes are almost nonexistent, except lattice versions of IV algorithms. This paper deals with the derivation and the properties of fast transversal algorithms. Hence, by establishing a relationship between classical (IV) methods and cumulant-based AR estimation problems, new fast adaptive algorithms, (fast transversal recursive instrumental variable-FTRIV) and (generalized least mean squares-GLMS), are proposed for the estimation of AR processes. The algorithms are seen to have better performance in terms of convergence speed and misadjustment even in low SNR. The extra computational complexity is negligible. The performance of the algorithms, as well as some illustrative tracking comparisons with the existing adaptive ones in the literature, are verified via simulations. The conditions of convergence are investigated for the GLMS.

Boosting 3-D-Geometric Features for Efficient Face Recognition and Gender Classification

We utilize ideas from two growing but disparate ideas in computer vision-shape analysis using tools from differential geometry and feature selection using machine learning-to select and highlight salient geometrical facial features that contribute most in 3-D face recognition and gender classification. First, a large set of geometries curve features are extracted using level sets (circular curves) and streamlines (radial curves) of the Euclidean distance functions of the facial surface; together they approximate facial surfaces with arbitrarily high accuracy. Then, we use the well-known Adaboost algorithm for feature selection from this large set and derive a composite classifier that achieves high performance with a minimal set of features. This greatly reduced set, consisting of some level curves on the nose and some radial curves in the forehead and cheeks regions, provides a very compact signature of a 3-D face and a fast classification algorithm for face recognition and gender selection. It is also efficient in terms of data storage and transmission costs. Experimental results, carried out using the FRGCv2 dataset, yield a rank-1 face recognition rate of 98% and a gender classification rate of 86% rate.

Hypergraph imaging: an overview

Hypergraph theory as originally developed by Berge (Hypergraphe, Dunod, Paris, 1987) is a theory of finite combinatorial sets, modeling lot of problems of operational research and combinatorial optimization. This framework turns out to be very interesting for many other applications, in particular for computer vision. In this paper, we are going to survey the relationship between combinatorial sets and image processing. More precisely, we propose an overview of different applications from image hypergraph models to image analysis. It mainly focuses on the combinatorial representation of an image and shows the effectiveness of this approach to low level image processing; in particular to segmentation, edge detection and noise cancellation.

Source separation contrasts using a reference signal

In this letter, we consider contrast functions, which are useful tools for the sources separation problem. We present new generalized contrasts by considering a so-called reference signal. In particular, this allows us to show that a criterion recently proposed in the literature is a contrast. Furthermore, a link with a joint-diagonalization criterion is also emphasized. Finally, we show that another classical contrast can also be extended by considering a reference signal.

Textural feature selection by joint mutual information based on Gaussian mixture model for multispectral image classification

Textural features play increasingly an important role in remotely sensed images classification and many pattern recognition applications. However, the selection of informative ones with highly discriminatory ability to improve the classification accuracy is still one of the well-known problems in remote sensing. In this paper, we propose a new method based on the Gaussian mixture model (GMM) in calculating Shannons mutual information between multiple features and the output class labels. We apply this, in a real context, to a textural feature selection algorithm for multispectral image classification so as to produce digital thematic maps for cartography exploitation. The input candidate features are extracted from an HRV-XS SPOT image of a forest area in Rabat, Morocco, using wavelet packet transform (WPT) and the gray level cooccurrence matrix (GLCM). The retained classifier is the support vectors machine (SVM). The results show that the selected textural features, using our proposed method, make the largest contribution to improve the classification accuracy than the selected ones by mutual information between individual variables. The use of spectral information only leads to poor performances.

Local appearance based face recognition method using block based steerable pyramid transform

In this paper, an efficient local appearance feature extraction method based on Steerable Pyramid (S-P) wavelet transform is proposed for face recognition. Local information is extracted by computing the statistics of each sub-block obtained by dividing S-P sub-bands. The obtained local features of each sub-band are combined at the feature and decision level to enhance face recognition performance. The purpose of this paper is to explore the usefulness of S-P as feature extraction method for face recognition. The proposed approach is compared with some related feature extraction methods such as principal component analysis (PCA), as well as linear discriminant analysis LDA and boosted LDA. Different multi-resolution transforms, wavelet (DWT), gabor, curvelet and contourlet, are also compared against the block-based S-P method. Experimental results on ORL, Yale, Essex and FERET face databases convince us that the proposed method provides a better representation of the class information, and obtains much higher recognition accuracies in real-world situations including changes in pose, expression and illumination.

HOS-based image sequence noise removal

In this paper, a new spatiotemporal filtering scheme is described for noise reduction in video sequences. For this purpose, the scheme processes each group of three consecutive sequence frames in two steps: 1) estimate motion between frames and 2) use motion vectors to get the final denoised current frame. A family of adaptive spatiotemporal L-filters is applied. A recursive implementation of these filters is used and compared with its nonrecursive counterpart. The motion trajectories are obtained recursively by a region-recursive estimation method. Both motion parameters and filter weights are computed by minimizing the kurtosis of error instead of mean squared error. Using the kurtosis in the algorithms adaptation is appropriate in the presence of mixed and impulsive noises. The filter performance is evaluated by considering different types of video sequences. Simulations show marked improvement in visual quality and SNRI measures cost as well as compared to those reported in literature.

Feature Selection on Supervised Classification Using Wilks Lambda Statistic

Variable and feature selection have become the focus of much research in areas of application for which datasets with tens or hundreds of thousands of variables are available. This paper addresses the feature selection problem for supervised classification. We operate this feature selection step by step, which leads to search for a criterion to quantify the most relevant variable and its contribution compared to the others already selected. In this article we present a feature selection method based on the Wilks lambda criterion which is a statistical one used in discriminant analysis. Our objective is to evaluate the performances of this method when used in another application different from its classical one i.e. the discriminant analysis. This criterion is compared to other very known algorithms in the field of the feature selection on various real data sets. The obtained results with the criterion of Wilks lambda are satisfactory and even better in some cases.