samir belaid | Fac. Sciences Monastir University (original) (raw)

Papers by samir belaid

Neurocomputing, 2015

ABSTRACT The aim of the proposed algorithm is to expedite three-dimensional objects indexing and ... more ABSTRACT The aim of the proposed algorithm is to expedite three-dimensional objects indexing and retrieval. Investigating the Bag-of-Features (BoF) paradigm, we focus on a set of extracted local descriptors from 3D objects. Using scalar function calculated on the surface mesh, the algorithm extracts the salient points, and then associates each of these points with a local Fourier descriptor. The descriptor is computed on the neighboring salient points by projecting the geometry onto the first eigenvectors of Laplace-Beltrami operator. Additionally, through an offline learning step, a visual dictionary is built by clustering a large set of feature descriptors. Then, each 3D shape is described by a histogram of these visual words occurrences weighted using the number of their local descriptors. Experimental results show the highly discriminative capability of the proposed approach against rigid and nonrigid transformations, noise and geometry changes. The performance of our algorithm is also demonstrated on global and partial shape retrieval.

Research efforts in digital watermarking have been mainly on audio, image and video data. Only re... more Research efforts in digital watermarking have been mainly on audio, image and video data. Only recently, 3D watermarking has attracted the attention, due to the increasing proliferation of the 3D object scans and models, and the new trend of 3D imaging and communication witnessed in many areas, such as medicine, architecture and entertainment. In this paper, we present a compilation and description of the current state of the art in 3D triangular mesh watermarking. After a description of the different categories, paradigms, and methods develop so far, we provide a comparative study highlighting their merits and the shortcomings with respect to a set of standard criteria.

3ème rencontre francophone d’holographie numérique appliquée à la métrologie des fluides (Holophi 3), Ecully France, Nov 13, 2014

Nous proposons une méthode permettant la suppression d’images jumelles lors de la reconstruction... more Nous proposons une méthode permettant la suppression d’images jumelles lors de la
reconstruction des hologrammes numériques enregistrés dans la configuration in-line. Cette
méthode est basée sur la séparation aveugle de mélanges convolutifs de sources. Le procédé
est composé de deux étapes: un schéma de lifting en quinconce basé sur la transformée en
ondelettes, dont le rôle est de permettre une décorrélation des images holographiques d'entrée,
et un algorithme de déconvolution géométrique pour la tâche de séparation. Les résultats
expérimentaux confirment l’efficacité de la séparation aveugle de sources pour l’élimination
de l’image jumelle à partir d'hologrammes numériques de particules.

6th International Conference of Soft Computing and Pattern Recognition (SoCPaR2014)

This paper interests to present a new efficient method for indexing and retrieval of three-dimens... more This paper interests to present a new efficient method for indexing and retrieval of three-dimensional objects in general databases. The proposed method is based on Bag of Features (BoF) paradigm, and salient characteristics of objects. We first extract the salient points of target object using a scalar function calculated on the surface. Then, each interest point is associated with local Fourier descriptor. This descriptor is computed in its neighborhood by projecting the geometry onto the first eigenvectors of Laplace-Beltrami operator. Additionally, in an offline learning stage, a visual dictionary is built by clustering a large set of feature descriptors. Then each 3D shape is described by a histogram of occurrences of these visual words weighted by his local descriptors number. The obtained results show the highly discriminative capability of the proposed approach against rigid and non-rigid transformations, noise and geometry changes. The performance of our approach is also demonstrated for global and partial shape retrieval.

IET Signal Processing, Apr 10, 2015

ABSTRACT This paper describes a new multilevel decomposition method for the separation of convolu... more ABSTRACT This paper describes a new multilevel decomposition method for the separation of convolutive image mixtures. The proposed method uses an Adaptive Quincunx Lifting Scheme (AQLS) based on wavelet decomposition to preprocess the input data, followed by a Non-Negative Matrix Factorization whose role is to unmix the decomposed images. The unmixed images are, thereafter, reconstructed using the inverse of AQLS transform.
Experiments carried out on images from various origins showed that the proposed method yields better results than many widely used blind source separation algorithms.

We propose a robust autofocus method for reconstructing digital holograms and twin-image removal ... more We propose a robust autofocus method for reconstructing digital holograms and twin-image removal based on blind source separation approach. The method is made up of two components: an efficient quincunx lifting scheme based on wavelet packet transform, whose role is to maximize a sharpness metric related to the sparseness of the input holograms, and a geometric unmixing algorithm, which achieves the separation task. Experimental results confirm the ability of sparse blind source separation to discard the unwanted twin-image from in-line digital holograms of particles.

Electronics, Circuits, and Systems (ICECS), 2013 IEEE 20th International Conference on

This paper describes a new multi-resolution approach for the blind source separation of convoluti... more This paper describes a new multi-resolution approach for the blind source separation of convolutive image mixtures in transform domain. The proposed method uses an Adaptive Quincunx Lifting Scheme based on wavelet decomposition and a Complex ICA unmixing algorithm. It proceeds in three steps: first, the mixed signals are decomposed by an adaptive lifting scheme. Then, the unmixing algorithm is applied to the more relevant component. The unmixed signals are, thereafter, reconstructed using an inverse transform. Experiments carried out on images from various origins showed that the proposed method yields better results than many widely used blind source separation algorithms.

Signal Processing: Algorithms, Architectures, Arrangements, and Applications (SPA), 2013, Sep 28, 2013

This paper describes a new multilevel-decomposition method for the blind source separation of con... more This paper describes a new multilevel-decomposition method for the blind source separation of convolutive image mixture. The proposed method uses an Adaptive Quincunx Lifting Scheme based on wavelet decomposition and a geometric unmixing algorithm. It proceeds in three steps. In the first step, the mixed signals are decomposed by an adaptive lifting scheme. The unmixing algorithm therefore applied to the most relevant component. The unmixed signals are, thereafter, reconstructed using an inverse transform. Experiments carried out on images from various origins showed that the proposed method yields better results than many widely used blind source separation algorithms.

ROIS’13-Hammam-Sousse TUNISIA

This work presents a new approach for face rep-resentation. Our purpose is to implement a fast an... more This work presents a new approach for face rep-resentation. Our purpose is to implement a fast and efficient 3D face recognition system, which is insensitive to the captur-ing conditions and no need to normalization phase.
Our approach shows that the HalfEdge collapse techniques, used in 3D object compression, can be considered as a suita-ble form for multi-resolution analysis.
The experiments performed on real data illustrate the possi-bility to represent the human face shape into different levels of resolution. Inspired from far and near natural human perception, we process the comparison stage throw three-analysis level of model face. Then we describe the face repre-sentation using novel metrics to compare the 3D face mesh model and those of the base. Experimental results using the FRAV3D database was presented showing a remarkable outperformance of our approach against studied works.

Modeling, Simulation and Applied Optimization (ICMSAO), 2013 5th International Conference on, Apr 28, 2013

The detection and recognition of human face are a vital step to achieve several objectives, such ... more The detection and recognition of human face are a vital step to achieve several objectives, such as intrusion detection, face recognition which require high accuracy. The presence or absence of a beard changes a person's appearance dramatically and has a huge effect or the ability to detect and to recognize him. This paper introduces a novel approach for the detection of bearded face and removal of beard from three dimensional
bearded faces. The proposed method preprocess to extract the informative part of face ...

N. Werghi, S. Selim, S. BELAID, Mar 2004

N. Werghi, Y. Xiao, S. BELAID, Mar 2002

S. BELAID, N. WERGHI, Mar 2001

High-speed in-line holography is used to visualize the trajectories fibers which are drawn out in... more High-speed in-line holography is used to visualize the trajectories fibers which are drawn out in a turbulent flame. In order to improve the SNR, the images are not observed by a conventional reconstruction setup. In our case, the holographic plate is directly placed on the input plane of a wavelet-transform optical system. This processing system is based on a VanderLugt correlator including an Electrically Addressed Spatial Light Modulator (EASLM). The shape of matched filters is deduced by successive rotation and dilatation operations of wavelet functions in the Fourier domain. The 3-D location of a fiber element and its orientation are estimated by searching the daughter wavelet which yields the maximum intensity on the output plane of the correlator, which also contains the reconstructed image. The results are compared with those obtained by a conventional optical reconstruction. The SNR of the images observed on the output plane is improved. Moreover, it is shown that the axis coordinate accuracy is improved to (Delta) z equals plus or minus 50 micrometer, instead of plus or minus 0,5 mm for holographic reconstruction.

S.BELAID, D.LEBRUN ET C.ÖZKUL, Aug 1996

$Research paper thumbnail of Localisation 3D de fibres de verre en étirage par analyse des figures de diffraction a l'aide de la transformée en ondelettes$

S.BELAID, D.LEBRUN ET C.ÖZKUL, Oct 1996

S. ADELINE, J.B. BLAISOT, S. BELAID, M. LEDOUX, Aug 1997

D. LEBRUN, S. BELAID, C. OZKUL, Jun 1998

T.GIRASOLE, K.F.REN, D.LEBRUN, S.BELAID, G.GOUESBET, G.GREHAN , Dec 1998

C. BURAGA, S. COËTMELLEC, D. LEBRUN, S. BELAID, C. OZKUL, May 2000