Institute Superieur dInformatique

Saint-Étienne-de-Baïgorry, France

Institute Superieur dInformatique

Saint-Étienne-de-Baïgorry, France

Time filter

Source Type

Wuhrer S.,National Research Council Canada | Azouz B.Z.,Institute Superieur dInformatique | Shu C.,National Research Council Canada
Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition | Year: 2010

We propose a posture invariant surface descriptor for triangular meshes. Using intrinsic geometry, the surface is first transformed into a representation that is independent of the posture. Spin image is then adapted to derive a descriptor for the representation. The descriptor is used for extracting surface features automatically. It is invariant with respect to rigid and isometric deformations, and robust to noise and changes in resolution. The result is demonstrated by using the automatically extracted features to find correspondences between articulated meshes. ©2010 IEEE.

Amri R.,RIADI Laboratory | Bellamine Ben Saoud N.,Institute superieur dinformatique
Proceedings - 2014 4th International Conference on Advances in Computing and Communications, ICACC 2014 | Year: 2014

Since many years sustainability is becoming a challenging issue in software engineering domain. However till now, no clear nor exhaustive characterization was proposed to the concept of "sustainable software". Without it, sustainability remains an intangible ideal for software systems and consequently can't be assessed nor controlled nor enhanced. To cover this gap, we propose a Generic Sustainable Software Star Model (GS3M) that forms the basis towards a "complete" view of sustainable software. The model covers different sustainability dimensions: environmental, technical, social, individual and economic. For each dimension we define corresponding software sustainability values. To each value are associated software attributes promoting it. And each attribute can be composed of sub-attributes. To this last is associated a well defined metric. © 2014 IEEE.

Lassoued I.,Institute Superieur dInformatique | Zagrouba E.,Institute Superieur dInformatique | Chahir Y.,University of Caen Lower Normandy
Communications in Computer and Information Science | Year: 2011

Action recognition in video and still image is one of the most challenging research topics in pattern recognition and computer vision. This paper proposes a new method for video action classification based on 3D Zernike moments. These last ones aim to capturing both structural and temporal information of a time varying sequence. The originality of this approach consists to represent actions in video sequences by a three-dimension shape obtained from different silhouettes in the space-time volume. In fact, the given video is segmented in space-time volume. Then, silhouettes are extracted from obtained images of the video sequences volumes and 3D Zernike moments are computed for video, based on silhouettes volumes. Finally, least square version of SVM (LSSVM) classifier with extracted features is used to classify actions in videos. To evaluate the proposed approach, it was applied on a benchmark human action dataset. The experimentations and evaluations show efficient results in terms of action characterizations and classification. Further more, it presents several advantages such as simplicity and respect of silhouette movement progress in the video guaranteed by 3D Zernike moment. © 2011 Springer-Verlag.

Jabra S.B.,Institute Superieur dInformatique | Zagrouba E.,Institute Superieur dInformatique
Proceedings - 5th International Conference on Signal Image Technology and Internet Based Systems, SITIS 2009 | Year: 2010

This paper proposes 3D mesh watermarking using Maximally Stable Meshes detection and multi-signatures embedding that is robust against numerous attacks, blind, invisible and permits to detect attack type applied on marked mesh. In the proposed algorithm, The Maximally Stable meshes (MSMs) are detected using an extension of Maximally Stable Efficient Regions detection (MSER). After, these MSMs are sorted by their vertices number and three MSMs are selected to be marked. Finally, three different signatures are embedded using three different watermarking schemes. This allows to know the type of the used attack by detecting which of the signatures resisted and to maximize robustness by profiting from advantages of every scheme. Experimental results verified the robustness of the proposed algorithm based on watermark extraction after various types of attack including Rotation, Translation, Zooming, remeshing, simplification, noise and smoothing. In plus, given an attacked mesh, the proposed algorithm can detect the applied attack type by knowing the resistant signature. © 2009 IEEE.

Zoubeir N.,Institute Superieur dInformatique | Khalfallah A.,Institute Superieur dInformatique
MODELSWARD 2015 - 3rd International Conference on Model-Driven Engineering and Software Development, Proceedings | Year: 2015

Software architecture's interoperability faces many problems when it comes to integrating different components or formalisms in describing the architecture. Even within the same modeling language such as UML, the diversity of notations and the lack of semantic information make the interoperability between models difficult. In this paper, we propose semantic foundations that unify the notations of classes, interactions and constraints, and hence provide a robust basis for models interoperability. We rely on graphs and graph transformations to describe systems structure and behavior, semantics and constraints in a combined form within an integrated framework, which constitutes a robust basis for automated software architecture analysis. Copyright © 2015 SCITEPRESS - Science and Technology Publications.

Wafa M.,Institute Superieur dInformatique | Zagrouba E.,Institute Superieur dInformatique
International Conference on Signals and Electronic Systems, ICSES'10 - Conference Proceeding | Year: 2010

In this article, an image segmentation method based on Dempster-Shafer evidence theory and fuzzy logic theory is presented. The neighborhood relationship between the voxels is exploited to provide a more accurate modeling of the information and to determine the class of a pattern to classify. Fuzzy logic theory is used to automatic estimate mass function using pixel membership degrees witch is estimated at pixel level. We illustrate the effectiveness of the method in segmentation of synthetic images. Copyright © 2010 by Institute of Electronics, Silesian University of Technology.

Zoubeir N.,Institute Superieur dInformatique | Khalfallah A.,Institute Superieur dInformatique | Benahmed S.,Faculte des science de Tunis
International Journal of Software Engineering and its Applications | Year: 2014

Design patterns recognition and injection constitute challenging tasks in software engineering, since they are generally conducted in a non-formal way. In this paper, we will present an approach for decomposing and formalizing design patterns using graph transformation systems. We will propose a combined graph-based description for design patterns structure, interactions and constraints. Then, based on this description, we will introduce a set of Elementary Transformations whose diverse combinations lead to the design patterns injection. These elementary transformations are formal, precise and presented in a generic form that allows to describe the 23 design patterns defined by the GOF. Detailed descriptions and examples are presented in this paper in order to illustrate our approach. © 2014 SERSC.

Chihi H.,Institute Superieur dInformatique | Arous N.,Institute Superieur dInformatique
International Journal of Digital Content Technology and its Applications | Year: 2011

The aim of this work is to design a hierarchical model which represents a multi-layer extension of Self-Organizing Map (SOM) variant. The purpose of the proposed system is to create autonomous systems that can learn independently and cooperate to provide a better decision of the phoneme classification. The basic SOM variant is a hybrid model of SOM and Genetic Algorithm (GA) using a growing incremental technique to adapt the map structure and extra information in the map units to optimise the map codebooks. The hierarchical evolutionary learning algorithm classifies data according to tow hierarchical levels ensuring the representation of the hierarchical relations of the data. Our experiments yielded a high recognition rate of 77.73% on TIMIT acoustic-phonetic continuous speech corpus.

Amri S.,Institute Superieur dInformatique | Barhoumi W.,Institute Superieur dInformatique | Zagrouba E.,Institute Superieur dInformatique
2010 IEEE International Conference on Imaging Systems and Techniques, IST 2010 - Proceedings | Year: 2010

We propose in this paper a novel iterative approach for unsupervised reconstruction of static background from a complex video shot. After aligning some key-frames of the video onto a reference plane in order to compensate the camera motion, the basic idea of the suggested scheme is to iteratively reconstruct a precise image of the background using median blending and spatial segmentation. In each iteration, coarse binary masks, representing foreground moving objects, are estimated by comparing each motion-compensated key-frame with the corresponding part in the input background image. These masks are then refined by spatial segmentation while profiting of the semantic information offered by region maps. The iterative process allows the blending operator to eliminate the detected moving objects while reconstructing the output background image. Several experiments have been carried out to prove the effectiveness of the suggested unsupervised approach for precise background reconstruction of complex dynamic scenes after a relatively small number of iterations. © 2010 IEEE.

Moualhi W.,Institute Superieur DInformatique | Ezzeddine Z.,Institute Superieur DInformatique
Proceedings of SPIE - The International Society for Optical Engineering | Year: 2015

The motivation of this work is to register a tumor brain magnetic resonance (MR) image with a normal brain atlas. A normal brain atlas is deformed in order to take account of the presence of a large space occupying tumor. The method use a priori model of tumor growth assuming that the tumor grows in a radial way from a starting point. First, an affine transformation is used in order to bring the patient image and the brain atlas in a global correspondence. Second, the seeding of a synthetic tumor into the brain atlas provides a template for the lesion. Finally, the seeded atlas is deformed combining a method derived from optical flow principles and a model for tumor growth (MTG). Results show that an automatic segmentation method of brain structures in the presence of large deformation can be provided. © 2015 SPIE.

Loading Institute Superieur dInformatique collaborators
Loading Institute Superieur dInformatique collaborators