Dspace de universite Djillali Liabes de SBA >
Thèse de Doctorat en Sciences >
Informatique >
Veuillez utiliser cette adresse pour citer ce document :
http://hdl.handle.net/123456789/1318
|
Titre: | A highly efficient algorithm for fast motion detection and estimation using parallel processing |
Auteur(s): | ELBAHRI, Mohamed Encadreur: TALEB, Nasreddine |
Mots-clés: | Détection de mouvement représentation parcimonieuse architecture parallèle cartes graphiques |
Date de publication: | 30-nov-2015 |
Résumé: | Visual surveillance of dynamic scenes is one of the difficult current research topics in computer vision. It is a key technology in the fight against terrorism, crime, to ensure public safety and to the effective management of road traffic. This work is to design an effective system of detection and motion estimation for object tracking in a complex environment. In video surveillance, detection of moving objects from a video is important for objects classification, target tracking, recognition of the activity, and understanding behavior. In this thesis we used background subtraction for motion detection. In order to recognize objects, we used a classification system based on sparse representation. A faster processing is proposed in this work, using the performance of graphics cards. This acceleration consists to parallelize the "Othogonal Matching Pursuit" algorithm to speed up the processing and make it operate in real-time. |
Description: | Doctorat en sciences |
URI/URL: | http://hdl.handle.net/123456789/1318 |
Collection(s) : | Informatique
|
Fichier(s) constituant ce document :
|
Tous les documents dans DSpace sont protégés par copyright, avec tous droits réservés.
|