séminaire Pr. Aouatif AMINE

Bonjour à tous, Vous êtes cordialement invités à participer au premier séminaire du LATECE de la saison 2016/2017. Qui? Professeur Aouatif AMINE, de l’université Ibn Tofail de Kenitra au Maroc. Quand? Mercredi 14 septembre 2016 à 12h45 (12h15 pour la pizza) Où? PK-7210 Titre: Object tracking using Cuckoo Search and Hypovigilance Analysis to Extract Driver’s Micro Sleep Periods Résumé: Object Tracking (MOT) has been an attractive research topic for the video processing community, given the amount of information that a user can get from it, and the number of possibilities that a successful tracking performance can trigger. Unfortunately, tracking many moving objects is still a complex task, regarding the computation cost that each object may require. Metaheuristics, which are a high level computation processes, have been introduced in order to deal with the problem of running time reduction. Many population-based metaheuristics have been successfully introduced in the field of image processing. The hypovigilance reduces driver’s capacity to react, judge and analyze information and it is often caused by fatigue and/or drowsiness. However, drowsiness and fatigue are different. Drowsiness refers to the inability to stay awake while fatigue refers to a cumulative process producing difficulty to pay attention. Therefore, it is very important to monitor driver’s vigilance level and issue an alarm when he is not paying enough attention to the road. Several visual behaviors can be used to characterize hypovigilante driver. In this research, we propose a system that uses two criteria. The first criterion detects drowsiness by extracting microsleeps using the presence of the iris. The second one detects fatigue using yawning deduced from large mouth opening. In this work, the proposed scheme begins by a face detection using local Successive Mean Quantization Transform (SMQT) features and split up Sparse Network of Winnows (SNoW) classifier. After performing face detection, the novel approach for eye/mouth detection, based on Circular Hough Transform (CHT), is applied on eyes and mouth extracted regions. Biographie: Pr. Aouatif AMINE est depuis 2014 professeur habilité à l’université Ibn Tofail de Kenitra, Maroc. En juillet 2010, elle a rejoint l’École Nationale des Sciences Appliquée de kenitra au Maroc. Ex-Vice Présidente de IEEE SPS Morocco Chapter, secrétaire générale de l’Association Marocaine pour le Développement de l’Électronique, l’Électrotechnique, l’informatique et de l’Automatique (AMADEIA). Les domaines de recherche sont la classification des données, le suivi d’objet, la sécurité routière pour les véhicules intelligents et reconnaissance de forme SLIDES

Comments are closed.