DSpace
 

Depot Institutionnel de l'UMBB >
Publications Scientifiques >
Publications Internationales >

Veuillez utiliser cette adresse pour citer ce document : http://dlibrary.univ-boumerdes.dz:8080/handle/123456789/6108

Titre: A new technique based on 3D convolutional neural networks and filtering optical flow maps for action classification in infrared video
Auteur(s): Khebli, A.
Meglouli, H.
Bentabet, L.
Airouche, M.
Mots-clés: Artificial neural networks
Image classification
Infrared imaging
Machine learning
Date de publication: 2019
Editeur: Control Engineering and Applied Informatics Journal
Collection/Numéro: Control Engineering and Applied InformaticsVolume 21, Issue 4, 2019;pp. 43-50
Résumé: Human action in video sequences provides three-dimensional spatio-temporal signals that characterize both visual appearance and motion dynamics. The aim of this work is to recognize human action in infrared video by focusing mainly on dynamic information. We developed a new technique based on deep 3D convolutional neural networks (3D CNNs) that take optical flow maps as input. Our approach consists mainly of three parts: 1) computation of optical flow maps; 2) filtering of these maps, using an entropy measurement in order to increase the classification rate and reduce the run time by eliminating sequences that do not contain human action; and 3) classification using 3D CNN. The experimental results obtained by our approach on the InfAR dataset show considerable improvement in comparison with results obtained by existing models.
URI/URL: https://www.scopus.com/record/display.uri?eid=2-s2.0-85081741782&origin=SingleRecordEmailAlert&dgcid=raven_sc_affil_en_us_email&txGid=48ab6c0c24a36042f762c1e57eefb7b6
http://dlibrary.univ-boumerdes.dz:8080/handle/123456789/6108
ISSN: Control Engineering and Applied Informatics Journal
Collection(s) :Publications Internationales

Fichier(s) constituant ce document :

Fichier Description TailleFormat
Khebli, A.pdf471,81 kBAdobe PDFVoir/Ouvrir
View Statistics

Tous les documents dans DSpace sont protégés par copyright, avec tous droits réservés.

 

Valid XHTML 1.0! Ce site utilise l'application DSpace, Version 1.4.1 - Commentaires