An efficient multi-resolution SVM network approach for object detection in aerial images

Abstract : In this paper, we deal with the problem of object detection in aerial images. A lot of efficient approaches uses a cascade of classifiers which process vectors of descriptive features such as HOG. In order to take into account the variability in object dimension, features at different resolutions are often concatenated in a large descriptor vector. This prevents from taking into account explicitly the different resolutions but results in losing some valuable information. To overcome this problem, we propose to use a new method based on a SVM network. Each resolution is processed , regardless to the others, at the input layer level, by a dedicated SVM. The main drawback of using such a network is that the computational complexity for the classification phase drastically increases. We propose then to foster an incomplete exploration of the network by defining an activation path. This activation path determines an order to activate the network neurons, one after the other, and introduces a rejection rule which allows the process to end before crossing the whole network. Experimental results are obtained and assessed in an industrial application of urban object detection. We can observe an average gain of 17% in precision while the computational cost is divided by more than 5, with respect to a standard method.
Type de document :
Communication dans un congrès
MLSP: Machine Learning for Signal Processing, Sep 2015, Boston, United States. Machine Learning for Signal Processing (MLSP), 2015 IEEE 25th International Workshop on, 2015
Liste complète des métadonnées

Littérature citée [24 références]  Voir  Masquer  Télécharger

https://hal-lirmm.ccsd.cnrs.fr/lirmm-01234225
Contributeur : Marc Chaumont <>
Soumis le : jeudi 26 novembre 2015 - 14:37:43
Dernière modification le : jeudi 11 janvier 2018 - 06:26:18
Document(s) archivé(s) le : samedi 29 avril 2017 - 04:04:12

Fichier

MLSP2015_PASQUET_CHAUMONT_SUBS...
Fichiers éditeurs autorisés sur une archive ouverte

Identifiants

  • HAL Id : lirmm-01234225, version 1

Citation

Jérôme Pasquet, Marc Chaumont, Gérard Subsol, Mustapha Derras. An efficient multi-resolution SVM network approach for object detection in aerial images. MLSP: Machine Learning for Signal Processing, Sep 2015, Boston, United States. Machine Learning for Signal Processing (MLSP), 2015 IEEE 25th International Workshop on, 2015. 〈lirmm-01234225〉

Partager

Métriques

Consultations de la notice

78

Téléchargements de fichiers

139