DocumentCode :
3287184
Title :
A new multiclass SVM algorithm and its application to crowd density analysis using LBP features
Author :
Fradi, Hajer ; Dugelay, Jean-Luc
Author_Institution :
EURECOM, Sophia Antipolis, France
fYear :
2013
fDate :
15-18 Sept. 2013
Firstpage :
4554
Lastpage :
4558
Abstract :
Crowd density analysis is a crucial component in visual surveillance for security monitoring. In this paper, we propose to estimate crowd density at patch level, where the size of each patch varies in such way to compensate the effects of perspective distortions. The main contribution of this paper is two-fold: First, we propose to learn a discriminant subspace of the high-dimensional Local Binary Pattern (LBP) instead of using raw LBP feature vector. Second, an alternative algorithm for multiclass SVM based on relevance scores is proposed. The effectiveness of the proposed approach is evaluated on PETS dataset, and the results demonstrate the effect of low-dimensional compact representation of LBP on the classification accuracy. Also, the performance of the proposed multiclass SVM algorithm is compared to other frequently used algorithms for multi-classification problem and the proposed algorithm gives good results while reducing the complexity of the classification.
Keywords :
compensation; distortion; feature extraction; image classification; image representation; pedestrians; security; support vector machines; video surveillance; LBP feature; LBP low dimensional compact representation; PETS dataset; classification accuracy; complexity reduction; crowd density analysis; discriminant subspace; distortion effect compensation; local binary pattern; multiclass SVM algorithm; multiclassification problem; patch level; patch size variation; relevance score; security monitoring; visual surveillance; Crowd density; dimensionality reduction; local binary pattern; multiclass SVM;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Image Processing (ICIP), 2013 20th IEEE International Conference on
Conference_Location :
Melbourne, VIC
Type :
conf
DOI :
10.1109/ICIP.2013.6738938
Filename :
6738938
Link To Document :
بازگشت