A new multiclass SVM algorithm and its application to crowd density analysis using LBP features

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.

[1]  Ian T. Jolliffe,et al.  Principal Component Analysis , 2002, International Encyclopedia of Statistical Science.

[2]  Xuran Zhao,et al.  Crowd density analysis using subspace learning on local binary pattern , 2013, 2013 IEEE International Conference on Multimedia and Expo Workshops (ICMEW).

[3]  Yangsheng Xu,et al.  Crowd Density Estimation Using Texture Analysis and Learning , 2006, 2006 IEEE International Conference on Robotics and Biomimetics.

[4]  Chih-Jen Lin,et al.  A comparison of methods for multiclass support vector machines , 2002, IEEE Trans. Neural Networks.

[5]  David J. Kriegman,et al.  Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projection , 1996, ECCV.

[6]  Hong Liu,et al.  Crowd Density Estimation Based on Local Binary Pattern Co-Occurrence Matrix , 2012, 2012 IEEE International Conference on Multimedia and Expo Workshops.

[7]  Lei Huang,et al.  Advanced Local Binary Pattern Descriptors for Crowd Estimation , 2008, 2008 IEEE Pacific-Asia Workshop on Computational Intelligence and Industrial Application.

[8]  Bill Triggs,et al.  Histograms of oriented gradients for human detection , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[9]  R. Y. Tsai,et al.  An Efficient and Accurate Camera Calibration Technique for 3D Machine Vision , 1986, CVPR 1986.

[10]  Abishai Polus,et al.  Pedestrian Flow and Level of Service , 1983 .

[11]  S. A. R. Abu-Bakar,et al.  Crowd Estimation Using Histogram Model Classification Based on Improved Uniform Local Binary Pattern , 2012 .

[12]  Matti Pietikäinen,et al.  Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[13]  A. Marana,et al.  Estimation of crowd density using image processing , 1997 .

[14]  Hua Yang,et al.  The large-scale crowd density estimation based on sparse spatiotemporal local binary pattern , 2011, 2011 IEEE International Conference on Multimedia and Expo.

[15]  Bernard Mérialdo,et al.  A Multimedia Retrieval Framework Based on Automatic Graded Relevance Judgments , 2012, MMM.

[16]  Wen Gao,et al.  Review the strength of Gabor features for face recognition from the angle of its robustness to mis-alignment , 2004, Proceedings of the 17th International Conference on Pattern Recognition, 2004. ICPR 2004..