Abstract:In the paper,a novel hybrid method of support vector machine and pedestrian representation model is proposed to thermal infrared images process , which can solve the shortcoming of insufficient information
represented by the current pedestrian representation model in infrared images.Firstly, we create the pedestrian representation model. Then,pedestrian body in infrared images can be used for tracking
purpose in the study. The experimental imagies are applied to testify the effectiveness of thermal infrared images process based on support vector machine and pedestrian representation model. In order
to acquire the optimal number of particles, several experiments are performed. The experimental results show that the proposed method is suitable for thermal infrared images process.
Xuejun Yue;Tiansheng Hong;Bin Lu;Ling Yin;Songbin Zhai. Thermal Infrared Images Process Based on Support Vector Machine and Pedestrian Representation Model[J]. , 2010, 7(14): 3069-3073.
Xuejun Yue;Tiansheng Hong;Bin Lu;Ling Yin;Songbin Zhai. Thermal Infrared Images Process Based on Support Vector Machine and Pedestrian Representation Model. , 2010, 7(14): 3069-3073.