Performance Improvement on Edge-Based Human Detection Using Local Contrast Enhancement
Xue Yuan, Xue Ye Wei, Yong Duan Song · Advanced materials research · 2011
This paper presents a local contrast enhancement method, which is able to improve the detection performance of edge-based human detection. First, a neighborhood dependent local contrast enhancement method is used to enhance the images contrast. Next, the cascade AdaBoost classifier is used to discriminate between human and non-human. Experimental results show that the performance of our method is about 5% better than that of the conventional method.