Classifiers fusion of pedestrian detection based on Choquet integral

Pedestrian detection is a hot research topic because of its importance in many applications, especially in the fields of automobiles, monitoring and robotics. In this paper, we come up with new a new approach to pedestrian detection by fusing SVM classifier and CNN classifier based on Choquet integral. We use fuzzy measures to represent the attribution of SVM classifier and CNN classifier to the final output, and fuse the results of two kinds of classifiers by Choquet integral to get more accurate results.