Passenger detection for subway transportation based on video

The purpose of this paper is to analyze passengers´ moving direction through the video shot in the entrances and exits of the subway stations. The results of the analysis will be helpful to relevant departments to manage the traffic condition, making a decision in the face of emergency. First of all, this paper adopts Haar features and Adaboost algorithm to implement the detection of human´s head through OpenCV; Secondly, this paper uses color histogram in the head recognition and an improved algorithm that adds the step of comparing the pixel value of the location coordinates in consecutive frames is proposed; At last, the paper realizes the human tracking through the establishment of target tracking chain and puts forward to analyze passengers´ moving direction through space coordinate information.