Pedestrian Detection Algorithm Using Local Feature Cascade Classifier in Traffic Scenes

Aiming at the low recognition rate of pedestrian detection algorithm in traffic scene, this paper designs a pedestrian recognition algorithm based on local feature cascade classifier. The pedestrian candidate area is extracted, and the imaging intensity of the scattered light in the camera sensor is calculated. A local feature cascade classifier is used to segment the image boundary, and the rectangular box whose height is smaller than the height of the HOG window is enlarged to design a pedestrian detection algorithm for traffic scenes. The experimental results show that the average recognition rate of the pedestrian detection algorithm proposed in this paper is 94.764%. The results show that the proposed method can achieve better pedestrian recognition in traffic scenes after introducing local feature cascade classifier.