Research on Pedestrian Detection Technology Based on AdaBoost-RBFSVM Algorithm Intelligent Monitor

Abstract

Make a cascade classification by AdaBoost and RBFSVM algorithm, the HOG features are extracted using HOGDescriptor of OpenCV. From the pedestrian detection experiment with the specified samples, the result shows that the AdaBoost-RBFSVM cascade classification is superior in Positive rate and False Positive rate of classification, the training time is greatly reduced.