Research on Pedestrian Detection System based on Tripartite Fusion of "HOG+SVM+Median filter"

Aiming at the problem of low accuracy of traditional pedestrian detection methods in a complex monitoring environment, based on the histogram of oriented gradient (HOG) and support vector machine (SVM) pedestrian detection algorithms, incorporating median filter (MF), a pedestrian detection model based on tripartite fusion (PDMTF) is proposed. The model first uses the median filter algorithm to denoise the image, effectively reducing the impact of noise on the HOG feature descriptor. Then, uses the extracted pedestrian features to train the SVM classifier. In addition, in order to optimize the SVM classifier, the model conducts a secondary training on the misidentified pedestrian area. The final experimental results show that the pedestrian false detection rate of the PDMTF model is only 7%, which has a good pedestrian recognition rate in complex environment.