A method for detecting pedestrians in video surveillance scenes

Detecting pedestrian accurately from natural scenes makes the important impact on intelligent video surveillance. In this paper, we combine motion information, human skin color information, human shape information and variation of ambient lighting to detect pedestrians for the application of automated video surveillance. The moving objects in the video sequence images are extracted using the multi-frame differencing method with adaptive ambient illumination changes. The adaptive ambient illumination human skin feature extraction algorithm extracts human skin color in different lighting changes in order to tackle the problem that skin color is susceptible to illumination. Improve Hough transform is used to automatically determine the size of human head in different scenes. The experimental results show that the method presented in this paper is feasible and is suitable for online applications in moving human detection in natural scenes.