This paper proposes a vision-based pedestrian detection in crowded situations based on a single camera. The main idea behind our work is to fuse multiple cues so that the major challenges, such as occlusion and comple...This paper proposes a vision-based pedestrian detection in crowded situations based on a single camera. The main idea behind our work is to fuse multiple cues so that the major challenges, such as occlusion and complex background facing in the topic of crowd detection can be successfully overcome. Based on the assumption that human heads are visible, circle Hough transform (CHT) is applied to detect all circular regions and each of which is considered as the head candidate of a pedestrian. After that, the false candidates resulting from complex background are firstly removed by using template matching algorithm. Two proposed cues called head foreground contrast (HFC) and block color relation (BCR) are incorporated for further verification. The rectangular region of every detected human is determined by the geometric relationships as well as foreground mask extracted through background subtraction process. Three videos are used to validate the proposed approach and the experimental results show that the proposed method effectively lowers the false positives at the expense of little detection rate.展开更多
文摘This paper proposes a vision-based pedestrian detection in crowded situations based on a single camera. The main idea behind our work is to fuse multiple cues so that the major challenges, such as occlusion and complex background facing in the topic of crowd detection can be successfully overcome. Based on the assumption that human heads are visible, circle Hough transform (CHT) is applied to detect all circular regions and each of which is considered as the head candidate of a pedestrian. After that, the false candidates resulting from complex background are firstly removed by using template matching algorithm. Two proposed cues called head foreground contrast (HFC) and block color relation (BCR) are incorporated for further verification. The rectangular region of every detected human is determined by the geometric relationships as well as foreground mask extracted through background subtraction process. Three videos are used to validate the proposed approach and the experimental results show that the proposed method effectively lowers the false positives at the expense of little detection rate.
