Study on vehicle front pedestrian detection based on 3D laser scanner
Lisheng Jin, Qingning Niu, Haijing Hou, Hu Shunxi, Wang Fangrong · 2011
3D laser scanning system not only has the imaging characteristics as visual sensors, but also has the ability of precise range measurement and position as the radar. In this paper, three-dimensional laser scanner was used to detect pedestrian in front of the host vehicle. The mapping algorithm was proposed to convert the laser data points from range data space to the gray image space. According to characteristics of laser gray range image, process algorithms such as noise filtering, histogram equalization, edge sharpening and threshold segmentation were used. The linear classifier was designed to recognize pedestrians in the gray range image based on shape features. Coordinate matching algorithm was used for pedestrian location. Experiment results show that the proposed pedestrian detection system based on the three-dimensional laser scanner has good reliability and robustness.