Pedestrian detection and localization using 3D range data
Bin Li, Jiadong Shi, Minghe Cao, Rongkai Zhang, Jianzhong Wang · 2016
This paper presents an approach to exploit three-dimensional laser range finder to detect and locate pedestrian over time. The sensor is employed on a vehicle. The approach uses geometric and motion features to detect human signatures. According to the different characteristics in different height of the human body to establish a model of human body. We establish a number of sample models to reflect the diversity of the human body, to improve the probability of a variety of pedestrians detection. Moving objects should be matched with the human body model. Experimental carry out with 3D data illustrate the efficiency and accuracy of our approach.