Human tracking based on vision and laser sensor

Abstract

In order to solve the problem of human tracking, we present a real-time person tracking system for a mobile robot based on the fusion of monocular camera and laser sensor. Firstly, the mark model constructed to detect the target by the combination of color and geometric features is developed, and the robot obtains the coordinates of the target by using the principle of monocular distance measurement. Then, a method is proposed to detect the human leg by using the method of geometric feature detection based on self-adapting threshold segmentation of laser information. Finally, the weighted fusion strategy is used to fuse the results of laser and monocular vision to determine the target person, and a Kalman filter is used to obtain the stable trajectory. Experimental results performed in indoor environments show the effectiveness and robustness of the developed approach.