WCSE 2017
ISBN: 978-981-11-3671-9 DOI: 10.18178/wcse.2017.06.221

Traffic Measurement of Traffic Light Based on Binocular Vision

Yue Wu, Linli Zhou, Yue Duan

Abstract— The accidents of the traffic light intersection have posed a great threat to the safety of people's life and property. In this paper, we study a kind of the front of the traffic light distance measuring technology based on binocular vision, through two CCD cameras capture the image information in front of traffic lights from different perspectives, we get two pieces of binocular images at the left and right , which are processed、positioned,and then, we find the matching points of the corresponding feature with using a global matching method of feature point extracted, which obtain parallax,then according to the principle of triangulation again, we restore the three-dimensional information of the measured object and depth of information to achieve the distance in front of traffic lights measurement. The experimental simulation results show that the design can effectively measure the distance in front of the traffic light, simple and effective, and has the high accuracy of measurement.

Index Terms— binocular vision, camera calibration, stereo matching, three-dimensional distance measurement

Yue Wu, Linli Zhou, Yue Duan
Institute of Intelligent Machines, Chinese Academy of Sciences, CHINA

ISBN: 978-981-11-3671-9 DOI: 10.18178/wcse.2017.06.17Xsrc="http://www.wcse.org/uploadfile/2019/0823/20190823055609629.png" style="width: 120px; height: 68px;" />[Download]

Cite: Yue Wu, Linli Zhou, Yue Duan, "Traffic Measurement of Traffic Light Based on Binocular Vision," Proceedings of 2017 the 7th International Workshop on Computer Science and Engineering, pp. 1268-1275, Beijing, 25-27 June, 2017.