Research on Vehicle Forward Target Recognition Algorithm Based on Vision and MMW Radar Fusion

Vehicle forward target recognition is the most concerned part in the field of environmental perception. In order to overcome the limitation of single sensor in target recognition, this paper proposes a forward target perception algorithm based on fusion of camera and millimeter wave (MMW) radar. Considering the characteristics of object and sensor, this paper divides vehicle forward targets into two categories: close-range target and distant target. For the close-range target, the target information obtained by the two sensors is matched and fused at the data level by using object recognition, monocular vision ranging, Kalman filter and other algorithms. For the distant target, the initial position is determined by the radar detection point, and the target is accurately classified by the visual algorithm. Experimental results show that the proposed algorithm can effectively reduce the rate of missed detection and improve the target stable recognition distance to 90 m. Besides, more accurate and abundant target information can be obtained by this method.

[1]  Shan Ouyang,et al.  Power Allocation and Relay Selection for Multisource Multirelay Cooperative Vehicular Networks , 2016, IEEE Transactions on Intelligent Transportation Systems.

[2]  Zhengyou Zhang,et al.  A Flexible New Technique for Camera Calibration , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  Lian Xiaomin Study on Real-time Distance Detection Based on Monocular Vision Technique , 2006 .

[4]  Janne Heikkilä,et al.  A four-step camera calibration procedure with implicit image correction , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[5]  Ricardo Omar Chávez García,et al.  Fusion at detection level for frontal object perception , 2014, 2014 IEEE Intelligent Vehicles Symposium Proceedings.

[6]  Wei Liu,et al.  SSD: Single Shot MultiBox Detector , 2015, ECCV.

[7]  Denis F. Wolf,et al.  Fast Metric Tracking by Detection System: Radar Blob and Camera Fusion , 2015, 2015 12th Latin American Robotics Symposium and 2015 3rd Brazilian Symposium on Robotics (LARS-SBR).

[8]  Nanning Zheng,et al.  Frontal object perception for Intelligent Vehicles based on radar and camera fusion , 2016, CCC 2016.

[9]  Nanning Zheng,et al.  On-Road Vehicle Detection and Tracking Using MMW Radar and Monovision Fusion , 2016, IEEE Transactions on Intelligent Transportation Systems.

[10]  Y. Bar-Shalom Tracking and data association , 1988 .