摘 要：基于边缘计算设备 MEC 接入部署在路口的毫米波雷达和摄像头，通过机器视觉深度学习技术和多传感器目标融合技术，实现对路口本地的车辆、行人、路况的精细化、实时性感知，构建路口的泛感知体系，从而实现道路交通多维度、多来源、全要素的全息感知。通过利用这些全息感知数据可以有效提升路口、区域整体通行能力，提高车辆平均速度，带来城市交通管理成效的显著提升。
关键词：边缘计算；机器视觉；深度学习；目标融合；全息感知
DOI:10.19850/j.cnki.2096-4706.2022.05.026
中图分类号：TP393 文献标识码：A 文章编号：2096-4706（2022）05-0098-03
Holographic Perception at Intersections Based on Fusion of Millimeter-Wave Radar and Machine Vision
CHEN Xinyun
(Zhejiang Hikalllink Technology Co., Ltd., Hangzhou 311100, China)
Abstract: Based on the edge calculation device MEC with millimeter wave radar and camera device deployed on road intersections, this paper realizes the refined and real-time perception on intersection of local vehicles, pedestrians, road refinement through machine vision deep learning technology and multi-sensor target fusion technology, builds intersection general perception system, so as to realize the road traffic multi-dimensional, multi-source, total elements of holographic perception. By using these holographic perception data, it can effectively improve the overall traffic capacity of intersections and regions, improve the average speed of vehicles, and bring a significant improvement in the effectiveness of urban traffic management
Keywords: edge calculation; machine vision; deep learning; target fusion; holographic percepti
参考文献：
[1] 李原 . 毫米波雷达在车路协同系统中的应用研究 [J]. 工业控制计算机，2020，33（1）：44-46+50.
[2] 王宇宁，庞智恒，袁德明 . 基于 YOLO 算法的车辆实时检测 [J]. 武汉理工大学学报，2016，38（10）：41-46.
[3] WANG C Y, LIAO H Y M. YOLOv4：Optimal Speed and Accuracy of Object Detection [J/OL]. arXiv:2004.10934v1 [cs.CV]. (2020-04-23).https://arxiv.org/abs/2004.10934v1.
[4] OH J, KIM K S, PARK M, et al. A Comparative Study on Camera-Radar Calibration Methods [C]//15th International Conference on Control,Automation,Robotics and Vision (ICARCV).Singapore： IEEE，2018：1057-1062.
[5] ZHANG X，ZHOU M，QIU P，et al. Radar and vision fusion for the real⁃time obstacle detection and identification [J].Industrial Robot An International Journal，2019，46（3）：391-395.
作者简介：陈信云（1979—），男，汉族，浙江宁波人，工程师（中级），学士学位，研究方向：车路协同、智慧交通。