Computer Engineering and Applications ›› 2022, Vol. 58 ›› Issue (10): 276-282.DOI: 10.3778/j.issn.1002-8331.2011-0032

• Engineering and Applications • Previous Articles     Next Articles

Vehicle Detection Algorithm Based on Pulse Coherent Radar

ZHANG Xiancai, ZHANG Zusheng, LI Yiguang, GUO Xiaohong, CHEN Wei   

  1. 1.School of Cyberspace Security, Dongguan University of Technology, Dongguan, Guangdong 523808, China
    2.School of Computer Science, Guangdong University of Technology, Guangzhou 510006, China
  • Online:2022-05-15 Published:2022-05-15

基于脉冲相干雷达的车辆检测算法

张先才,张足生,李益广,郭小红,陈伟   

  1. 1.东莞理工学院 网络空间安全学院,广东 东莞 523808
    2.广东工业大学 计算机学院,广州 510006

Abstract: Vehicle detection technology is the foundation of roadside intelligent parking management system. This paper presents an outdoor parking detection algorithm based on a new pulse coherent radar. Deploy a sensor node in the center of each parking space, then perform feature extraction and multiple predictions on the radar sampled signal, and integrate the results of multiple predictions to realize parking space detection. In each prediction, use all characteristic information of the radar signal and different parameters which are calculated by the particle swarm optimization algorithm according to the radar data set. Finally, deploy an experimental system. In the non-interference environment, the experimental results show that the detection accuracy of the algorithm is 99.9%. When there is interference caused by water or wet leaves, the accuracy of the algorithm is 91%. Compared with the existing algorithms, this algorithm not only improves the accuracy of parking detection in interference scenarios, but also can detect whether the radar is interfered by water or wet leaves.

Key words: intelligent transportation system, parking detection, radar sensor, Internet of things, particle swarm optimization

摘要: 车辆检测技术是路边智慧停车管理系统的基础。提出一种基于新型脉冲相干雷达的室外停车检测算法。在每个停车位中央部署传感器节点,然后对雷达采样信号进行特征提取和多次预测,综合多次预测的结果实现停车检测。每次预测时,利用雷达信号所有的特征信息,并且使用不同的参数,这些参数由粒子群优化算法根据雷达数据集计算得到。部署了实验系统,实验结果表明,在无干扰情况下,算法准确率为99.9%;在水和湿树叶等干扰情况下,算法准确率为91%。相比已有算法,该算法不仅提高了干扰场景下停车检测的精度,而且还可以检测出雷达是否受到水或湿树叶的干扰。

关键词: 智能交通系统, 停车检测, 雷达传感器, 物联网, 粒子群优化