智慧路灯场景下车祸检测预警系统设计与实现  

Design and Implementation of Traffic Accident Detection and Early Warning System Based on Intelligent Street Lamps

作　　者：宋鑫 张珣[1] SONG Xin;ZHANG Xun(School of Electronic Information,Hangzhou Dianzi University,Hangzhou 310018,China)

机构地区：[1]杭州电子科技大学电子信息学院,浙江杭州310018

出　　处：《软件导刊》2021年第10期197-201,共5页Software Guide

摘　　要：随着社会经济的快速发展,车在人们的生活中越来越普及。为解决许多车祸由于发生在偏僻道路处或晚上人流稀少时,往往会很晚才被发现导致救治不及时的问题,设计一个智慧路灯场景下的车祸检测预警系统。该系统主体算法采用模式匹配算法,先采用YOLOv3算法对获取的图像信息进行筛选,并结合背景差分法和卡尔曼滤波法对视频图像进行处理,再根据车辆速度信息进行辅助判断。实验结果表明,该方法的识别精度达到90%左右,相比传统方法提高了约20%,识别时间减少了约0.3s,有效减少了环境因素的影响。With the rapid development of social economy,Cars are becoming more and more popular in people's lives.In order to solve many accidents occuring on the way of remote place and tend to be discovered untimely for saving lives,a vehicle detection based on intelligent street light traffic accident warning system is designed.The system main algorithm uses pattern matching algorithm,advance the YOLOv3 algorithm is adopted to obtain the filtered image information,combining background difference method and the method of Kalman filter for video image processing,along with the vehicle acceleration and speed information of the auxiliary judgment.The experimental results show that the recognition accuracy of the proposed method reaches about 90%,which is about 20%higher than that of the traditional method,and the recognition time is about 0.3s less than that of the traditional method,which excludes the influence of environmental factors.

关 键 词：人工智能 物联网 智慧路灯 车祸检测 

分 类 号：TP319[自动化与计算机技术—计算机软件与理论]