摘要
近些年来,中国经济飞速增长,汽车产业也已进入持续发展新阶段。在此背景下,对于驾驶人来说除了拥堵的困扰之外,停车难、找车难的问题也日益突出。因此,本文提出了基于机器视觉的停车场智能寻车系统,旨在为复杂停车场环境下快速寻车、降低用户时间成本、及推进现有智能停车场建设等提供有效支持。该系统包含三大模块,车牌识别模块负责输出车牌文本,与移动终端模块人工输入的车牌信息进行匹配,进而确定车辆所在位置,最后由路径规划模块实现寻车路径规划。系统在移动终端触控操作,并显示最短路线,完成智能寻车。测试实验表明,该系统算法在改进后具有良好的鲁棒性和更快的检测准确率,车牌检测准确率可达到96.86%,并且路径规划时长大幅缩短,较好地实现了复杂停车场环境下快速智能寻车的目的。
In recent years,China’s economy has grown rapidly,and the development of the automobile industry has entered a new stage of sustainable development.In this context,in addition to the problems of congestion,the problems of difficulty in parking and finding a car have become increasingly prominent for drivers.Therefore,this article proposes the machine vision-based parking lot intelligent car-finding system,which aims to quickly find a car in a complex parking environment,reduce user time costs,and promote the construction of existing smart parking lots.The system contains three modules.The license plate recognition module is responsible for outputting the license plate text,and matching it with the license plate text manually input by the mobile terminal module,then determining the location of the vehicle to be found.Finally,the path planning module implements the shortest path planning.The system touches the mobile terminal and displays the shortest route to complete the intelligent car search.The test results show that the improved algorithm has good robustness and faster detection accuracy,the license plate detection accuracy can reach 96.86%,and the time of path planning is greatly shortened,which better realizes the purpose of fast intelligent car search in complex parking environment.
作者
杜嘉豪
黄孝慈
邢孟阳
DU Jiahao;HUANG Xiaoci;XING Mengyang(School of Mechanical and Automotive Engineering,Shanghai University of Engineering Science,Shanghai 201620,China)
出处
《智能计算机与应用》
2022年第5期102-106,共5页
Intelligent Computer and Applications
关键词
机器视觉
车牌识别
路径规划
停车场管理
machine vision
vehicle license plate recognition
path planning
parking lot management
