基于改进的高斯模糊算法的车牌识别系统开发与验证  

Development and Verification of License Plate Recognition System Based on Improved Gaussian Fuzzy Algorithm

作　　者：张利芬 时培成[2] 葛胜升[1] ZHANG Li-fen;SHI Pei-cheng;GE Sheng-sheng(College of Automobile and Aviation,Wuhu Institute of Technology,Wuhu Anhui 241006,China;College of Mechanical and Vehicle Engineering,Anhui University of Engineering,Wuhu Anhui 241000,China)

机构地区：[1]芜湖职业技术学院汽车与航空学院,安徽芜湖241006 [2]安徽工程大学机械与车辆工程学院,安徽芜湖241000

出　　处：《萍乡学院学报》2023年第3期66-70,共5页Journal of Pingxiang University

基　　金：安徽省优秀青年骨干教师国内访学研修项目“四轮毂驱动电动汽车线控电制动特性研究”(gxgnfx2021189);安徽省职业与成人教育学会教育教学研究规划课题“产教深度融合背景下新工科智能制造汽车类专业人才培养模式的探究与实践”(Azcj2021059)。

摘　　要：车牌识别是车辆识别的重要内容,交通管理系统的智能化的发展对车牌识别的准确度、速度、鲁棒性、检测的空间范围等提出了更高的要求。为实现快速、高准确率及高适应性的车辆识别效果,课题组基于图片处理的车牌识别方法,利用改进的高斯模糊算法,基于OpenCV软件进行车牌识别系统设计,以检测距离和检测角度为主要变量,搭建试验平台,以识别的速度,识别准确率及鲁棒性作为主要评价指标进行车牌识别试验验证。38组测试结果显示,除一次极限状态,和一次试错外,该系统均能快速、准确地完成车牌识别工作。License plate recognition is an important part of identification,and the development of intelligent traffic management system brings about higher requirements on the accuracy,speed,robustness and spatial range of license plate recognition.In order to achieve fast,high accuracy and high adaptability of the vehicle recognition effect,the image processing license plate recognition method and the improved Gaussian fuzzy algorithm were used to design a license plate recognition system based on the OpenCV software.A test platform was built to test the system.Taking the distance and angle of detection as the main variables,the main evaluation indexes of the license plate recognition,such as the recognition speed,recognition accuracy and robustness,were verified.The 38 sets of test results show that except for one ultimate state,and one trial and error,the system can quickly and accurately complete the license plate recognition.

关 键 词：图像处理 高斯模糊 OPENCV 车牌识别 

分 类 号：U495[交通运输工程—交通运输规划与管理] TP391.41[交通运输工程—道路与铁道工程]