基于Matlab的多色车牌识别系统优化设计  

Optimized design of multiple colors license plate recognition system based on Matlab

作　　者：卢朝辉 王仁平[1] 蔡挺 LU Chaohui;WANG Renping;CAI Ting(College of Physics and Information Engineering,Fuzhou University,Fuzhou 350108,China)

机构地区：[1]福州大学物理与信息工程学院,福建福州350108

出　　处：《电子设计工程》2024年第21期137-140,共4页Electronic Design Engineering

基　　金：福建省自然科学基金项目(2023J01398)。

摘　　要：以车牌识别专用芯片设计为前提,应用图像处理的方式设计和优化多色车牌识别系统。在车牌定位算法中将数学形态学定位与颜色定位相结合,弥补了传统单一定位算法定位不准确的缺陷;在车牌二值化过程中采用去背景二值化算法,减弱了光照不均匀等因素对车牌二值化的影响;在字符分割算法中对投影法进行优化,对字符进行粗分割后,对可能受到影响的字符进行细分割,提高了字符分割的准确性;在字符识别算法中采取二次匹配等操作对传统的模板匹配法进行优化,进一步提高了车牌识别率。仿真结果表明,该系统对不同颜色车牌号综合识别率达到了94.5%,为后续芯片设计提供算法依据。Based on the design of a dedicated chip for license plate recognition,a multiple colors license plate recognition system is designed and optimized using image processing.Combining the mathematical morphological positioning with the color positioning in the license plate positioning algorithm,which makes up for the inaccurate positioning of the traditional single positioning algorithm.In the process of license plate binarization,the de⁃background binarization algorithm is adopted,which weakens the influence of uneven illumination and other factors on the binarization of license plates.The projection method is optimized in the character segmentation algorithm,the characters are first coarsely segmented and then the characters that may be affected are finely segmented,which improves the accuracy of character segmentation.The character recognition algorithm adopts secondary matching and other operations to optimize the traditional template matching method,which further improves the license plate recognition rate.The simulation results show that the comprehensive recognition rate of different color license plate numbers reaches 94.5%,which provides an algorithm basis for subsequent chip design.

关 键 词：车牌识别 车牌定位 字符分割 字符识别 

分 类 号：TN492[电子电信—微电子学与固体电子学]