摘要
针对实际应用中受污损的模糊车牌文字识别过程存在着计算复杂或识别率欠佳等问题,提出一种基于模糊集IFScom贴近度的模糊车牌文字识别算法。IFScom贴近度是区分三种不同否定模糊集的相似性度量,该算法对污损车牌图像进行预处理得到像素点的分布密度函数,提取出车牌字符的模糊向量特征,利用IFScom贴近度来实现模糊文字的识别,提高车牌的识别准确率。通过实验结果表明,该方法在模糊车牌文字的识别中是合理有效的。
In order to solve the problems of complex calculation or poor recognition rate in the text recognition process of defaced fuzzy license plate,a fuzzy text image recognition algorithm based on IFScom closeness degree was proposed.IFScom degree is a similarity measure of fuzzy sets which distinguish three different kind of negation,the algorithm preprocesses the image of defilement license plate to obtain the distribu⁃tion density function of pixels,the fuzzy vector features of the license plate characters were extracted,and the IFScom closeness degree was used to realize the recognition of the fuzzy characters,which improved the recognition accuracy of the license plate.The experimental re⁃sults show that this method is reasonable and effective in the recognition of fuzzy license plate characters.
作者
吴晓刚
汪静
WU Xiao-gang;WANG Jing(School of Information technology,Xingyi Normal University for Nationalities,Xingyi 562400;Department of Computer Science and Technology,Tongji University,Shanghai 201804)
出处
《现代计算机》
2020年第21期46-49,共4页
Modern Computer
基金
贵州省人工智能和医学诊断重点实验室(黔教合KY字[2018]006)
贵州省科学技术基金资助项目(黔科合基础[2019]1458号)
贵州省教育厅重点项目(黔教合KY字[2015]403)。
