摘要
本文提出了一种改进的模板匹配的污损车牌识别方法,该识别方法对于污损车牌和模糊车牌识别效果较好。研究时充分利用车牌定位和字符分割过程中得到的信息对车牌识别过程进行反馈。利用改进的模板匹配方法进行字符的匹配识别,通过将字符的差异值累加进行扩大,以便于比较字符之间的差异值,通过对比差异值的大小来判断字符识别结果。通过改进的模板匹配方法来识别车牌字符,提高了污损车牌识别的准确性,实验结果表明本方法具有很强的环境适应性和鲁棒性。
In this paper,an improved template matching method is proposed for fouling license plate recognition.The recognition method is better for smeared license plates and fuzzy license plate recognition.More specifically,the method makes full use of the information obtained in the process of license plate location and character segmentation to feed back the process of license plate recognition.The improved template matching method is used to match and recognize characters.By enlarging the difference values of characters,the research compares the difference values between characters,and compares the difference values to determine the result of character recognition.The improved template matching method is used to identify license plate characters,which could improve the accuracy of dirty license plate recognition.Experimental results show that this method has strong environmental adaptability and robustness.
作者
李强
张娟
LI Qiang;ZHANG Juan(School of Electronic and Electrical Engineering,Shanghai University of Engineering Science,Shanghai 201620,China)
出处
《智能计算机与应用》
2019年第3期113-117,共5页
Intelligent Computer and Applications
关键词
模板匹配
车牌识别
污损车牌
车牌字符
模糊车牌
字符匹配识别
template matching
license plate recognition
fouling license plate
license plate character
fuzzy license plate
character matching recognition
