基于一维灰度投影的拼接币左右编号匹配识别算法  

An Algorithm for Matching Recognition of Left and Right Numbers of Spliced Money Based on One-Dimensional Gray Scale Projection

作　　者：王志杨 刘金龙 WANG Zhi-yang;LIU Jin-long(Department of Electronics Engineering,Anhui Vocational College of Electronics&Information Technology,Bengbu 233030,China;School of Electronics Engineering,Jiangsu Ocean University,Lianyungang 222005,China)

机构地区：[1]安徽电子信息职业技术学院电子工程系,安徽蚌埠233030 [2]江苏海洋大学电子工程学院,江苏连云港222005

出　　处：《安徽师范大学学报（自然科学版）》2021年第5期426-433,共8页Journal of Anhui Normal University(Natural Science)

基　　金：2020年度安徽省高校优秀青年人才支持计划重点项目(gxyqZD2020073).

摘　　要：针对自动存款机中的拼接币问题,提出了一维灰度投影模板匹配算法识别左右编号。首先对图像进行预处理,以使后面的处理更加简单;然后应用精准定位法分割出左右编号区域;再计算两编号区的一维灰度投影曲线,并根据投影曲线分割每一个字母或数字;最后进行一维灰度投影模板匹配运算,求出每个字母或数字的相关系数,与门限值进行比较,判断左右编号是否互相匹配。实验表明该算法具有快速、高效的特点,并且识别正确率达到100%。To solve the problem of spliced money in ATM,the paper puts forward a template matching algorithm for recognition of left and right numbers based on one-dimensional gray scale projection.Firstly it preprocesses the image to simplify the after treatments;secondly it applies the precise location to divide the area of left and right numbers;thirdly calculates the one-dimensional gray scale projection curves of left and right numbers,based on which every letter or number is divided;finally it carries out the operation of template matching to obtain correlation coefficients of every letter or number,compares them with threshold and judges whether left and right numbers are matched.The experiments show that the algorithm is fast and efficient,and its recognition accuracy rate is up to 100%.

关 键 词：形态学分析 灰度均衡 平滑滤波 反色 图像二值化 

分 类 号：TN391.41[电子电信—物理电子学]