摘要
变造币是一种利用拼接技术合成的假币。针对第5套人民币中100元的图像特征,文章提出了两种利用模板匹配识别变造币横竖条码的方法。一种方法是运用传统的二维模板匹配算法,依次截取横条码中的字符与竖条码中的对应字符比较,依据模板和目标图像的相关性判断它们是否相同;另一种方法是分别得到横竖条码在水平和垂直方向的一维灰度投影曲线,根据它们的相关性判断两幅图像是否匹配。实验结果表明这两种识别方法均精确快速有效。
Composite-RMB is false money which stick some parts of real money and some parts of counterfeit money together by adhesive tape. In this paper, two algorithms based on cyclostyle matching are proposed for recognizing horizontal and erect numbers of composite-RMB, especially for I00 yuan RMB of fifth series. One method uses 2 D cyclostyle matching algorithm to select characters of horizontal list and compares with erect list if its not homology; The other method uses 1-D cyclostyle matching algorithm to obtain horizontal and erect gray scale projection curves of two list images, then compares with them if its not homology. The methods are proved to be highly effective by many simulated data.
出处
《计算机工程》
EI
CAS
CSCD
北大核心
2006年第10期183-185,共3页
Computer Engineering
关键词
变造币
横竖条码
数学形态学
模板匹配
灰度投影曲线
Composite-RMB
Horizontal and erect characters
Mathematical morphology
Cyclostyle matching
Gray scale projection curve
