【作者】 韩广良； 陈小云；

【Author】 Han Guangliang Chen Xiaoyun (Changchun Institute of Optics,Fine Mechanics and Physics,Chinese Academy of Sciences,Changchun,130033,China)

【机构】 中国科学院长春光学精密机械与物理研究所；

【摘要】 本文提出一种纸币图像自动识别定位的方法,在这种图像中,纸币可能存在旋转、正反面、正倒放、新旧、破损等各种情况,使得其对应的图像多种多样,本文针对这种情况,提出一种普遍适用的基于图像特征点匹配的纸币状态自动识别方法,利用纸币上的几个典型特征点进行特征匹配,并通过特征匹配的结果进行融合分析,从而能够区分出纸币的各种位置,进而检测出其自身位置状态,为后续的进一步识别提供充分的信息,本文通过一些人民币图像样本进行了实验,验证了此方法的有效性。更多还原

【Abstract】 This paper presents recognition and positioning method for paper currency image.In such images,there may be a variety of situations such as rotation,positive and negative,upside down,old,damaged and etc.,therefore,a variety of corresponding image is possible.So,this paper presents a auto recognition method of paper currency based on image feature point matching,that is. several typical feature points were used to perform the characteristics matching,and the matching result were used to fusion analysis, so various locations of this paper currency in this image were obtained,and finally its position status can be detected.In the last of this paper,the result of some experimental of dozens RMB image samples are given,to verify the effectiveness of this method.更多还原