谷歌浏览器插件
订阅小程序
在清言上使用

The Anti Counterfeiting Algorithm for Overall Security Code Based on Registration Principle

2021 IEEE 6th International Conference on Signal and Image Processing (ICSIP)(2021)

引用 0|浏览15
暂无评分
摘要
In view of the fact that many security codes can not be correctly identified due to interference factors such as character adhered, complex background and strong light, the anti-counterfeiting method is changed to take the specific area containing the security code as the whole object, and then the IORA(Improved ORB Registration Algorithm) based on the registration principle is designed for the anti-counterfeiting method. Firstly, FAST corner detection algorithm with adaptive threshold and non-maximum suppression processing are used to extract non-clustered feature points; secondly, rBRIEF and KNN algorithm are used to describe and coarse match feature points; finally, the Triangle Like algorithm is proposed to fine match feature points by judging whether two triangles composed of three pairs of feature points are similar, so as to improve the target matching rate. The experimental results show that IORA algorithm has stronger robustness, accuracy and anti-light interference ability in anti-counterfeiting function.
更多
查看译文
关键词
overall security code,registration principle,anti-counterfeiting,triangle like
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要