摘要
指纹图像细化指将指纹脊线削减为单像素宽的线条,它能够有效地提高自动指纹识别系统(AFIS)的性能.根据OPTA算法及其改进算法,结合AFIS对图像细化算法在运算速度和系统资源两方面的要求,提出一种基于快速细化算法并引入OPTA算法优点的改进算法,该算法对指纹二值图像的细化结果满足8连接条件.在FVC指纹数据库上的测试结果表明,文中算法能够获得正确的指纹细化图像,运算速度满足AFIS的要求.
Fingerprint image thinning is the operation to delete the width of ridge to one-pixel, which could improve the performance of automated fingerprint identification system （AFIS） effectively. Based on the advantages and disadvantages of OPTA algorithm and its variants, combined with the system requests on calculating velocity and resources of thinning algorithm, an improvement of quick thinning algorithm is proposed, which still contains the advantages of OPTA algorithm and could be used to get 8-connected thinned image. Experiments on FVC database showed the proposed algorithm could produce correct thinned image and have an effective velocity meeting the requirement of AFIS.
出处
《计算机辅助设计与图形学学报》
EI
CSCD
北大核心
2009年第2期179-182,189,共5页
Journal of Computer-Aided Design & Computer Graphics
关键词
自动指纹识别系统
图像细化
OPTA算法
快速细化算法
automated fingerprint identification system（AFIS）
image thinning
OPTA algorithm
quick thinning algorithm
