指纹自动识别中图像分割算法研究  

Studies on the image segmentation algorithm of AFIS

作　　者：甘树坤[1] 吕雪飞[1] 吕颖[2] 杨文惠 

机构地区：[1]吉林化工学院机电工程学院,吉林吉林132022 [2]吉林省计量科学研究院,吉林长春130022 [3]中油吉林吉化公司合成树脂厂,吉林吉林132022

出　　处：《电子技术应用》2007年第11期65-67,70,共4页Application of Electronic Technique

摘　　要：为了可靠、准确地实现指纹的自动识别,提出了一种简便、可行的指纹图像的分割算法。通过合理地运用指纹图像的灰度特性,以较低的计算代价有效地解决了指纹图像的分割问题,从而使算法的处理效果好、运行速度快。实验表明,这种分割算法对于指纹图像的预处理十分有效。In this paper, a simpler image pre-processing algorithm for segmentation of fingerprint image is proposed, which can improve the accuracy and reliability of the automatic recognition of fingerprint images. This paper proposes an algorithm, which by reasonable using the gray-level-statistics of fingerprint images, resolves the problem of fingerprint image segmentation with low computational cost. The algorithm has good process result and runs faster and easily to be progrmmned. The experimental result shows that this segmentation algorithm is more effective and robust the pre-process of the fingerprint images.

关 键 词：指纹识别 图像分割 预处理 灰度特性 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]