一种有效的基于八邻域查表的指纹图像细化算法  

An Efficient Index Thinning Algorithm of Fingerprint Image Based on Eight Neighbourhood Points

作　　者：杨威[1] 郭科[1] 魏义坤[1] 

机构地区：[1]成都理工大学信息管理学院,成都610059

出　　处：《四川理工学院学报（自然科学版）》2008年第2期61-63,共3页Journal of Sichuan University of Science & Engineering(Natural Science Edition)

摘　　要：指纹图像预处理是指纹识别的第一步,它的好坏直接影响着后续指纹特征值提取和匹配的准确性,而细化是又是预处理过程的核心环节。因而文章重点研究了二值化后指纹图像细化算法基础上,提出更有效的基于"八邻域"查表细化算法及其改进算法。实验表明基于八邻域查表细化算法效率更高、处理后几乎没有毛刺、效果更理想,为后续的工作奠定了良好的基础。The preprocessing is the first phase in the process of fingerprint recognition. Only after the preprocessing is performed perfectly, could the subsequent phases be reliable and satisfying. The thinning of fingerprint image is the most important step of the preprocessing, so the thinning algorithm is researched in this paper. During the thinning stage, the thinning method is used based on eight neighbourhood points. Experiments show the last method is simple and the results of thinning are very ideal.

关 键 词：指纹识别 预处理 查表 细化 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]