摘要
指纹识别是一种生物识别技术,指纹匹配是指纹识别的一个重要过程,如何解决指纹匹配过程中出现的待识别指纹出现的指纹平移、旋转与挤压引起的形变问题,是本文重要的研究内容。提出了一种基于局部特征点的三角形模型匹配算法,根据指纹中图像中的细节点,与其相邻的局部细节点构成一个三角形,并引入了它与邻近细节点间的距离、脊线数目作为判断因素,从而有效解决指纹图像匹配过程中遇到的因平移、旋转与挤压引起形变导致的识别率较低的问题,提高了指纹识别的精确度和可靠性。最后通过3个实验的结果表明,本文提出的算法比另外2种算法具有拒识率低、正确识别率高(正确率达到98.92%)的特点,即基于三角形局部特征点模型的指纹匹配算法有较好的性能。
Fingerprint identification as one of biometric technology, fingerprint matching is an important process. It is a question on how to solve the identified fingerprint problem caused by the fingerprint translation, rotation and deformation because of different pressure of fingerprint, and it is the important subject of this article resolved. In the paper I propose a local triangular model of the feature point matching algorithm based on neighbor point. It is based on the image of fingerprint minutiae and adjacent local minutiae to form a triangle, and import two judgment factors for the algorithm which is distance factor and the ridge line number factor between the two neighboring minutiae points of triangle, in order to effectively address encountered the low recognition rate in the matching process of the fingerprint image caused by the translation, rotation and extrusion deformation. The fingerprint matching algorithm based on local triangular feature point model improved the precision of the fingerprint identification and made it more reliability. Finally, the experimental comparison analysis results obtained the triangle local feature points model based fingerprint matching algorithm has better performance.
出处
《重庆师范大学学报（自然科学版）》
CAS
CSCD
北大核心
2013年第2期69-73,共5页
Journal of Chongqing Normal University:Natural Science
基金
河南省科技项目(No.112400450137)
关键词
指纹识别
局部特征点
匹配算法
极坐标
可靠性
fingerprint recognition
local feature points
matching algorithm
polar coordinates
reliability
