指纹识别算法实现本科毕业论文.doc

PAGE 8 - PAGE II - 摘 要 生物识别技术已经成为身份识别和网络安全的发展技术之一，其中指纹识别技术是目前公认的安全、准确、方便的身份认证技术之一，是人们研究的热点。 本文对指纹图像识别系统的原理和基本过程进行了较为深入的分析与研究并给与实现。首先对指纹数字图像的分类、基本特征与识别原理进行了详细地论述。其次，本文在指纹数字图像预处理部分从指纹图像的预处理、特征提取、特征匹配这三个必要环节对指纹识别算法进行了详细地研究，最后实现了指纹图像处理的算法。在指纹图像预处理阶段，本文使用基于灰度的算法对图像进行了分割，同时，针对二值化后图像中仍然存在的噪声，也进行了相应的修饰处理，尽可能的为以后指纹特征的提取打好基础。指纹特征提取阶段，采用模板法提取指纹中所有的端点和分叉点，并进行了细化的处理。特征匹配阶段，采用基于细节点的方式进行特征点的匹配。本文设计并实现了一个指纹识别系统，完成了对指纹数字图像的处理、特征提取，保存和匹配等功能，利用编码对指纹数字图像进行入库、匹配等操作，从而完成图像的识别来判断是否同一指纹，其比对结果令人满意。 【关键词】指纹识别 预处理 二值化 特征提取 特征匹配  ABSTRACT Biometric technology has become one of the developing technologies for identity recognition and network security. And fingerprint identification technology is now recognized as one of the most safe, accurate and convenient authentication technologies, and it is a focus for researchers. This paper mainly research the basic principles of digital image of the fingerprint identification system and do more in-depth analysis on process, and finally achieved the algorithm of digital fingerprint precessing. First, this paper explains the classification of fingerprints digital images, basic characteristics and identify the principle in detail. Secondly, in the digital fingerprint image pre-processing part of this paper, three essential aspects of fingerprint : ?fingerprint image?preprocessing, feature?extraction and feature matching are discussed and researched in detail in this paper. Fingerprint image pre-processing stage, This paper presents a gray-based algorithm for image segmentation, at the same time, the noise still exists for the image after binarization, also carried out the appropriate modification of treatment, as much as possible lay a solid foundation for the fingerprint feature extraction . Fingerprint feature extraction stage, the use of all of the endpoints and the bifurcation point in the fingerprint template extraction and refinement processing. Feature matching stage, minutiae-based approach for matching feature points. In this p