基于Matlab的手背静脉图像特征提取.doc

. PAGE . 基于Matlab的手背静脉图像特征提取 摘 要 人体手背静脉识别技术作为一种全新的非接触式生物特征识别技术，与以往传统的指纹以及虹膜识别技术相比表现出了许多明显的优势，近年来得到了广泛的关注。 手背静脉识别是通过分析手背上的静脉特征来进行身份识别的。首先对手背静脉图像预处理，得到细化图像，利用分割思想把细化图像分割为多个子图像，然后分别提取平均原点静矩，组合成特征向量，最后再结合最近邻法，用最小欧氏距离的方法来进行匹配，实验结果验证了该方法的有效性和可行性。 关键词：静脉识别，原点静距，特征提取，匹配 The?Feature?Extraction??of the ?Hand?Vein?Image?Based?on?Matlab Author：Cai Hongshan Tutor：Cai Chaofeng ABSTRACT With the arrival of information age, information security has become increasingly important. Many occasions need to identify their visitors and biometric identification technology has provided effective solutions to such problems. As a new non-contact Biometric identification technology, hand vein verification technology has many obvious advantages compared with the traditional fingerprint and iris recognition technology and has received attention widely. Human hand vein recognition technology can improve the reliability of identity resolution in virtue of vein feature on the back of the hand. Based on the research and analysis of the recent biometric identification technology research results, this paper conducts deep research on some of the key technologies of the hand vein recognition technology, especially on feature extraction and feature matching. In the feature extraction stage, this paper’s algorithm based on thinned vein image takes origin static moment as recognition features. We can get the refining image through the preprocessing of the hand vein image, divide the refining image into multiple sub-images, extract the average origin static moment into the feature vector, use the smallest Euclidean distance to match it. The experimental result verifies the effectiveness and feasibility of the method. KEY WORDS: Vein?Verification ,origin?static?moment ,feature?extraction . . 目 录 TOC \o 1-3 \h \z \u HYPERLINK \l _Toc356839646 1绪论 PAGEREF _Toc356839646 \h 1 HYPERL