掌纹识别算法综述  

Survey of Palmprint Recognition Algorithms

作　　者：岳峰[1] 左旺孟[1] 张大鹏[2] 

机构地区：[1]哈尔滨工业大学计算机学院生物信息技术研究中心,哈尔滨150001 [2]香港理工大学计算学系生物识别技术研究中心

出　　处：《自动化学报》2010年第3期353-365,共13页Acta Automatica Sinica

基　　金：国家高技术研究发展计划(863计划)(2006AA01Z308);国家自然科学基金(60872099;60902099)资助~~

摘　　要：掌纹识别作为一种新兴的生物识别技术,近年来得到了广泛的关注与研究.与其他生物特征相比,掌纹有许多独特的优势,包括识别率高、采集设备价格低廉、用户可接受性好等.这些优势使得掌纹识别成为一种有着广泛应用前景的生物识别方法.本文首先介绍了掌纹的特点、掌纹的采集设备和预处理方法,之后详细介绍了近几年来提出的各种掌纹识别方法.根据特征提取以及匹配方法的不同,本文将掌纹识别方法分为基于结构的、基于子空间的、基于编码的和基于统计的四类方法.在回顾和比较了各种算法的特点之后,对未来的掌纹识别方法的发展方向作了展望.As an emerging biometric technology, palmprint recognition has recently received considerable research interest. Because of its low-price capture device, fast execution speed, and high verification accuracy, palmprint recognition is very competitive in biometric research and is expected to have a wide range of security applications. This paper first introduces the characteristics of the palmprint, the acquisition and the preprocessing method for palmprint image, and then presents a survey of the palmprint recognition algorithms. According to different feature extraction and matching approaches, we roughly group these algorithms into four categories： structure based, statistics based, subspace based and coding based methods. After surveying and comparing these algorithms in each category, we conclude this paper with a discussion of future trends.

关 键 词：生物识别 掌纹识别 低分辨率掌纹图像 特征提取 特征匹配 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TP391.4[自动化与计算机技术—计算机科学与技术]