【作者】 周召敏；

【导师】 余成波；

【作者基本信息】 重庆理工大学 ， 信号与信息处理， 2010， 硕士

【摘要】 现代社会中,计算机及网络技术的高速发展,信息安全显示出前所未有的重要性。人们希望有一种更加方便可靠的办法来进行身份鉴定,生物特征识别技术给这一切带来可能。手指静脉血管藏匿于身体内部,具有速度快、精度高、安全等级高、活体识别、内部特征、非接触式等独特的优势,有效地解决了传统的生物特征识别技术如指纹等所面临的诸多难题。随着静脉识别技术向着产品化的方向发展,对静脉提取算法的要求越来越高。提取算法在兼顾准确性的基础上,还要重点考虑算法的实时性。本研究的主要工作集中在以下几个方面:分析了现阶段国内外手指静脉特征提取算法,并通过仿真比较了其优缺点;根据手指静脉图片在变换域中的特点,提出了一种脊波特征提取技术;根据手指静脉对近红外的吸收特性,提出了一种背景图像的构造以得到第一次残差图像,通过调节比例系数得到第二次残差图像,以间接的方式获取静脉的特征;针对低质量静脉图片特征提取容易将噪声错判成静脉信息,将较差的静脉信息错判成噪声。提出了一种基于线性加权(LWF)免疫克隆的手指静脉特征提取算法。该算法促进静脉信息的生长和抑制噪声的干扰;吸取了免疫克隆算法的核心思想,简化了免疫克隆应用到了手指静脉特征识别的片上系统上。更多还原

【Abstract】 In modern society, the high-speed development of computer and network has shown unprecedented importance of information security. People want to have a more convenient and reliable way for identification, Biometric identification technology has brought it all possible. Finger vein, hidden within the body, has unique advantages in high speed, high precision, live identification and internal features. It’s an effective solution to the traditional biometrics such as fingerprints which face many problems. With the vein recognition technology develops in the direction towards products, the demanding of extraction algorithms on the veins become increasing. Extraction algorithm based on the accuracy, also need to consider the balance of real-time detection.This study focused on the following aspects: Analyzed the current finger vein feature extraction algorithms,Through simulation and compares their advantages and disadvantages;According to features of the human finger vein images, the algorithm of image feature extraction based on the Ridgelet transformation is presented in this paper;In the consideration of the accuracy of the Extraction-Algorithm, the real time ability of the algorithm is also taken into account. According to the absorption feature of our finger towards the near-infrared, Based on the construction of background image, this paper proposed an approach to obtain the first residual error image, get the second residual error image through adjusting the scale factors, and finally indirectly gain the features of the vein;To solve the misjudgment of noise and vein information in features extraction from low quality images, a novel method based on LWF (Linear weighting function) immune-clone algorithm is proposed in this paper. The method can helps to boost the growth of the vein information and suppress the interference of noise.;Absorbed the idea of the immune clonal algorithm simplified the immune clone method and applied to the finger vein recognition on-chip system.更多还原