节点文献
手指静脉身份识别智能系统的研究
The Study of Finger Vein Recognition Intelligient System
【作者】 杨数强;
【导师】 余成波;
【作者基本信息】 重庆理工大学 , 信号与信息处理, 2009, 硕士
【摘要】 现代社会中,随着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了可能。目前,指纹识别技术相对来说是生物特征识别中比较成熟的一种身份确认技术,但由于指纹易于伪造和破环以及被污染等致命缺点,使其应用范围受到限制。[0]而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,可有效地解决传统的生物特征识别技术所面临的诸多难题。本文通过分析手指静脉的生物识别系统所面临的人体手指差异和多变的问题,提出了一些针对性的解决方案并付诸实施,构建了由前端采集装置和基于PC的主机系统所构成的手指静脉身份识别智能系统。重点针对手指静脉采集装置中自动信号处理、智能参数调整以及主机系统中Matlab与VB接口的关键技术进行了一系列研究,研究工作的主要内容集中在如下几个方面:(1)根据人体手指及静脉的特征选择了适合与手指静脉成像的红外光源,确定了一套针对于手指静脉采集的速度、性能、成本、功耗等综合指标比较理想的光源控制方案。(2)根据手指静脉成像的特点,结合性能、成本、市场供求等,选择了适合于采集手指静脉图像的红外相机和红外滤光片。(3)分析了由于个体之间差异、个体本身不同状态下手指静脉差异及每次图像采集时手指放置位置差异等手指静脉身份识别系统中不可避免的问题,有针对性地逐个提出方案进行解决。提出了基于多传感器融合的手指定位方法和基于模糊控制理论的采集参数自动调整智能算法,实现了手指准确定位和采集参数的自整定。(4)为了提高系统的便携性和稳定性,针对主机系统与前端采集装置的通信接口,提出了基于USB的图像传输技术和虚拟串口控制方法,大大提高了系统的集成度。利用模块化设计思想进行系统软硬件的设计,既有效地缩短了系统设计实现的周期,又大大增强了系统设计的鲁棒性。(5)研究了Windows环境下Matlab与VB的接口技术,确定了基于COM组件的DLL接口技术作为智能系统的接口方案。利用模块化设计思想实现主机系统主机人机界面的设计、各部分软件模块的无缝连接及调度,最终构建了一套适合手指静脉身份识别的智能系统。
【Abstract】 In modern society, with the rapid development of computer and network technology, the information security becomes more and more important. People expect a more convenient and reliable method for identify authentication. The biometrics makes this possible. Currently, fingerprint recognition is considered as the comparative maturity technology in biological feature recognition. However, its fatal defect is easy to forge, destroy or contaminate, which would restrict its application. Finger vein conceal body interior, which has many merits such as fast, higher precision, high level of security, living body recognition, interior characteristic, non-contacting and so on, which can effectively resolve the many problems in the traditional biometrics ( e.g. Fingerprint ).In this paper, the problem of human finger differences and easy to change was analyzed for biological recognition system based on finger vein, a number of targeted solutions were made and the implementation was given. Finally, we constructed a smart finger vein identification system that consisted of the front-end collection device and the host PC system. We did a series of studies mainly focused on automatic processing and intelligent parameter adjustment about this finger vein collection device, as well as the interface between Matlab and VB in host PC system. Main works are as followings:(1) According to characteristics of human fingers and veins as well as the devices that easier to buy, we selected a suitable infrared light source for finger vein imaging, then identified a set of finger vein collection light source control way considered the parameters including collection speed, performance, cost, power consumption and so on.(2) According to the characteristics of a finger vein image, combined to performance, cost and market, infrared camera and infrared filter were selected based on finger vein image and acquisition.(3) By through analyzing several inevitable problems such as individual differences, finger vein differences under different states and differences in the location of finger placement in the process of acquisition, we proposed target solutions one by one. Moreover, we designed an intelligent fuzzy control algorithm based on multi-sensor finger position and acquisition parameters automatically adjust, achieving detection of the finger in place information automatically of the front-end acquisition device, accurately positioning finger, automatically adjust the finger vein image acquisition parameters for image acquisition.(4) In order to improve the portability and stability of the system, we proposed and implemented the method that, in the part of communication interface between host PC system and front-end collection device, the USB-based image transmission and the virtual serial port control, achieving portability and hot-swappable connection, greatly enhance the system integration. Using modular design thought of hardware and software for system design, not only fast the system design and implementation cycle, but also enhance the stability of the system greatly.(5) We carefully studied Matlab interface technology within VB in the Windows environment and ultimately determine to use COM-based DLL interface technology as a system call interface way. Using modular design thought to achieve host system human-machine interface design, various parts of software modules to connect seamlessly and scheduling, eventually build a set of intelligent finger vein identification system.