节点文献
基于DSP的虹膜识别系统研究
The Research of Iris Recognition System Based on DSP
【作者】 付云峰;
【导师】 赵国良;
【作者基本信息】 哈尔滨工程大学 , 控制理论与控制工程, 2005, 硕士
【摘要】 基于虹膜的生物识别技术是一种最新的鉴别技术,通过一定的虹膜识别算法后,可以达到十分优异的准确度。虹膜识别技术的高精度使它能够在众多识别技术中占有一席之地。由于其本身的高精度性、广阔的使用前景和很好的经济效益,使得它越来越受到国内外有关研究人员的重视。随着虹膜识别技术的发展,它的应用领域越来越宽,不仅在高度机密场所得到应用,并将逐步推广应用到机场、银行、金融、公安、出入境口岸、安全、网络、电子商务等场合。但由于虹膜识别系统的高成本,已经成为制约它发展的重要瓶颈。 首先,本文开发了一种低成本的虹膜采集装置。该装置采用低廉CCD摄像机,配以可见和红外发光二极管组成。由此装置获得的虹膜图像,纹理清晰,图像清楚,达到了虹膜识别的图像要求。 其次,完成了基于DSP虹膜识别部分的硬件的开发调试和相应软件的开发调试。在硬件上完成了视频输入、视频缓冲、RS232、USB等硬件接口的测试。在软件上完成了I~2C总线的模拟、两通道EDMA的数据传输、CPLD的逻辑控制、DSP自举引导、USB的固件程序和驱动程序、中断服务程序等相应软件的编制工作。 最后,在虹膜识别系统上进行虹膜图像质量的评价工作。这里主要针对一种客观的虹膜图像质量评价方法进行了研究。并对它所提的“图像清晰度”、“内外偏心度”以及“虹膜可见度”三个指标进行了分析。最后,将上述三个指标综合为一个“虹膜图像综合质量指标”。本文结合实际的虹膜识别系统,给出了相应的实验结果。并证明了该“虹膜图像综合质量指标”可以很好的反应虹膜图像的真实质量水平。
【Abstract】 The iris recognition is a new biometrics recognition technology. It can reach the precision veracity. The high veracity of iris recognition is the special feature in numerous biometrics recognition technologies. It is paid attention to by more and more domestic and international relevant researchers because of high veracity and good economic benefits of iris recognition. It can not only be used in secret places, but also in the airport, bank, finance, police, entry and exit ports, security, network, e-business. The barrier for iris recognition in practice is its high cost.At first, the paper designed a cheap collection device for iris image. The device consists of a cheap CCD camera, visible light leds and infrared leds. The iris image in the CCD camera is very clear so that the demand of iris recognition is satisfied.Secondly, the paper developed the hardware and software of iris recognition device based on the DSP. The test of video-in , video-out, asynchronous serial interface(RS232), universal serial bus(USB) interface and others are completed on printed board. The software of I2C bus simulation, data transmission between two channels of EDMA, the logic control with CPLD, the booting and guiding of DSP, the interrupt program and others are tested.At last, the appraisal of iris image quality on the iris recognition system is discussed in the paper. A kind of evaluation method for iris image quality is introduced. And, the indexes of image clearity, eccentricity of inner and external boundaries and the iris visibility are discussed. Finally, the three indexes are synthesized into a comprehensive quality index for iris image. The experimental results are given. It shows that the true quality level of iris images can be represented by the comprehensive quality index of the iris picture exactly.
【Key words】 iris images; iris recognition; DSP; images quality evaluation;
- 【网络出版投稿人】 哈尔滨工程大学 【网络出版年期】2005年 08期
- 【分类号】TP391.4
- 【被引频次】9
- 【下载频次】438