【作者】 高贵朋；

【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】 现代生物识别技术是指通过计算机利用人类自身生理或行为特征进行身份认定的一种技术。而虹膜的身份识别技术是一种最新的鉴别技术。虹膜作为重要的身份鉴别特征,具有唯一性、稳定性、可采集性、非侵犯性等优点,与其他的生物识别方法相比,虹膜具有更高的可靠性、准确性,所以被认为是最具前途的一种生物识别技术。本文对虹膜识别算法做了深入的研究,希望能够将它运用到保密机关、机场、自动取款机等等需要它的部门。 首先,本文简单介绍了几种典型的生物识别技术的发展、详细地阐述了虹膜识别技术的发展及研究意义、虹膜生理结构及虹膜识别系统的组成。之后较深入地研究了虹膜识别算法,并实现了其各个步骤,即定位、预处理、特征提取和匹配。 在定位方面,提出了一种新的定位算法,即利用灰度阈值分割瞳孔结合边界跟踪定位虹膜内边缘,利用Canny算子和Hough变换方法定位虹膜外边缘。 在预处理方面,实现直角坐标到极坐标的转换,使虹膜识别具有伸缩和平移不变性。同时利用直方图均衡化的方法消除了由于光照不均所造成的影响。 在虹膜纹理的特征提取方面,提出了采用Gabor变换和二维小波变换对虹膜纹理进行编码的两种方法,同时利用方差倒数加权欧式距离分离器来识别,取得了很好的识别效果。 最后,将本文的成果及结论加以总结,并对今后的研究工作提出自己的几点期望。更多还原

【Abstract】 The modern technology of biometrics recognition could identify person by using powerful computer based on the biologic feature or behavior of human being. The personal identification system based on iris is a new personal identification technology. As a personal identification technology, iris recognition has some advantages, such as uniqueness, stability, ease of acquisition and no infringing. Comparing with the recognition of face, sound and others, iris recognition has higher accuracy so that it has more booming perspective comparing with other numerous biometrics recognition technologies. The iris recognition is studied deeply in this thesis.In this thesis, the history of biometric recognition and some kinds of biometric recognition techniques are introduced. Then, the development of iris recognition technique and the structure of iris recognition system are discussed in detail. The algorithms of iris recognition have also been studied deeply and all of the processes that include location, pre-processing, feature extraction, and pattern match have been carried out.In the iris location respect, a new location method is put forward which can enhance the speed of location. The threshold of grey and edge tracing are used to get the centre and radius of inner circle of iris. And, the Hough transform and Canny operators are used to extract that of outer circle of iris. Testing results have shown that the proposed method can improved iris location and is not sensitive to noise. In order to remove the effects caused by displacement and zoom of iris image, iris image is transformed from Cartesian coordinates to polar coordinates. For feature extraction, two kinds of algorithms are considered. One of them is Gabor encoding method and the other is Wavelet transform. For iris recognition, the weighted Euclidean distance is used. Experiments show that the method is valuable in iris recognition.更多还原