Design And Implementation Of A High Resolution Iris Recognition System

Biometric identification is a burgeoning technology which authenticate personal identity using their unique physical or behavior characteristics as medium. Among the numerous biometric technologies, iris recognition gets most attention as it has the highest recognition rate over other methods within biometric identification technologies. At present, there are deficiencies in most of the existing iris recognition system which restricts the development of the iris recognition mostly. Based on the research with iris image acquisition and iris feature extraction algorithm of existing iris recognition system, we exploited and realized a high resolution iris recognition system with highly user friendly which is convenient, fast, high accuracy, low cost, low power consumption, etc.The main works and innovations are presented as follows:1. By studying on the difficulty of iris image acquisition in iris recognition system, we acquire high-resolution image using high resolution CMOS camera MT9J003 and USB3.0. Acquired image is showed real-timely on a LCD through a high performance FPGA, give a feeling of "what you seen is what you get!".The acquired high-resolution image will be transferred to computer using "Super Speed" USB3.0 simultaneously. Auto-focus module and measure distance circuit make the system more intelligent.2. Whether is there a qualified iris region or not is judged using the combining method of Adaboost and Circumference Radial Symmetry since heavy computation with high resolution image.3. With the further study of iris feature extraction method, we put forward an iris feature extraction algorithm based on Quaternion Orthogonal Filter. The method extracts iris feature using 2-D quadrature Log Gabor filter obtained by Quaternion Fourier transform. The method could get reliable iris phase information by orthogonal directions and calculate analytic signal of the filtered iris image as feature. The method characterizes iris feature in multidirectional ways, thus iris feature encoding with high complexity that makes the iris feature more comprehensive was obtained, improve the accuracy of the system. The experimental results show that this method has good recognition performance.4. A GUI runs on computer was programmed using Qt Creator. The GUI contains some basic functions of user registration, identification and management. The GUI also have the function of displaying captured iris image, even though the iris identification device has already a real-time display LCD, making iris identification device and computer can observe the captured iris image at the same times.