Font Size: a A A

Research And Realization On Key Technology Of Iris Recognition

Posted on:2011-05-21Degree:MasterType:Thesis
Country:ChinaCandidate:X ZhouFull Text:PDF
GTID:2248330338996309Subject:Aviation Aerospace Manufacturing Engineering
Abstract/Summary:PDF Full Text Request
With the high-speed development in economics and the continuing progress of science, traditional identification methods such as password, identification cards cannot satisfy the higher security request in modern society for the shortcomings like low-security and easily-copying. Biometrics identification technology provides a kind of more secure and accurate way for identification. Among them, Iris recognition has become the focus in information security field for the advantage of its high security and stability. The main contents of the paper are as follows:1. According to the request of iris recognition, the iris image gathering equipment consists of industrial camera, fixed-focus lens and infrared LED which being used as lighting device. The Iris images taken from this equipment have clear texture and proper size, meanwhile, they can be used to set standard iris database.2. Traditional iris localization algorithm is improved. Firstly, the image is processed through binarization, and centroid detecting method is used to locate the center of Iris pupil. Secondly, employ least square method to fit the pupil edges. Finally, Canny operator is used to detect the outer edges.3. Analyze the character of 2D-Gabor filter on location, scale and frequency. A set of Gabor filters with different scales are designed, which can well extract the overall and partial character of Iris textures. According to phrase information around sample points, unique iris code from different people is generated.4. Ellipse Gaussian filter with proper direction is designed based on the scale changing law of iris texture in order to extract the region grayscale near sampling points to generate code sequence. The hamming distance from different code sequence is calculated to make criteria for classification.5. Experiment results on the CASIA, Bath and NUAA-IDB iris database show that the proposed is efficient and robust. On the basis of the recognition algorithm, software has been designed, and is connected with Iris-gathering device to complete the whole Iris recognition system.
Keywords/Search Tags:Iris recognition, Iris acquisition, Image preprocessing, Feature extraction, Code matching, Hamming distance
PDF Full Text Request
Related items