The Research On Iris Recognition Algorithm And System

With the rapid development of new information technologies, life is becoming more and more organized and convenient. At the same time, to prevent unauthorized access and abuse, new security systems and secure personal recognition are becoming the most challenging problems for modern society. Biometrics is an emerging field of technology which uses inherent physiological or behavioral characteristics to perform automated recognition of a person. Personal characteristics are universal, unique, permanent, collectible and measurable, so that they offer greater security and convenience than traditional methods of personal recognition. In some applications, biometrics can replace or supplement the existing technology. Biometrics has a great application potential in the security controlling for secured areas and the information. Now it is gaining increasingly extensive attention.Iris recognition is one of the most reliable biometrics technologies. The iris has highly complex and unique structure and provides many interlacing minute characteristics. Even the left and right irises of the same person or the irises from two twins also contain a lot of distinguishable features. Compared with other biometric features, the iris is more stable and reliable for identification. A typical iris recognition system includes four main modules: iris acquisition, image preprocessing, feature extraction and classification.Firstly, a set of non-contact iris capture equipment is designed. The special illumination and display can assist the users to focus their eyes on the proper position. Through image quality assessment we get the clearest iris image from the picture sequence obtained.Secondly, a novel algorithm for iris localization is proposed. This method can improve pupil location's speed and accuracy by means of detecting two pairs of chords. Then, the outer boundary is detected with a deformable circular template. By contrast with the usual methods, this algorithm reduces the blindness of the searching process, and enhances the speed.Finally, a novel iris feature extraction method based on the dissimilar division of two iris regions has been presented. By using multi-channel 2D Gabor filters, we get...