Study On Identification Recognition Based On Iris

Biometric recognition technique based on the human's behavioral or inherent characteristic is applied to distinguish one's identification. Comparing to traditional methods, the biometric recognition can solve those problems of bearing obliviousness, inconvenience, losing and it is hard to be faked, so it is highly reliable and safe.Among various external features of human body such as fingerprint, face, eye and iris, iris that regards as an important characteristic has the advantage of uniqueness, stability, high reliability and non-invasion, which provides the iris identification with excelsior accuracy. Iris identification technology is regarded as one of the most promising biometric recognition technology.In this paper, the development of iris recognition technique and the structure of iris recognition system are discussed in detail. Then all of the processes that included image preprocessing, feature extraction and coding, matching and recognition are introduced. Feature extraction and coding have been focused on researching.In image preprocessing of iris, two steps of coarse location and exact location is used to locate the center and radii with the inner and outer circle of iris. Binary process is used to coarse location of the inner circle of iris. According to the principal that a circle can be confirmed by three points which are not colinearity, the outer circle of iris is coarse located. Hough transformation is used to the exact location of iris. In feature extraction and coding matching of image, first, two-dimension normalized iris image is transformed to a special 1-D intensity signals to decrease the quantity of calculation. Then wavelet which function is a quadratic spline of a finite support is adopted to represent the resulting 1-D intensity signals. The results of suitability scales are transformed to binary to record as features, so the complexity is reduced and the recognize speed is enhanced. Finally, Iris matching is accomplished by Hamming distance. It can achieve right recognize when the iris images are circumvolved.In this paper, massive confirmation experiments approved that the algorithm is feasible.