Biometrics and Bioinformatics

Kshamaraj P. Gulmire ¹, Sanjay R. Ganorkar ¹

Affiliations
1 Department of ETC Engineering, Sinhgad College of Engineering, M.S, Pune, IN

Abstract

The iris recognition is a kind of the biometrics technologies based on the physiological characteristics of human body, compared with the feature recognition based on the fingerprint, palm-print, face and sound etc, the iris has some advantages such as uniqueness, stability, high recognition rate, and non-infringing etc. The iris recognition is using principal component analysis can produce spatially global features. In this paper we use the feature extraction algorithm based on PCA fo2r a compact iris code. Here for segmentation of iris image is based on Daugman’s method using integro differential operator finally, sorts the different iris patterns by improved Euclidean distance method and gives the recognition results.

We use PCA method to generate optimal basis elements which could represent iris signals efficiently. In practice the coefficient of this method are used as feature vectors. Then iris feature vectors are encoded into the iris code for storing and comparing individual's iris patterns.

Keywords

Biometric Recognition, Iris Recognition, PCA.

Full Text

     

Mayuri M. Memane ¹, Sanjay R. Ganorkar ¹

Affiliations
1 Department Electronic & Telecommunication, Sinhgad College of Engineering, Pune, MH, IN

Abstract

The first and most popular iris recognition algorithm was introduced by pioneer Dr. J. Daugman which is not available for open use, so an alternative for research purposes can be found in the implementation created by Ives et al. to perform iris recognition, referred as Ridge Energy Direction (RED) algorithm that will be the focus of this work. The first step is to collect the database of iris, than carry out the various preprocessing steps which includes conversion of color image to gray scale image, histogram equalization, intensity adjustment and segmentation i.e. separation of pupil from the image and select the area of interest. Features are extracted from the area of interest using two directional filters (i.e. vertically and horizontally oriented) which generates two individual templates. The final template is generated by comparing the results of two different directional filters and writing a single bit that represents the filter with the highest output at the equivalent location. This final template is compared with the stored one using Hamming distance. If the template is match with the stored one than the match ID number of the person is displayed and non-match image is indicated by displaying zero.

Keywords

Biometrics, Iris Recognition, Hamming Distance, Histogram, RED Algorithm, Segmentation.

Full Text