Biometric Identification Based on Conjuctival Vasculature Pattern using Contourlet Transform

The performance of any biometric system depends on the reliable and robust feature extraction.  Biometric recognition refers to the use of distinctive anatomical (e.g., fingerprints, face, iris, palm) and behavioral (e.g., speech) characteristics for automatically recognizing individuals. Iris recognition was found to be most accurate bio-metrics technology. Apart from the red, green, and blue (RGB) format, we analyze significance of using HSV, Otsu’s multi thresholding is applied on V channel, K-means clustering is applied to merge over segmented region to get sclera mask. Pyramidal directional filtering approach (Contourlets) for feature extraction for ocular biometrics are proposed. In this paper, we pursue a “true†two-dimensional transform that can capture the intrinsic geometrical structure that is key in visual information. Our approach starts with a discrete domain construction. For Classification, linear discriminant analysis (LDA) is used.