Author : Akku Anna Rajan and Vipin V

Personal Authentication using Finger Knuckle patterns has received substantial attention from researchers with prominent applications in biometrics and forensics. This paper explores the performance evaluation of texture based feature extraction techniques for personal authentication using minor finger knuckle patterns, which are formed on the finger surface joining distal phalanx and middle phalanx bones. The texture pattern produced by the finger knuckle blending is quite unique and has high capability to discriminate different individuals. Finger knuckle matching scheme is developed with key steps for region of interest segmentation, feature extraction and robust matching. Feature extraction techniques based on texture features, like Local binary patterns (LBP), Three patch local binary patterns (TLBP), 1-D log Gabor filter is evaluated and compared. The algorithms have been tested on database available from Hong Kong Polytechnic University of both male and female volunteers. The experimental results show that TLBP and 1-D log Gabor filter yields better accuracy with reduced computational time.

Keywords: Texture, Feature Extraction, Segmentation, Finger Knuckle, Matching

Article published in International Journal of Current Engineering and Technology, Vol.5, No.5 (Oct-2015)