Nirpjit Kaur and Nirvair Neeru, International Journal of Advanced Computronics and Management Studies (IJACMS), Volume 1,Issue 5, August, 2016, pp26-33,ISSN:2456-1835 26 Bimodal Biometric System Based On SIFT With GA for Recognition of Hand and Palm Nirpjit Kaur M.Tech Scholar CE department Punjabi University, Patiala, Punjab, India sandhunirpjit@gmail.com Nirvair Neeru Assistant Professor CE Department Punjabi University, Patiala, Punjab, India nirvair_neeru@yahoo.com Abstract: Automated biometric systems are most popular to characterise a person for identification and verification purposes because of its robustness. Among biometric characteristics like face, fingerprint, iris etc Hand geometry is well known used because of its simplicity. In this paper, we propose an approach which combines hand geometry and palm print characteristics and Scale Invariant Feature Transform (SIFT) is used to extract features. To improve the performance optimization is performed using genetic algorithm which reduces the complexity as by this e can find more accurate feature dataset . In proposed work features extracted from scale invariant feature transform are invariant to scale are passed to genetic algorithm for optimization. The optimized features from both the hand and palm print are then classified individually using one of the distance classifier called hamming distance. Finally both the modalities are fused together for identification purpose. Experimental results are improved as compare to previous proposed work. Experiments have been taken under MATLAB software toolbox. Keywords: Hand Biometrics, palm print, scale invariant feature transform, genetic algorithm, fusion information 1. Introduction Biometric is a measurement of life and life is related to our characteristics which can be use for measurement. Biometric is a branch of digital image processing which deals with the processing of digital images. Biometric is a class divided into categories based on its characteristics like our physical appearance or it can be our behaviour. Generally biometrics is classified as physiological characteristics and behavioural characteristics. Former studies the life based on physical appearance like face, iris, fingerprint, hand, palm, retina etc measurements are taken from physical traits while latter studies the life based on behaviour like gait, keystroke, signature, voice etc measurements are taken form behavioural traits. Illustration of biometric characteristics is shown in figure 1 below. Figure1: Biometric Characterization The remainder of the paper is organised in following sections. Section 2 covers the related work done by other researchers and their evaluations. Section 3 describes methodology to be followed in the proposed approach and the algorithms which are used are explained in steps. Section 4 exhibits the experimental