H.S. Fadewar Assistant Professor, School of Computational Sciences, S.R.T.M . University, Nanded-431606, M aharashtra, India. fadewarhsf@gmail.com Finger Vein Recognition System Based on Multi- algorithm of Fusion of Gabor Filter and Local Binary Pattern Abstract — Finger vein is another biometric innovation that contends with other ground-breaking biometrics modalities, for example, the face, palm print, fingerprint, iris, and voice. The system used for the finger vein recognition contains 5 phases that contain data acquisition, pre-processing, feature extraction, matching and decision. In any bio-metric system, feature extraction phase is significant. The multi-algorithms for feature extraction are used. This work implies three scenarios based on Two-Dimension Gabor Filter (2DGF), Local Binary Pattern (LBP), and fusion of 2DGF+LBP for feature extraction. The experimental outputs prove that the fusion of 2DGF and LBP obtained the optimal output with a high- performance accuracy of 94.94% which improved by 0.81 % and 0.88% respectively for 2DGF and LBP when presented individually. Keywords— Finger Vein; Biometrics; Genetic Algorithm; Feature Extraction; Gabor Filter; LBP; Correlation Coefficients; FAR, FRR. I. INT RODUCT ION The present innovation is developing every day, on the other hand, the security scheme additionally is increment identified with the advances. The system of biometric is extraordinary exploration work at present, which including numerous biometrics attributes, for example, (biological and behaviour) [1],[2]. Passwords, PINs, tokens, and smart cards are the traditional security system which is not significant for application on frameworks that need high security. The biometrics recognition process supplanting customary techniques by using physical traits or behavior traits of people that speak to an individual's personality and points of interest that are hard to copy, taken, and adulterated [3]. Accuracy, scale, and usability are the principal challenges confronting any biometric recognition system[4]. Many researchers proposed techniques and methods to improve accuracy, for example, using more than one trait or multi- algorithms for feature extraction and this is called a multimodal biometric system [5]. A biometric recognition system can be divided into two different types depending on the extracted features. Face, iris, palm print, fingerprint and gait are the first category is called extrinsic biometric features. The second category is intrinsic biometric features such as hand vein, finger vein and palm vein [6]. The intrinsic features are high secured than extrinsic features. For example, the intensity of light effects on the retina surface when features are extracting from iris [7]. The same thing is happening when the face recognition system is used. Illumination of a discrepancy, fashion of facial, stoppage in blood vein, and pose causes disfiguring of accuracy in face recognition. [8]. Table I contains various typical extrinsic and intrinsic features describing various advantages, disadvantages, and other attributes. In 2002, Kono [9]a medical researcher belonging to Japan formulated various methods for Finger vein recognition. Consequently, these methods have been applied in many Japanese cities, and recognition of finger vein systems was developed in various countries [10]. The vein- based identification or verification system has models for a biometric pattern for security and suitability for authentication of individuals. biometric finger vein recognition (FVR) system is depicted in Fig. It is difficult for the vein is difficult to replicate and forge, because it is a portion of intrinsic traits. Generally, using the trans- illumination method [11] the Finger veins will typically be detected with the help of Near-Infrared (NIR) light (700– 900 nm). Vein-based systems frequently make use of various anatomical traits like hand vein, finger vein, palm vein, or foot vein for personal identification/verification. The vein-based method also uses specific anatomical features such as the hand vein, the finger vein, the palm vein, or the foot vein for human identification/verification. Finger vein is suitable. The least is a vein in finger sensor, as well as that fingers, have numerous veins compared to the hand and palm [12], the finger vein trait is distinctive for identical twins and occurs for individuals [13]. Most substantially, during a lifetime, the finger vein pattern does Fig. 1. Framework for finger vein recognition [15]. Waleed Dahea Research Scholar, School of Computational Sciences, S.R.T.M. University, Nanded-431606, M aharashtra, India. Computer Science and Information System, Thamar University, Dhamar, Yemen . dahea.waleed@gmail.com not change. [14]. Proceedings of the Fourth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC) DVD Part Number:CFP20OSV-DVD; ISBN: 978-1-7281-5463-3 978-1-7281-5464-0/20/$31.00 ©2020 IEEE 403