I.J. Image, Graphics and Signal Processing, 2015, 4, 51-59 Published Online March 2015 in MECS (http://www.mecs-press.org/) DOI: 10.5815/ijigsp.2015.04.06 Copyright © 2015 MECS I.J. Image, Graphics and Signal Processing, 2015, 4, 51-59 Accuracy Improvement in Palmprint Authentication System Jyoti Malik 1 , Dhiraj Girdhar 2 1 National Institute of Technology, Kurukshetra, India, 2 Computer Associates, Bangalore, India E-mail: jyoti_reck@yahoo.com, girdhar.dhiraj@gmail.com Ratna Dahiya 3 , G. Sainarayanan 4 3 National Institute of Technology, Kurukshetra, India, HCL Technologies Pvt. Ltd, Chennai, India E-mail: ratna_dahiya@yahoo.co.in, sai.jgk@gmail.com Abstract—Biometric authentication has been emerged as a reliable means to control a person’s access to physical and virtual places. Despite the various efforts made on biometrics, accuracy of the authentication/identification is the main concern and it has to be completely investigated. The paper presents critical analysis of the matching score values in such a manner that system accuracy is increased. Min Max Threshold Range (MMTR) technique is proposed that provides two levels of authentication and increase in accuracy is observed. The methodology of increase in accuracy is observed on various feature extraction methods. Index Terms—Biometric system, palmprint, accuracy measurement, authentication. I. INTRODUCTION Biometric systems are being used for access-control, e- commerce and m-commerce activities and it is being considered as safe, secure and fast source for personal authentication. Biometric authentication system is dependent on various factors like cost, security, user acceptance, speed and accuracy etc. For identification/authentication, biometric system has to be evaluated on the parameter of accuracy because accurate authentication can prevent unauthorized access. A typical biometric system needs lots of volunteers for enrolment to make a large database. The various stages in biometric system like image acquisition, pre-processing etc. can affect system accuracy directly or indirectly. A biometric system is to be designed that can address various problems/factors affecting accuracy. The factors have to be resolved in such a manner that the accuracy of the system can be increased. Security is an important issue with the advancement in information technology. USA, UK and several other countries are using biometric passport to control access from country borders [1, 2]. If the system is not accurate, an innocent person can be doubted/questioned as intruder/impersonator. Using biometric systems for access control or online banking, highly accurate judgment of person is required otherwise it can lead to great loss in terms of money and security. Improvement and increase in accuracy is desired in biometric systems [3, 4]. The aim of this paper is to present various factors affecting accuracy and improvement in accuracy validated by experimental results on palmprint biometric system. Section II presents accuracy and the factors affecting accuracy. Section III describes the proposed accuracy improvement framework implemented on various feature extraction based palmprint biometric system and concluded in section IV. II. ACCURACY AND FACTORS AFFECTING ACCURACY OF A BIOMETRIC SYSTEM A. Accuracy of a biometric system In password and token based authentication system, perfect comparison of user input data with stored template (password/token value) is possible. However, biometric authentication systems decision making is affected at every stage by various factors like noise in biometric sensor, illumination, environmental conditions, type of biometric used, feature extraction method, matching algorithm etc. Accuracy of biometric system is measured in terms of image acquisition errors and image matching errors. Image matching errors are False match rate (FMR) and False non-match rate (FNMR). Image acquisition errors include Failure-to-enrol (FTE) and Failure-to-acquire (FTA). Accuracy can be defined in terms of FAR and FRR that considers both image matching and image acquisition errors.     2 / (%) (%) 100 (%) FRR FAR Accuracy    (1) where, FAR is the percentage of number of wrongly accepted individuals over the total number of wrong matching, FRR is the percentage of number of wrongly rejected individuals over the total number of correct matching. B. Factors affecting Accuracy There are several aspects that affect the accuracy in a