IJCSNS International Journal of Computer Science and Network Security, VOL.9 No.5, May 2009 38 Manuscript received May 5, 2009 Manuscript revised May 20, 2009 Online Multi-Parameter 3D Signature Verification through Curve Fitting P. M. Rubesh Anand 1 , Gaurav Bajpai 2 and Vidhyacharan Bhaskar 1 1 Department of Electronics and Communication Engineering, SRM University, Kattankulathur – 603203, India 2 Faculty of Engineering, Kigali Institute of Science and Technology, B.P. 3900, Kigali, Rwanda Summary Biometric based personal identification is a reliable and widely accepted method for authentication. Human 3D signature is distinct from other biometric authentication methods due to the presence of third dimensional hidden information. In this paper, a new model for on-line 3D signature verification using multiple parameters is proposed. The parameters considered from human signature namely, velocity, acceleration, pressure, direction, pen ups/downs, total time taken, length and depth of the signature are unique for each person. The proposed model highlights the third dimensional distinctive parameter, depth of the signature for its hidden information. Curve fitting is performed on the points obtained from different non-linearly spaced layers of the signature pad. The best fitted curves from all the layers and other signature parameters are used in the process of verification of 3D signature. The digital multi-parameters of the 3D signature are further encrypted with cryptographic algorithms to protect from cryptanalysis. The attempts for 3D signature expert forgery by satisfying both the global and local parameters of the signature are difficult. The application of the 3D signature verification broadly ranges from authentication of financial transactions to authorization of administrative documents. Key words: 3D Signature verification, Authentication, Biometric Security, Curve fitting, Forgery prevention 1. Introduction The rapid growth of internet has lead to numerous on-line business transactions and administrative works through computers. The need to ensure that only the right person gets access to the highly secured information, the requirement for reliably effective security methods to protect the information transferred through insecure channel leads to various authentication mechanisms [1], [2]. The term biometrics refers to individual recognition based on person’s unique characteristics. In the biometric techniques, the individual is identified by his/her physiological or behavioural characteristics. The physiological identification is based on the biological individuality of users, like, fingerprint, face, hand geometry, vein patterns, retina and iris. The behavioural biometric identification considers voice or handwritten signature [3]. Although physiological biometrics have consequently become more integrated into commercial products, behavioural biometrics exhibit the quality of memory that make them attractive for security applications. The techniques used for authentication in computer systems, like, token based, knowledge based identity verification requires the possession of token, remembering of the password/pass phrase are prone to be forgotten, disclosed or compromised. In contrast to the knowledge or token based verification techniques, the biometric based identification/verification offers the advantage of presenting the individual personality, whose attributes are hard to steal or forge [4]. Human hand written signature is used as a traditional way of authentication in business and financial transactions due to its unique nature of individuality. The static and dynamic signature verification for the paper-based documents and transactions are done by humans. The challenges faced in that verification are: any signature can be learnt; it can be changed by the owner and has several versions of the signature depending on the level of importance or intent of the signer [5]. Most humans are a relatively poor judge of handwritten signature authenticity leading to the success of the expert forgers. Presently, the writing pad with dedicated pen for 2D handwriting recognition is available for email signing and handwriting recognition [6], [7]. The 2D signature verification methods are vulnerable to spoof [8], [9]. The main reason for the failure is due to the fact that signatures are verified in 2D. This paper proposes a new model for reliable and accurate identification/verification of 3D hand written signature by considering the depth parameter in different layers of z- axis in the signature pad to enhance secure transactions. Signature verification has a number of statistical features that can be derived from the basic set of data from the signature pad. This paper uses only the parameters which have more uniqueness like velocity, acceleration, pressure, direction, pen ups/downs, total time taken, length of the signature. Apart from these parameters, the hidden parameters like depth of the signature, curve fitting, surface fitting and solid angle are calculated by the mathematical functions. The application of this proposed