Searching for an Optimal Reference System for On-Line Signature Verification Based on (x, y) Alignment

Abstract

Abstract. The handwritten signature is the expression of the will and consent in daily operations such as banking transactions, access control, contracts, etc. However, since signing is a behavioral feature it is not invariant; we do not always sign at the same speed, in the same position or at the same orientation. In order to reduce the errors caused in verification by these differences between original signatures we have introduced a new concept of reference system for the (x, y) coordinates from experimental results. The basis of this concept lies on using global references (centre of mass and principal axes of inertia) instead of local references (initial point and initial angle) for a recognition system based on local parameters. The system is based on the hypothesis that signing is a feedback process, in which humans react to our own signature while writing it following patterns stored in our brain.

Automatic Signature Verification: The State of the Art

“…Igarza et al [130] used a left-to-right HMM for online signature verification and verified its superiority with respect to ergodic HMMs. The superiority of PCA and MCA for online signature verification with respect to DTW and Euclidean-based verification was also investigated by Igarza et al [131]. Jain et al [138] used a set of local parameters-describing both spatial and temporal information.…”

“…For this purpose, Fourier transform [146], [147], [379], mathematical morphology [115], and Gaussian functions [37], [138], [180] have been used. Signature normalization procedures using global reference systems (center of mass and principal axes of inertia) [131] and Fourier transform [7], [146], [147], [149], [203], [273] have been considered to standardize signatures in the domain of position, size, orientation, and time duration. A crucial preprocessing step, that strongly influences all the successive phases of signature verification, is segmentation.…”

An Efficient DTW Matching for On-line Signature Verification

“…Mahalonobis distance, distances statistics, membership function, neural networks (NN), hidden markov model (HMM), support vector machine (SVM), etc.) [6,7,8], and structural based methods (e.g. string matching, tree matching, structural description graph analysis (SDG), etc.)…”

Methods of Alignment for On-Line Signature Verification

“…Signatures would not keeping higher consistent for a long time since external and internal environment exchanging. Thus, it is important to effectively align the test signatures to references, and reduce the influences the inconsistency of the size, location and rotation angle on the performances of online handwritten signature verification [3][4][5][6][7]. During online signature verification process, the authenticity of test signature is evaluated by matching its features against those stored in knowledge base for given individual.…”