Online Signature Verification Using the Information Set Based Models
Urvashi Choudhary, Sanjay Kumar Dhurandher, Vinesh Kumar, Isaac Woungang, Joel J. P. C. Rodrigues · 2018
This paper proposes a new online signature verification system based on fuzzy modelling which involves the modification of the fuzzy membership function with the help of structural parameters. By using these structural parameters, the signature of a given user can be easily verified. The proposed approach relies on the extraction of features from the sample data collected from users under different situations and time intervals. Two distinct methodologies have been suggested to obtain the structural parameters: the first method is based on the Shannon entropy functions coupled with an objective function defined in terms of error while the second is based on the interaction among the input fuzzy sets which is computed using s-norms. Both genuine and forged signatures were tested using the proposed techniques with encouraging results.