The Off-line Signature Verification Based on Structural Similarity
Margarita N. Favorskaya, Baranov Roman · Frontiers in artificial intelligence and applications · 2014
The off-line signature verification is demanded in business and marketing, bank transactions, security control, and document authentication. The off-line signature verification is the difficult process in against of the on-line verification. In this paper, the extension of feature set including global, local, and special features for simple and cursive types of signatures is proposed. The global features are required to create a decision tree, which limits a field of the search. The pre-processing procedures remove noises and artifacts of scanning and luminance. The Structural Similarity Index Measure (SSIM) and detection of special points are based on a skeleton signature representation. SSIM shows the well results of verification and demonstrates low values of False Rejection Rate (FRR) and False Acceptance Rate (FAR) for the GPDS-100 signature corpus.