Discrete Radon Transform Based Static Handwritten Signature Verification

Biometric security devices are now permeating all facets of modern society. Hand-written signature is widely used for authentication and identification of individual. The proposed algorithm uses automatic cropping system and Local Radon Transform. The method uses Radon Transform locally as feature extractor and Hidden Markov Model as classifier. To avoid interpersonal variation 3 signature images of the same person are taken and feature points are trained. These trained feature points are compared with the test signature images and based on a specific threshold, the signature is declared as original or forgery. When being trained using 3 genuine signatures of each person and 90 forgeries taken from our database, the proposed method obtained an equal error rate (EER) of 4.29%. The false acceptance rate (FAR) and false rejection rate (FRR) for proposed method was also kept as low as 5.00% and 4.44% respectively.