ABSTRACT
Handwritten signature is an important biometric attribute of a human being and the most natural method of authenticating a person’s identity. Verification of handwritten signature can be performed either off-line or online based on application. In this paper, offline handwritten signature recognition and verification using neural network is projected, where the signature is written on a paper and is obtained using a scanner and presented in an image format. There are various approaches to signature recognition with a lot of scope of research. The method presented in this paper consists of global feature extraction, neural network training with extracted features and verification. A verification stage includes applying the extracted features of test signature to a trained neural network. Signatures are verified based on parameters extracted from the signature using various global features. The method is implemented using MATLAB.
