The main aim of proposed system is to increase the security in banking environment using signature verification. For some real world personal identification, signature can be used for personal identification. It is used for authentication or concluding document. In order to reduce frauds in banks, signature verification is very much important. It increases accuracy and efficiency. Various methods are used for classification such as hidden markov model, support vector machine, neural network and contour method. In the proposed system, support vector machine and hamming distance measurement are used for signature verification. Verification can be offline or online. In proposed system, we used offline signature verification. Offline signature method is based on scanned image of signature. The proposed method consist of various steps such as pre-processing, feature extraction and comparison between stored and input signature.