OFFLINE SIGNATURE FORGERY DETECTION USING CONVOLUTIONAL NEURAL NETWORK

Handwritten Signature is considered as one of an integral part of security as it can be used for verification and authentication. Precision is not maintained every time a person does the signature, different parameters like signature strokes, length, pixel depth, continuity, etc may vary. Such Properties of the signature has to be checked before verification and authentication. So authenticating a fake signature becomes a challenging task. A Signature Capturing and Recognition System will take the image of the signature as an input and will train the image by extracting various features and will store it in the database then using Convolutional Neural Networks it will be compared with the original source signature and recognize whether it is the original signature. For feature extraction algorithms like Grayscale and Binarization are used. Once the image is captured, it will be converted into a black and white image and then processed. This system needs to be trained very well in order to have better results. Signatures samples will be fed into the system for identification tests in order to maintain high accuracy in the system. Feature extraction is an important stage where the features of each signature are captured using the CNN algorithm. The idea of this step is to identify each and every minor detail of a signature. Subsequently identifying the features and extracting them properly will lead to a better or more accurate verification. A centralized database of correct signatures of the customers will be available. This particular database can be used by a lot of systems that require customer information and signature information.