In the 21st century, it has become a trend that machines are replacing the man in many fields.
But, there are a lot many fields still untouched to machines. But the evolution of modern computers and the development of the branch of Artificial Intelligence, gives man a chance to make a machine that can really replace man from the field which, up till are considered to be the fields reliable on human intellectual power.
Signature is the characteristic of the particular person & hence used globally for identifying a person, validity of the documents signed, banking etc. Up till now, in banks where signature of a person is the basic code for transaction, the validity of the signature is generally checked by a man.
This is a project, which simulates the ability of a man to recognize a signature from the standard signature he has. We have tried to implement a system which recognizes the signature. We deal with the signature as an image which is scanned through scanner.
The image undergoes different normalization techniques and then we extract some features of it to be used as inputs to the fuzzy.
Signature of a person is an Important Biometric Attribute of a human being and is used for authorization purpose for decades. With a lot of computing power available with modern computers there is a vast scope to develop fast algorithms for signature recognition. There is a lot of research work is being conducted in this field.
Various approaches are possible for signature recognition with a lot of scope of research. In this project we deal with an Off-line signature recognition technique, where the signature is capture and presented to the user in the format of image only. We use various image processing techniques to extract the parameters of signatures and verify the signature based on these parameters.