Signature Verification

Handwriting recognition is a process in recognizing handwritten letter images. For this
project, the main purpose is to identify signature's owners to prevent from skilled forger.
Therefore, the project is more focused on the signature verification rather than the
character recognition. Signature verification can prevent falsification by detecting the
flow ofthe curve ofthe signature and using the distance similarity. The main objective of
this project is to verify the signature. The secondary objectives are to make life easier by
having the signature verification system to identify the skilled forger from using someone
else's credit card and to increase the security measure on credit card. The method that
will be used in this project is using the neural network classification in the
backpropagation network. This is because backpropagation can get the input to give the
correct output, which has been used by many researchers. In implementing the prototype,
a distance measure is used as the verification method but backpropagation is one of the
suitable methods in designing it for future expansion. Matlab software is used in
developing the system. Basically, this system will be using the Matlab software and for
the hardware part by using the digitized tabiet with the pen tip in order to capture the user
signature image. Beside that, some calculations will be used in measuring the signature
attributes andas for the error part; there will bethepercentage ofthe error occurs.