Offline Handwritten Signature Recognition Using Deep Convolution Neural Network

In the modern age, technological advancement reached a new limit where authentication plays vital role in security management. Biometric-based is most referenced procedure for signature verification significant part of it person. To prevent falsification signatures on important documents & legal transactions necessary to recognize person's accurately. This paper focused recognizing offline handwritten original forged using deep convolution neural network. We use completely dataset also downloaded datasets train system verify random as genuine or forgery. All testing samples are collected from several individuals after steps preprocessing model fed with resultant image our system, experimental results give us an accuracy 95.5% dataset.