Offline signature verification

Abstract

Signature images are not rich in texture however they have much vital geometrical information. The technique presented in this thesis harnesses the geometrical features of a signature image like center isolated points connected components etc and with the power of Artificial Neural Network classifier newlineclassifies the signature image based on their geometrical features. In todays era newlinedeep learning is the emerging field in case of feature extraction object detection classification etc. So rather than depending on the hand engineered features newlineor geometrical features we should adapt the techniques which will automatically newlineextract the relevant features. We have proposed a convolutional neural network based language independent shallow architecture sCNN Shallow Convolutional Neural Network for signature verification. The proposed architecture is very simple but extremely efficient in terms of accuracy.