Criminalistic conclusions on signature forgery process during building an offline signature verification intellectual system

Abstract

The article describes details of the NSP-SigVer project, such as qualitative indicators of humans’ ability to identify a signature forgery, which can be used to build an offline signature verification system based on an artificial neural network. The average accuracy of this action is 69.29%. The article also provides a classifi cation of signature forgery and some features of the forgery process which are important for its identification.