Abstract:
Forensic handwriting analysis is the process of using scientific methods to determine the origin of documents. From broad domain of forensic handwriting analysis, this project will focus on offline signature verification. Aim of this project is to develop a system that can differentiate between genuine and forged signatures, and treat disguised signatures as genuine, by extracting global and local features from signature using concept of stability. Aim is to use pattern recognition techniques of automatic signature verification to meet the requirements of forensic science. For this various techniques were developed and tested on datasets collected by forensic handwriting examiners.
