Writer-Independent Offline Signature Verification using LBP and NN
Ashok Kumar, Rahul Rastogi · 2022 International Conference on Machine Learning, Big Data, Cloud and Parallel Computing (COM-IT-CON) · 2022
Verification of a signature is a problematic research topic since two or more people’s sign may appear to be identical, but a person’s sign may differ depending on the state. The purpose of this research is to examine how effectively an Artificial-Neural-Network and a Local-Binary-Pattern feature set can be combined to construct a Writer-Independent-Offline-Signature verification arrangement. The system’s performance is assessed using two signature datasets, each with 260 and 100 writers. Authentic signatures of a person, as well as skilled-forgery, nonskilled-forgery, and random-forgery signs, are used to test the performance of the developed system, and authentic signatures, as well as skilled-forgery, nonskilled-forgery, and random-forgery signs, are taken into account in the development of the desired system. In this study, a false-acceptance-rate of 21.00 percent, 11.00 percent, and 1.00 percent was obtained for skilled-forgery signs, nonskilled-forgery signs, and random-forgery signs, respectively, while a false-rejection-rate of 0.00 percent was obtained for 11 reference signatures using a database of 260 writer’s signatures.