Writer identification is the domain of documents image analysis which is popularly sound in many applications like banking, academic professional in Optical Character Recognition (OCR) Signature verification remains one of the most important entities to authenticate document in these applications. In view of technical breakthrough, we have focused on short words signatures which are very hard to verify as they raise issues of ambiguities. From the geometrical studies of signature-based images, it is stated that the morphology of directional transformations (MDT) is right to extract the suitable features in case of short words for writer identification. MDT takes the data in the form of Structure Element (SE). The directional morphological structures (DMS) of SE are used as a key factor for performing morphological operations on the signature images. We have adopted Markov chains and Fisher Linear Discriminant (FLD) for computing the gradients from line features corresponding to the word. Neural network is used to evaluate the proposed model. In case of training and testing of the signature, images of short words leave-one-out are followed. Our purposed model is tested on NIST database for extracting the words length with three letters. It is observed that a very simple architecture of neural network achieved 100% satisfactory results using short words.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.