Volume 20 No 12 (2022)
 Download PDF
Offline Writer Identification using Convolutional Recurrent Neural Network
Deepa Bendigeri, Gopal A. Bidkar, Jagadeesh D.Pujari
Writer identification can be used to find fakes and help with forensic science. Handwritten text is the only way to identify the author offline. Recently, convolutional neural networks (CNNs) have become the best way to classify images on a large scale. The recurrent neural network (RNN) models the spatial relationship between the fragment sequences in order to improve the local fragment feature's capacity for discrimination. To get the best of both models, combine CNN and RNN to create a Convolution Recurrent Neural Network (CRNN). Therefore, the CRNN model for offline writer identification is suggested in this paper. The proposed method achieved 96% accuracy in classifying 690 writers using CRNN. The suggested method can make efficient and strong writer identification based on different sizes, different orientations, or both. The efficiency of the proposed system has been demonstrated. Finally, an accuracy comparison of CRNN with CNN and RNN has been conducted
Writer identification, Offline analysis, Convolution Neural Networks, Deep learning, Recurrent Neural Network, Convolutional Recurrent Neural Network, Long short-term memory
Copyright © Neuroquantology

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Articles published in the Neuroquantology are available under Creative Commons Attribution Non-Commercial No Derivatives Licence (CC BY-NC-ND 4.0). Authors retain copyright in their work and grant IJECSE right of first publication under CC BY-NC-ND 4.0. Users have the right to read, download, copy, distribute, print, search, or link to the full texts of articles in this journal, and to use them for any other lawful purpose.