Please use this identifier to cite or link to this item: http://hdl.handle.net/10603/482016
Title: Accurate recognition of handwritten Punjabi characters using deep learning techniques
Researcher: Gurpartap Singh
Guide(s): Agrawal, Sunil and Sohi, Balwinder Singh
Keywords: Computer Vision
Convolutional Neural Networks
Deep Learning
Handwriting Recognition
Machine Learning
University: Panjab University
Completed Date: 2021
Abstract: Since the advent of Deep Learning the handwriting recognition accuracies have achieved new standards. With the newly established standards the main focus is now on collecting standard datasets which represent the general population efficiently. The work presented in this thesis is includes an autonomous methodology for handwritten data collection. For data collection three algorithms have been proposed each of which uses a universal format for handwritten data collection. The collected data consists of Digits, Alphabets and Words of Punjabi Language. These datasets have been statistically proven to be more competitive than existing datasets. For recognition, modifications in Deep Learning algorithms have been proposed which gives a close to accurate recognition accuracy. The proposed methodology is also found to outperform other recent state-of-the-art techniques reported in literature. newline
Pagination: xiii, 145p.
URI: http://hdl.handle.net/10603/482016
Appears in Departments:University Institute of Engineering and Technology

Show full item record


Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).

Altmetric Badge: