Georgia Southern University
Reduced Dataset Neural Network Model for Manuscript Character Recognition
Abstract
dc:description.abstractThe automatic character recognition task has been of practical interest for a long time. Nowadays, there are well-established technologies and software to perform character recognition accurately from scanned documents. Although handwritten character recognition from the manuscript image is challenging, the advancement of modern machine learning techniques makes it astonishingly manageable. The problem of accurately recognizing handwritten character remains of high practical interest since a large number of manuscripts are currently not digitized, and hence inaccessible to the public. We create our repository of the datasets by cropping each letter image manually from the manuscript images. The availability of datasets is the major obstacle in our experiment. However, we tackle the problem by using resampling and convolutional techniques for performing character recognition from manuscript images in our neural network model with this reduced training dataset. The experimental result shows that our proposed model outperforms the previous work.
Degree
thesis:*- Name thesis:degree_name
- Master of Science in Mathematics (M.S.)
- Level thesis:degree_level
- Thesis (open access)
- Discipline thesis:degree_discipline
- Department of Mathematical Sciences
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Islam, Mohammad Anwarul
- Contributors dc:contributor
-
- Goran Lesaja
- Felix Hamza-Lup
Subjects
dc:subject × 6Rights
- Licence dc:rights.license
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://digitalcommons.georgiasouthern.edu/etd/2138