NORMA eResearch @NCI Library

Gujarati Handwritten Character Recognition using Convolution Neural Network

Sorathiya, Dharavi Rasikbhai (2021) Gujarati Handwritten Character Recognition using Convolution Neural Network. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (866kB) | Preview
[thumbnail of Configuration manual]
PDF (Configuration manual)
Download (1MB) | Preview


Language is important in many facets of life as it enables people to communicate in a manner that permits them to share similar views. With 22 languages 'registered' in the Indian Constitution and an approximated 780 linguistics in daily usage, India boasts a broad list of spoken languages. One of the languages is Gujarati, which is used primarily by Gujaratis and belongs to the Gujarat state of India. When it comes to character recognition, Gujarati script is among the languages for which there is limited literature is available. This paper focuses on creating a Gujarati script recognition algorithm that can recognize handwritten Gujarati characters accurately from images. Given the scarcity of data for the Gujarati language, a data collection of around 30,000 images featuring writings from various people is an essential contribution to this research. This procedure contains building a neural network that accepts a picture as an input, extracts a feature from it, then identifies the words and outputs them on the system. Three distinct models are used to train the model named Custom CNN, Xception, Inception_v3. All the implemented models have been evaluated using evaluation metrics such as accuracy, loss, recall, precision, f1-measure, and computation period. With an accuracy of 97%, 81 minutes of computation time for 15 epochs, InceptionV3 outperformed all models with regards to accuracy.

Item Type: Thesis (Masters)
Subjects: P Language and Literature > PK Indo-Iranian
Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
P Language and Literature > P Philology. Linguistics > Computational linguistics. Natural language processing
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Tamara Malone
Date Deposited: 13 Mar 2023 13:15
Last Modified: 13 Mar 2023 13:15

Actions (login required)

View Item View Item