Feature Extraction and Selection in Batak Toba Handwritten Text Recognition

Pasaribu, Novie Theresia Br. and Hasugian, Meilan Jimmy (2020) Feature Extraction and Selection in Batak Toba Handwritten Text Recognition. International Journal of Artificial Intelligence, 18 (1). ISSN 0974-0635

[img] Text
2. Feature Extraction and Selection in Batak Toba Handwritten Text Recognition.pdf
Download (4Mb)
[img] Text
2. Turnitin_Feature Extraction and Selection in Batak Toba Handwritten Text Recognition.pdf
Download (4Mb)

Abstract

Batak Toba is one of the tribes in Indonesia that has its own language and alphabet. However, not many Batak Toba people are familiar with the alphabet. In this paper, we use several feature extraction methods in the recognition process of Batak Toba handwritten text. For some features, between/within-class scatter matrix criterion is used to select the significant features. The k-NN classifier is used in the recognition step. The results show that elliptic Fourier descriptor is the most superior features that has recognition percentage greater than the other categories.

Item Type: Article
Uncontrolled Keywords: Batak Toba text, Feature extraction, Elliptic Fourier descriptor, Feature selection, k-NN classifier.
Subjects: T Technology > T Technology (General)
Depositing User: Perpustakaan Maranatha
Date Deposited: 27 Oct 2021 09:33
Last Modified: 25 Mar 2023 06:37
URI: http://repository.maranatha.edu/id/eprint/27998

Actions (login required)

View Item View Item
MCUrepository is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.