Feature Extraction Comparison in Handwriting Recognition of Batak Toba Alphabet

Novie Theresia Br Pasaribu (Program Studi Teknik Elektro, Universitas Kristen Maranatha)
M. Jimmy Hasugian (Program Studi Teknik Elektro, Universitas Kristen Maranatha)

IJITEE (International Journal of Information Technology and Electrical Engineering)
Vol 1, No 3 (2017): September 2017

Offline handwriting recognition is one of the most prominent research topics due to its tremendous application and high variability as well. This paper covers the offline Batak Toba handwritten text recognition, from the noise removal, the process of feature extraction until the recognition by using several classifiers. Experiments show that elliptic fourier descriptor (EFD) is the most discriminative feature and Mahalanobis distance (MD) outperforms the two others classifier.