Recognition of Bima script handwriting patterns using the local binary pattern feature extraction method and K-nearest neighbour classification method

Abstract

The Bima script is one of the cultural heritage of the archipelago that needs to be preserved. Based on the results of a questionnaire conducted by the author online with a total of 81 respondents from Bima, 48.1% had never studied the Bima script and 45.7% did not even know the existence of the Bima script. This study aims to build a machine learning model that can recognize the hand-written Bima script pattern by using the Local Binary Pattern (LBP) feature extraction, and the K-Nearest Neighbour classification method (K-NN). From the experimental results, LBP extracted good features and K-NN successfully classified the handwritten Bima letters. The best model generated in this study can classify the Bima script with an accuracy of 86.056%, were using the LBP of 9 radius value, the image size of 64x64 pixels, the zone of 8x8, and the neighbor value of k=1 with an 80:20 ratio between train and test data.