EXTRACTION OF SHAPE CHARACTERISTICS ON JAVA KAWI Script USING L*A*B METHOD and K-Means Clustering

Abstract

Kawi Javanese script is one of the many cultural assets belonging to Indonesia that must be preserved and protected, one of which is by introducing it with a computer-based system, namely pattern recognition. In pattern recognition, shape extraction is a process that identifies and extracts shape features in digital images which can then be used as the initial classification process. This study aims to create a form extraction system for Kawi Javanese script which can then be used to classify Kawi Javanese script images so that they can be used for the process of reading Kawi Javanese script. Data collection in this study was taken from books using Javanese Kawi script with as many as 6 characters. In making this system using Matlab R2020a. Testing is carried out by processing 6 character images using the L*A*B and K-Means Clustering methods which will produce segmentation values ​​and then take shape feature values ​​including Area, Perimeter, Metric, and Eccentricity which can then be processed using the Artificial Neural Network method for classification. It is hoped that the values ​​of these parameters can be used as input values ​​for the classification of the Kawi Javanese script.