RANCANG BANGUN APLIKASI PENCARIAN CITRA BATIK BESUREK BERBASIS TEKSTUR DENGAN METODE GRAY LEVEL CO-OCCURRENCE MATRIX DAN EUCLIDEAN DISTANCE

Fathin Ulfah Karimah
Ernawati Ernawati
Desi Andreswari

Jurnal Teknologi Informasi Cyberku
Vol 11 No 1 (2015): Jurnal Teknologi Informasi CyberKU Vol.11 no 1

Image retrieval using text input is considered less effective because the search results sometimes do not represent the input. Thus, it is necessary to create a search application that processes image input to obtain the results representing the input image. The objective of this study is to build such application for Besurek Batik images of various patterns, a type of batik representing the traditional fabric of Bengkulu with MatLab R2012a programming language The models used to identify image patterns and to determine of similarity between tested images and training images are consecutively Gray Level Co-occurrence Matrix method oriented to the direction of 0º, 45º, 90º, and 135º and Euclidean Distance method, while the approach models used to develop and to design the system are respectively the Waterfall model and Data Flow Diagrams (DFD). The final result of this study is the image retrieval application based on the texture with the recall levels of 37.75 % and precision of 77 % with respect to the test of one batik besurek pattern and the recall levels of 30.26 % and precision of 82 % with respect to more than one pattern