Image Retrieval Based on Multi Structure Co-occurrence Descriptor

Abstract: This study presents a new technique for Batik cloth image retrieval using Micro-Structure Cooccurence Descriptor (MSCD). MSCD is one of feature extraction method. It is a developed method based on Enhanced Micro Structure Descriptor (EMSD). In previous study, EMSD has been used for extracting texture, color, and shape feature simultaneously. However, EMSD could not achieve an optimal precision for Batik cloth image retrieval. Therefore, MSCD is proposed to overcome the EMSD drawback using global feature approach, namely Gray Level Co-occurrence Matrix (GLCM). There are 300 batik cloth images, which contain 50 classes used for dataset. The performance result shows that MSCD can increase precision and recall, by 19.03% and 18.33%, respectively. It indicates that MSCD can retrieve Batik cloth images more effective than EMSD.
Keywords: batik, image retrieval, micro structure, gray level co-occurrence matrix
Author: Agus Eko Minarno
Journal Code: jptkomputergg160264

Artikel Terkait :