Texture Feature Extraction using Improved Completed Robust Local Binary Pattern for Batik Image Retrieval

Abstract

One of the robust texture feature extraction methods is Local Binary Pattern (LBP). LBP is a simple but efficient method and gray-scale invariant. Some studies have been proposed to improve the performance of LBP, such as Completed Robust Local Binary Pattern (CRLBP). CPLBP is proposed by Zhao to overcome the weaknesses of CLBP that is sensitive to noise. However, CRLBP is not invariant to rotation. From that problem, in this study, a new approach method of CRLBP is proposed, that proposed method is called Improved Completed Robust Local Binaty Pattterh (ICRLBP), in lCRLBP algorithm, CRLBP algorithm will be inserted by LBPROT algorithm. LBPROT is one of improved LĔP methods that proposed to overcome the LBP weakness which is not rotation invariant. Inserting LBPROT into CRLBP is simpty carried out by shifting the binary value obtained from CRLBP to get the smallest integer value. The performance of ICRLBP is evaluated in content-based image retrieval system using 4 datasets, namely Batik, Tektile, Brodatz, and Corel datasets. The result experiments shaw that the average of precision, recall, and speed of ICRLBP using Modified Canberra distance and M_C feature increased by 14.28%, 13.52%, and 3 times, respectively. Whereas, the average of precision, recall, and speed of ICRLBP using L1 distance and S_M_C feature increasĕd by 21.14%, 20.03%, and 56 times, respectivety, It show that ICRLBP is proven can improve the performance of CRLBP.