A Novel Reliable Approach For Image Batik Classification That Invariant With Scale And Rotation Using MU2ECS-LBP Algorithm

Abstract

Abstract Batik in Indonesia has various types of patterns, which are arranged repeatedly to illustrate the basic motifs of cloth as a whole. Types of batik motifs collected through several sources such as batik magazines, the internet or directly using a digital camera. Batik’s pattern automatic classification still requires some improvement especially in regards to invariant with scale and rotation. The batik‘s pattern of changing this dilemma at the same time needs a feature extraction algorithm that is reliable in supporting image classification. This algorithm is multiwindow and multiscale extended center symmetric local binary patterns (MU2ECS-LBP) which uses several windows such as the size of 6x6, 9x9, 12x12, and 15x15 or a combination between windows. To recognize the batik pattern automatically, we implement a batik classification method using kNN and ANN. After doing some experiments the results of the accuracy values with the kNN method, based on the effect of training images and image conditions, on multi-window 6-9-12-15, another multi-window 6-12-15 with overlapping images of 40 and 50 pixels and the number of image classes 5, 9 and 12 classes are 99.91% and 99.8%. With the ANN method, the accuracy classification value is based on the multi-window effect and image overlap, where the highest accuracy value on multiwindow 6-9-12-15 with ANN architecture 64-240-12 is 98.43%. A novel algorithm is a development of the local binary pattern algorithm, but by looking at the results of classification accuracy which is very good and reliable. So that the feature extraction algorithm is very feasible to be developed and continued for other research.