Batik is a traditional clothwith unique patterns applied to fabric using a wax-resist dyeing technique. Aside from preserving this rich cultural heritage, the automated recognition of Batik patterns would enable many interesting applications. This paper introduces an approach to batik pattern recognition using the Scale Invariant Feature Transform (SIFT) as a feature extraction method. The challenging issues that arise are due to the highly symmetrical and repetitive properties of batik patterns. The Hough transform, as an evidence-based method of object detection, is applied to handle mismatched keypoints resulting from symmetrical and repetitive patterns of batik. On a collection of 120 batik images generated from 20 basic batik patterns, the proposed method showsan improvement over the original SIFT matching method with an equal error rate of 8.47%.
|Number of pages||10|
|Journal||Procedia Computer Science|
|Publication status||Published - 2015|
|Event||1st International Conference on Computer Science and Computational Intelligence, ICCSCI 2015 - Jakarta, Indonesia|
Duration: 24 Aug 2015 → 26 Aug 2015
- Hough transform