Structure Extraction from Complex Textures using Gradient based Relative Total Variation
Tapinder Kaur1, Ashish Verma2

1Tapinder Kaur, Department of Computer Engineering, Punjab Technical University, SSIET, Derabassi (Punjab), India.
2Ashish Verma, Department of Computer Engineering, Punjab Technical University, SSIET, Derabassi (Punjab), India.
Manuscript received on 11 January 2014 | Revised Manuscript received on 20 January 2014 | Manuscript Published on 30 January 2014 | PP: 1-6 | Volume-3 Issue-8, January 2014 | Retrieval Number: H1418013814/14©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: It is universal that expressive arrangements are shaped by or seem over textured exteriors. Extracting them under the difficulty of texture patterns, which could be consistent, near regular, or uneven, is very challenging, but of great real world significance. This paper presents a review on texture segmentation and analysis of different techniques. The comparisons among available techniques are also drawn in order to find the best suitable one. To overcome the existing problems of texture extraction a new extended relative total variation technique is proposed. The proposed technique has ability to extract textures from complex background images using gradient based methods and median filtering. The proposed method has shown accurate results even in highly noisy images. The comparison has shown that the proposed method is quite significant over the available methods.
Keywords: Dark Regions, Gradients, Texture, Texture Patterns, Texture Segmentation.

Scope of the Article: Agent-Based Software Engineering