Multiscale Geometric Representation for Single Image Super-Resolution
S. H. Jagtap1, M. M. Patil2, S. D. Ruikar3
1Mr. S. H. Jagtap, Department of Electronic and Telecommunication Engineering, Pune University, Maharashtra, India.
2Prof. M. M. Patil, Department of Electronic and Telecommunication Engineering, Pune University, Maharashtra, India.
3Prof. S. D. Ruikar, Department of Electronic and Telecommunication Engineering, Pune University, Maharashtra, India.
Manuscript received on March 02, 2012. | Revised Manuscript received on March 31, 2012. | Manuscript published on April 30, 2012. | PP: 219-224 | Volume-1 Issue-4, April 2012 | Retrieval Number: D0348041412/2012©BEIESP
Open Access | Ethics and Policies | Cite
© 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: This paper presents comparative study of different single image SR algorithms and takes deep drive on a new approach to single-image super-resolution, based upon Multiscale Geometric Representations. Nowadays computational power of processors are increasing therefore, it has become feasible to use more robust and computationally complex algorithms that increase the resolution of images without distorting edges and contours. We present a novel image interpolation algorithm that uses the new contourlet transform to improve the regularity of object boundaries in the generated images. By using a simple wavelet-based linear interpolation scheme as our initial estimate, we use an iterative projection process based on two constraints to drive our solution towards an improved high-resolution image. This algorithm generates high-resolution images that are competitive or even superior in quality to images produced by other similar SR methods.
Keywords: Image super-resolution (SR), contourlet transform, geometric regularity, directional multi-resolution image.