Fast MAP-based super-resolution algorithm using multilevel interpolation and reconstruction
Authors:
- 13 September 2012
- journal article
- research article
- Published by Taylor & Francis in Journal of the Chinese Institute of Engineers
- Vol. 35 (6) , 711-722
Abstract
Super-resolution (SR) image reconstruction can produce a high-resolution (HR) digital image from multiple low-resolution (LR) photographs. Many applications require HR images, including medical imaging, satellite imaging, and video applications. Recovering lost details from down-sampled images is the main challenge of SR techniques. However, most studies have not taken computational complexity into consideration. Therefore, this article presents a new fast maximum a posteriori (MAP)-based SR image reconstruction method based on a multilevel algorithm. In particular, this work focuses on the case of input LR images that do not meet the requirements for the analysis. A two-step interpolation process is proposed to increase the quality of the constructed SR image. Experimental results show that the proposed algorithm reduces the blocking artifacts in the reconstruct SR image caused by a lack of LR images. Compared to the conventional MAP-based method, the proposed method also achieves up to 60% reduction in computational complexity with comparable peak signal-to-noise ratio and visual quality.Keywords
This publication has 8 references indexed in Scilit:
- A Fast and Robust Super Resolution Method for Intima Reconstruction in Medical Ultrasound2009
- An efficient, SELective, Perceptual-based super-resolution estimator2008
- Fast MAP-based multiframe super-resolution image reconstructionImage and Vision Computing, 2005
- Super-resolution image reconstruction: a technical overviewIEEE Signal Processing Magazine, 2003
- Fast superresolution for image sequences using motion adaptive relaxation parameters2003
- Limits on super-resolution and how to break themIEEE Transactions on Pattern Analysis and Machine Intelligence, 2002
- A computationally efficient superresolution image reconstruction algorithmIEEE Transactions on Image Processing, 2001
- High-resolution image reconstruction from a sequence of rotated and translated frames and its application to an infrared imaging systemOptical Engineering, 1998