An iterative method for preserving edges and reducing noise in high resolution image reconstruction

Authors:
Chanho Jung;Gyeonghwan Kim
Affiliations:
Department of Electronic Engineering, Sogang University, Seoul, Rep. of Korea;Department of Electronic Engineering, Sogang University, Seoul, Rep. of Korea
Venue:
ACCV'06 Proceedings of the 7th Asian conference on Computer Vision - Volume Part II
Year:
2006

Citing 5
Cited 0

A regularization approach to joint blur identification and image restoration

IEEE Transactions on Image Processing
Extraction of high-resolution frames from video sequences

IEEE Transactions on Image Processing
Restoration of a single superresolution image from several blurred, noisy, and undersampled measured images

IEEE Transactions on Image Processing
Regularized adaptive high-resolution image reconstruction considering inaccurate subpixel registration

IEEE Transactions on Image Processing
Super-resolution still and video reconstruction from MPEG-coded video

IEEE Transactions on Circuits and Systems for Video Technology

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we present simple iterative method for obtaining high resolution images with enhanced edges but reduced noise. In the method the trade off between the output noise and the edge preservation is being taken care of by employing an energy-based framework. In each iteration, two processes are involved: 1) the edge enhancement and reducing noise which occurs during the edge enhancement process, and 2) consideration of the fidelity to the low resolution images and the smoothness constraint of the restored high resolution image. In the implementation, the first process is designed to be embedded into the second process. And a termination condition is established by taking into account high frequency energy of the image being restored and error energy for each low resolution image. Experimental results show that the proposed method produces high resolution images in which edges are preserved with reduced noise, comparing to the ones produced by conventional methods. Moreover, it turns out that the approach is less sensitive to initialization factor in terms of PSNR and subjective visual quality.