Wavelet transform based error concealment approach for image denoising

Gupta, Pradeep K and Kanhirodan, Rajan (2006) Wavelet transform based error concealment approach for image denoising. In: 1st IEEE Conference on Industrial Electronics and Applications, MAY 24-26, 2006, Singapore, pp. 1552-1557.

PDF 123getPDF.pdf - Published Version Restricted to Registered users only Download (1MB) | Request a copy

Abstract

Denoising of images in compressed wavelet domain has potential application in transmission technology such as mobile communication. In this paper, we present a new image denoising scheme based on restoration of bit-planes of wavelet coefficients in compressed domain. It exploits the fundamental property of wavelet transform - its ability to analyze the image at different resolution levels and the edge information associated with each band. The proposed scheme relies on the fact that noise commonly manifests itself as a fine-grained structure in image and wavelet transform allows the restoration strategy to adapt itself according to directional features of edges. The proposed approach shows promising results when compared with conventional unrestored scheme, in context of error reduction and has capability to adapt to situations where noise level in the image varies. The applicability of the proposed approach has implications in restoration of images due to noisy channels. This scheme, in addition, to being very flexible, tries to retain all the features, including edges of the image. The proposed scheme is computationally efficient.

Item Type:	Conference Paper
Series.:	IEEE Conference on Industrial Electronics and Applications
Publisher:	Institute of Electric Electronics Engineering
Additional Information:	Copyright of this article belongs to Institute of Electric Electronics Engineering.
Department/Centre:	Division of Physical & Mathematical Sciences > Instrumentation Appiled Physics
Date Deposited:	26 Aug 2010 05:47
Last Modified:	19 Sep 2010 06:12
URI:	http://eprints.iisc.ac.in/id/eprint/30709

Actions (login required)

View Item