A Combined Spatial and Frequency Domain Approach for Removal of Impulse Noise from Images
Malini S1, Moni R. S2
1Malini S, Marian Engineering College, Thiruvananthapuram, Kerala, India.
2Moni R. S, Prof, Marian Engineering College, Thiruvananthapuram Kerala, India.
Manuscript received on September 28, 2014. | Revised Manuscript received on October 17, 2014. | Manuscript published on October 30, 2014. | PP: 63-67  | Volume-4 Issue-1, October 2014. | Retrieval Number:  A3456104114/2013©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: Visual quality of any natural image is lost when it is corrupted by noise, especially by impulse noise. Further, essential features of the image cannot be retrieved from noisy image. Considering the reality that noise is ubiquitous, image denoising is an unavoidable prerequisite for any type of higher level image processing. A variant of the existing method of impulse noise removal is proposed in this paper. The method has two stages. Detection of noisy pixels and then replacing the noisy pixels by one of its non-noisy neighbour is the first stage. In the second stage, a multiresolution technique of image denoising is employed. The proposed method is found to be very effective in image denoising of grey as well as color images, as is evidenced by the given experimental results. Also the method is shown to be effective in reducing mixed noise from images.
Keywords: Denoising frequency domain, Impulse noise, multiresolution, Spatial method, Stationary wavelet.