Removal of PCA Based Estimated Noise in Processed Images
| Author(s) | : | Neethu Mohan |
| Institution | : | chool of Computer Sciences, M G University, Kottayam |
| Published In | : | Vol. 1, Issue 9 — September 2014 |
| Page No. | : | 153-156 |
| Domain | : | Engineering |
| Type | : | Research Paper |
| ISSN (Online) | : | 2348-4470 |
| ISSN (Print) | : | 2348-6406 |
Noise is an important problem in image processing applications. This noise level is to be estimated and is tobe removed. Blind noise level estimation is an important image processing step. The proposed system is a new noiselevel estimation and removal method. It estimates noise based on principal component analysis (PCA) of image blocks.Principal component analysis is one of the statistical techniques frequently used in signal processing to the datadimension reduction or to the data correlation. In PCA first rearrange image blocks into vectors and compute thecovariance matrix of these vectors. Then select the covariance matrix eigen values, which correspond only to noise. Thisallows estimating the noise variance as the average of these eigen values. The blocks to process are selected from imageregions with the smallest variance. After noise level estimation the noise is removed using denoise function. It does notrequire images with homogeneous areas.
Neethu Mohan, “Removal of PCA Based Estimated Noise in Processed Images”, International Journal of Advance Engineering and Research Development (IJAERD), Vol. 1, Issue 9, pp. 153-156, September 2014.
