Noise Reduction of Hyperspectral Imagery Nonlocal sparse Representation
Milad Razmi1, Ali Rafiee2, Zoheir Kordrostami3
1Milad Razmi, Department of Electrical Engineering, Branch Bushehr slamic Azad University, Bushehr, Iran.
2Ali Rafiee, Electrical Engineering Department Kazeroun Branch Islamic Azad University, Kazeroun, Iran.
3Zoheir Kordrostami, Department of Electrical Engineering, Shiraz University of Technology, Shiraz, Iran.
Manuscript received on October 11, 2013. | Revised Manuscript received on October 15, 2013. | Manuscript published on October 25, 2013. | PP:82-85 | Volume-1, Issue-12, October 2013. | Retrieval Number: L05331011213/2013©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Published By: 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: Noise reduction is always an active research area in image processing due to its importance for the sequential tasks such as object classification and detection. In this paper, we develop a sparse representation based noise reduction method for hyperspectral imagery, which is dependent on the ssumption that the non-noise component in the signal can be approximated by only a small number of atoms in a dictionary while noise component has not this property. The main contribution of the paper is in introducing nonlocal similarity and spectralspatial structure of hyperspectral imagery into sparse representation. Non-locality means the self-similarity of image, by which the whole image can be partitioned into some groups containing similar patches. The similar patches in each group is sparsely represented with shared atoms making the signal and noise more easily separated. Sparse representation with spectral-spatial structure can exploit spectral and spatial joint correlations of hyperspectral imagery also making the signal and noise more distinguished, in which 3-D blocks are instead of 2-D patches for sparse coding. The experimental results indicate that the proposed method has a good quality of restoring the true signal from the noisy observation.
Keywords: Hyperspectral imagery, noise reduction, sparse representation, nonlocal similarity, spectral-spatial structure.