A Filter Selection Method in Hard Thresholding Recovery for Compressed Image Sensing
Compressed sensing has been widely researched since the beginning of 2000s. Although there are several well-known signal recovery algorithms, its reconstruction noise cannot be avoided completely, thus requiring good filters to remove the noise in the reconstructing process. Since each different filter has its own advantages and disadvantages depending on specific reconstruction algorithm, the reconstruction performance can be varied according to the choice of filter. This paper proposes an inner filter selection method according to the sampling rate and the property of image to be sensed.
KeywordsCompressed sensing Wiener filter Median filter
This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MEST) (No. 2011-001-7578).
- 2.Donoho DL (2006) Compressed sensing. IEEE Trans Inform Theory 52(4): 1289–1306Google Scholar
- 3.Candes EJ, Romberg J, Tao T (2006) Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information. IEEE Trans Inform Theory 52(2): 489–509Google Scholar
- 6.Kumar S, Kumar P, Gupta M, Nagawat AK (2010) Performance comparison of median and Wiener filter in image de-noising. Int J Comput Appl (0975–8887) 12(4): 27–31Google Scholar
- 7.Candes E, Romberg J (2005) ℓ1-magic: recovery of sparse signals via convex programming. Technical report, California Institute of TechnologyGoogle Scholar
- 12.Church JC, Yixin C, Rice SV (2008) A spatial median filter for noise removal in digital images. In: IEEE Southeastcon, Alabama, pp 618–623Google Scholar