Title :
Minimum Noiseless Description Length (MNDL) Thresholding
Author :
Fakhrzadeh, Azadeh ; Beheshti, Soosan
Author_Institution :
Dept. of Electr. & Comput. Eng., Ryerson Univ., Toronto, Ont.
Abstract :
In this paper, a new thresholding approach for data denoising is presented. The approach is based minimum noiseless description length (MNDL), a new method for optimum sub-space selection in data representation. By using the observed noisy data, this information theoretic approach provides the optimum threshold that minimizes the description length of the noiseless signal. Comparison of the new method with the existing thresholding methods is provided
Keywords :
signal denoising; data denoising; data representation; information theory; minimum noiseless description length thresholding; noiseless signal; optimum subspace selection; Additive noise; Computational intelligence; Data engineering; Image reconstruction; Noise reduction; Random variables; Signal denoising; Signal processing; Signal restoration; Upper bound;
Conference_Titel :
Computational Intelligence in Image and Signal Processing, 2007. CIISP 2007. IEEE Symposium on
Conference_Location :
Honolulu, HI
Print_ISBN :
1-4244-0707-9
DOI :
10.1109/CIISP.2007.369308
