Title :
A new approach to identify the nature of the noise affecting an image
Author :
Chehdi, K. ; Sabri, M.
Author_Institution :
Rennes Univ., Lannion, France
Abstract :
A method for the identification of the nature of noise and estimation of its statistical parameters from only the degraded image is proposed. This then makes it possible to apply the appropriate filtering algorithm. The method is based on the criterion of maximum likelihood in order to detect homogeneous regions, and this makes it possible to identify the nature of the noise. The identification criteria used are based on the analysis of local variations of the average and the standard deviation of the observed image. These statistical parameters are determined uniquely from regions of the image supposed to be homogeneous locally. After the nature of the noise has been identified, the statistical parameters of the noise are estimated
Keywords :
image processing; maximum likelihood estimation; noise; parameter estimation; additive noise; degraded image; filtering algorithm; homogeneous regions; impulsive noise; maximum likelihood estimation; multiplicative noise; noise identification; statistical parameters; Additive noise; Biomedical imaging; Degradation; Filtering algorithms; Image analysis; Maximum likelihood detection; Maximum likelihood estimation; Noise level; Pattern recognition; Radar imaging;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1992. ICASSP-92., 1992 IEEE International Conference on
Conference_Location :
San Francisco, CA
Print_ISBN :
0-7803-0532-9
DOI :
10.1109/ICASSP.1992.226195