Title :
Image filter method based on maximal entropy principle
Author :
Dong, Lichun ; Yang, Lianhe ; Liang, Chen
Author_Institution :
Comput. Sci. & Software Instn., Tianjin Polytech. Univ., Tianjin, China
Abstract :
This paper selects Haar wavelet as the basis, using the soft threshold method wavelet coefficients of compression processing, image filtering experiment is accomplished by wavelet compression de-noising method. The approach to determine the appropriate threshold coefficient affects the noise level directly, so the approach used to determine the threshold coefficient is very important. The maximum entropy principle is used to determine threshold coefficient in this paper. De-noising experimental results of a picture with noise shows that the method to determine the threshold coefficient through the maximum entropy principle and then wavelet compress de-noising is accomplished, which is used to complain combine the threshold coefficient through the maximum entropy principle and wavelet compression de-noising and the result is satisfied.
Keywords :
Haar transforms; image coding; image denoising; maximum entropy methods; wavelet transforms; Haar wavelet; image filter; maximal entropy principle; soft threshold method wavelet coefficients; wavelet compression denoising; Entropy; Image coding; Information filters; Noise; Noise reduction; Wavelet coefficients; maximum entropy; reducing image noise; threshold method;
Conference_Titel :
Computer and Information Application (ICCIA), 2010 International Conference on
Conference_Location :
Tianjin
Print_ISBN :
978-1-4244-8597-0
DOI :
10.1109/ICCIA.2010.6141527