A model of image denoising based on partial differential equations

Author

Liu, Jing ; Gao, Fei ; Li, Zuozhou

Author_Institution

Dept. of Math., Dalian Maritime Univ., Dalian, China

fYear

2011

fDate

26-28 July 2011

Firstpage

1892

Lastpage

1896

Abstract

Based on the isotropic diffusion denoising model and the total variation (TV) denoising model, an image denoising model, which is constructed by the partial differential equations assigned a weight function is proposed. This model can improve "staircase effect" and "block effect" compared to the isotropic diffusion denoising model and the total variation (TV) denoising model, respectively. Numerical experiments show that the model is able to remove noise effectively while preserving texture features and edge information well, so that the quality of the denoised image is enhanced.

Keywords

edge detection; feature extraction; image denoising; image texture; partial differential equations; block effect; edge information; image denoising model; image quality; isotropic diffusion denoising model; partial differential equations; staircase effect; texture feature; total variation denoising model; Image denoising; Mathematical model; Noise reduction; Numerical models; PSNR; TV; image denoising; partial differential denoising model; variational method; weight function;

fLanguage

English

Publisher

ieee

Conference_Titel

Multimedia Technology (ICMT), 2011 International Conference on

Conference_Location

Hangzhou

Print_ISBN

978-1-61284-771-9

Type

conf

DOI

10.1109/ICMT.2011.6002346

Filename

6002346