An improved anisotropic diffusion model for detail- and edge-preserving smoothing

被引:143
作者
Chao, Shin-Min [1 ]
Tsai, Du-Ming [1 ]
机构
[1] Yuan Ze Univ, Dept Ind Engn & Management, Tao Yuan, Taiwan
关键词
Edge-preserving smoothing; Image restoration; Image denoising; Anisotropic diffusion; IMAGE SEGMENTATION; DEFECT DETECTION; SCALE-SPACE; ENHANCEMENT; RESTORATION;
D O I
10.1016/j.patrec.2010.06.004
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
It is important in image restoration to remove noise while preserving meaningful details such as blurred thin edges and low-contrast fine features. The existing edge-preserving smoothing methods may inevitably take fine details as noise or vice versa. In this paper, we propose a new edge-preserving smoothing technique based on a modified anisotropic diffusion. The proposed method can simultaneously preserve edges and fine details while filtering out noise in the diffusion process. The classical anisotropic diffusion models consider only the gradient information of a diffused pixel, and cannot preserve detailed features with low gradient. Since the fine details in the neighborhood of the image generally have larger gray-level variance than the noisy background, the proposed diffusion model incorporates both local gradient and gray-level variance to preserve edges and fine details while effectively removing noise. Experimental results from a variety of test samples including shoulder patch images, medical images and artwork images have shown that the proposed anisotropic diffusion scheme can effectively smooth noisy background, yet well preserve edge and fine details in the restored image. (c) 2010 Elsevier B.V. All rights reserved.
引用
收藏
页码:2012 / 2023
页数:12
相关论文
共 46 条
[1]  
Almansa A, 2000, IEEE T IMAGE PROCESS, V9, P2027, DOI 10.1109/83.887971
[2]   IMAGE SELECTIVE SMOOTHING AND EDGE-DETECTION BY NONLINEAR DIFFUSION .2. [J].
ALVAREZ, L ;
LIONS, PL ;
MOREL, JM .
SIAM JOURNAL ON NUMERICAL ANALYSIS, 1992, 29 (03) :845-866
[3]  
Bakalexis SA, 2002, DSP 2002: 14TH INTERNATIONAL CONFERENCE ON DIGITAL SIGNAL PROCESSING PROCEEDINGS, VOLS 1 AND 2, P1203, DOI 10.1109/ICDSP.2002.1028309
[4]   Noise reduction for magnetic resonance images via adaptive multiscale products thresholding [J].
Bao, P ;
Zhang, L .
IEEE TRANSACTIONS ON MEDICAL IMAGING, 2003, 22 (09) :1089-1099
[6]   Robust anisotropic diffusion [J].
Black, MJ ;
Sapiro, G ;
Marimont, DH ;
Heeger, D .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 1998, 7 (03) :421-432
[7]  
Blanc-Feraud L., 1996, Vistas in Astronomy, V40, P531
[8]   THE EFFECT OF MEDIAN FILTERING ON EDGE ESTIMATION AND DETECTION [J].
BOVIK, AC ;
HUANG, TS ;
MUNSON, DC .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1987, 9 (02) :181-194
[9]   A review of image denoising algorithms, with a new one [J].
Buades, A ;
Coll, B ;
Morel, JM .
MULTISCALE MODELING & SIMULATION, 2005, 4 (02) :490-530
[10]   IMAGE SELECTIVE SMOOTHING AND EDGE-DETECTION BY NONLINEAR DIFFUSION [J].
CATTE, F ;
LIONS, PL ;
MOREL, JM ;
COLL, T .
SIAM JOURNAL ON NUMERICAL ANALYSIS, 1992, 29 (01) :182-193