Xue-zhen CHENG（程学珍）, Jing-zhao ZHANG（张京钊），Mao-yong CAO（曹茂永）
Shandong University of Science and Technology, Qingdao 266510, China)
Abstract-Based on the characteristics of impulse noises, the authors establish a new filter, Iterative Adaptive Median Filter (IAMF). Accordi ng to the characteristics of images polluted by impulse noises, they establish w eight function combined with iterative algorithm to eliminate noises. In IAMF fi lter process, because the noise spots do not participate in the computation, the y do not influence the normal points in the image, therefore IAMF can retain the detail well, maintain the good clarity after processing image, and simultaneous ly reduce the computation. Experiments showed that IAMF have ideal denoising eff ect for the images polluted by the impulse noises; especially when the noise rat es are more than 0.5, IAMF is more prominent, even when the noise rates are more than 0.9, IAMF can achieve a satisfactory results.
Key words-iterative adaptive median filter; impulse noi se; image processing
Manuscript Number: 1674-8042(2010)04-0326-04
dio: 10.3969/j.issn.1674-8042.2010.04.06
References
[1]G.C.Rafael, W.E.Richard, 2007. Digital Image Processing. Publishing House of Electronics Industry, Beijing.
[2]Chang-shing Lee,Yau-hwang Kuo,1997.Weighted fuzzy mean filters f or image processing. Fuzzy Sets and System, 89: 157-180.
[3]Shu-mei Guo, Chang-shing Lee, 2005. An intelligent image agent bas ed on soft-computing techniques for color image processing. Expert Sy stem with Applications, 28: 483-494.
[4]Tzu-chao Lin, Pao-ta Yu, 2004. Partition fuzzy median filter based on fuzzy rules for image restoration. Fuzzy Sets and System, 147: 75-97.
[5]Yau-hwang Kuo,Chang-shing Lee,Chao-lieh Chen,20-00.High-stabi lity AWFMlter for signal restoration and its hardware design. F uzzy Sets and Systems, 114: 185-202.
[6]Y. J. Zhang, 1996. A survey on evaluation methods for image segmenta tion. Pattern Recognition, 29(8): 1335-1346.
[7]H. Franke, J. Jann, J. Moreira, et al, 1999. An Evalu-ation of Para llel Job Scheduling for ASCI Blue-Pacic. Proceedings of SC'99, Portland, Oregon , ACM Press and IEEE Computer Society Press, p.11-18.
[8]E. Krevat, J. Castanos, J. Moreira, 2002. Job scheduling for the Blu e G ene/Lsystem. Job Scheduling Strategies for Parallel Processing, Lecture Notes in Computer Science, Springer, p.38-54.
[9]Eitan Frachtenberg, Fabrizio Petrini, Juan Fernandez, et al, 2002. S calable Resource Management in High-Performance Computers. Proceed-ings of the IEEE International Conference on Cluster Computing Cluster'02, Chicago, IL.
[full text view]
