基于方向信息測度的自適應多級中值濾波器
ADAPTIVE MULTILEVEL MEDIAN FILTER BASED ON DIRECTION INFORMATION MEASURE
-
摘要: 針對圖像濾噪中濾除噪聲和保護細節(jié)(邊緣)的矛盾,本文提出一種新的基于方向信息測度的自適應多級中值濾波器。方向性是邊緣和噪聲的本質區(qū)別之一,通過基于方向信息測度所構造的自適應結構,決定中值濾波器的形式和濾波窗口尺寸的大小,而不需要圖像和噪聲的先驗知識。文章比較了標準中值濾波器、單向多級中值濾波器、雙向多級中值濾波器和本文方法的結果。實驗表明本文方法具有更好的效果。
-
關鍵詞:
- 圖像濾噪; 信息測度; 中值濾波器
Abstract: This paper presents a new adaptive multilevel median filter based on direction information measure to solve the contradiction between noise attenuation and image details-preserving (edge-preserving). Direction is one of nature distinctions between noise and edge, and an adaptive algorithm based on direction information measure is constructed to decide the format and size of median filter without any knowledge of image and noise. A comparison of median filter, single direction multilevel filter, bidkection multilevel filter and the method in this paper has been done. The result indicates that the method has better effectiveness. -
D. Marr, T. Poggio, A computational theory of human stereo vision, Proc. Roy. Soc. London,1979, Vol. B-204, 301-328.[2]A. Nieminen, P. Heinonen, Y. Neuvo, A new class of detail-preserving filters for image processing,IEEE Trans. on PAMI, 1987, PAMI-9(1), 74-90.[3]G.R. Arce, R. E. Foster, Detail preserving ranked-order based filters for image processing, IEEE Trans. on ASSP, 1989, ASSP-37(1), 83-98.[4]楊煊,基于圖像信息測度的多尺度邊緣檢測方法研究,[博士論文],西安,西安交通大學,1998.[5]M.O. Ahmad, D. Sundarajan, A fast algorithm for two-dimensional median filtering, IEEE Trans.on CAS,1987, CAS-34(11), 1364-1374. -
計量
- 文章訪問數(shù): 2195
- HTML全文瀏覽量: 91
- PDF下載量: 717
- 被引次數(shù): 0
下載:
