方向性中值濾波術應用於調適影像對比增強法

Directional Median Filter for Adaptive Image Contrast Enhancement

10.30153/JCAGST.200803.0003

鄧樹楨(S. J Jerome Teng)；徐道義(Taoi Hsu)

方向性中值濾波器 ； 調適性對比增強 ； 影像雜訊 ； 區塊作用 ； Directional Median Filter ； Adaptive Contrast Enhancement ； Image Noise ； Blocking effect.

中華印刷科技年報

2008（2008 / 03 / 01）

12 - 22

英文

影像增強(image enhancement)是許多影像處理應用中的重要技術，調適性影像質方圖均衡法(Adaptive Histogram Equalization)是一種強效影像對比增強方法，尤其是對於具有高明暗度範圍的影像，然而此種影像對比增強法會產生散發性雜訊，此雜訊來自於調適區塊作用與影像對比的過度增強，因此我們希望能移除此種特殊的雜訊，但仍能保留影像細節，例如線條、邊角等。中值濾波器(median filter)是一種非線性濾波作用，它適用於清除黑白點性或性質不明的雜訊，使其常被應用於移除影像中的脈衝性雜訊，同時儘量少造成影像線條的模糊。本論文提出「方向性中值濾波術」(Directional Median Filter)應用於調適性影像對比增強法的移除雜訊結果，濾波器操作是根據中央圖素與鄰近四個方向的圖素相互比較，此方法也優於傳統的方形中值濾波器，本論文將首度報告「方向性中值濾波術」的此項應用成效與比較結果。

Contrast enhancements are important techniques in many imaging applications. Adaptive histogram equalization (AHE) has been shown a powerful algorithm for image contrast enhancement, especially for images having a wide dynamic range. The AHE method allows each pixel within a contextual region to be processed separately. However, such an adaptive contrast enhancement introduces sporadic noise features contributed both from the blocking effect and the contrast amplification during the image processing. It is very desirable to remove this special kind of noise while maintaining image fine details, such as fine lines and edges. Median filter performs a nonlinear operation and is suited for suppressing impulsive kind of noise and for data smoothing when the noise characteristics are not known. This property is often used for image filtering where impulse noise needs to be reduced but edge blurring is not desirable. In this paper, we report the application of the directional median filter for noise removal in adaptive image contrast enhancement. The filtering operation is based on the differences and ranking between the central pixel and its neighbors aligned in four directions. This directional median filter of size 5x5 performs better than the conventional square-box median filter and preserves more details and lines in the image. Experimental results and comparisons of using this filtering technique on the AHE images are reported.

1. Dong Y.,Xu S.(2007).A New Directional Weighted Median Filter for Removal of Random-Valued Ipulse Noise.IEEE Signal Processing Letters,14(3),193-196.
2. Kim J-Y.,Kim L-S.,Hwang S-H.(2001).An advanced contrast enhancement using partially overlapped sub-block histogram equalization.IEEE Transactions on Circuits and Systems fro Video Technology,11(4),475-484.
3. Kuan D.T.,Sawchuk A.A.,Strand T.C.,Chavel P.(1987).Adaptive Restoration of Images with Speckle.IEEE Trans. ASSP.,35(3),373-383.
4. Pitas I.,Venetsanopoulos A.N.(1990).Nonlinear Digital Filters: Principles and Applications.Kluwer Academic.
5. Stark J.A.(2000).Adaptive image contrast enhancement using generalizations of histogram equalization.IEEE Transactions on Image Processing,9(5),889-896.
6. Sun T.,Neuvo Y.(1994).Detail-preserving medianbased filters in image processing.Pattern Recognition Letters,15(4),341-347.
7. Teng S.J.J.,Hsu T.(2007).Applications of Adaptive Histogram Equalization for Image Contrast Enhancement.Journal of Chinese Association of Graphics Science & Technology (JCAGS),176-190.
8. Tukey J.W.(1974).Nonlinear (nonsuperposable) methods for smoothing data.Conference Rec.
9. Tukey J.W.(1977).Exploratory Data Analysis.Menlo Park, CA:Addison-Wesley.
10. Wiener filter.The Wikipedia, the free encyclopedia on the Internet.
11. Yuan S.-Q.,Tan Y.-H.(2006).Difference-Type Noise Detector for Adaptive Median Filter.Electronics Letters,42(8)
12. Zimmerman J.B.,Pizer S.M.(1988).An evaluation of the effectiveness of adaptive histogram equalization for contrast enhancement.IEEE Transactions on Medical Imaging,7(4),304-312.