彩色濾波器陣列補插之研究

A study of color demosaicking by high order interpolation

黃于豪(Yu- Hau Huang)

指導教授 ： 王鵬華

國立臺北大學/電機資訊學院/通訊工程研究所/碩士(2010年)

摘要

在本論文中，我們對於彩色影像內插法做了研究與改進。數位相機通常是透過彩色濾鏡陣列(color filter array)擷取彩色影像，每一個像素只擷取一種顏色，然後透過彩色影像內插法插補出缺少的顏色。常見的彩色濾鏡陣列排列方式為 Bayer 排列。本論文使用高階線性內插法以得到每一像素缺少顏色的估計值，對於顏色變化不大的區域，高階線性內插法可以準確地計算出顏色的估計值，但是對於顏色變化較大的邊緣部份，高階線性內插可能會產生模糊的邊緣以及錯誤的顏色，為了改進邊緣部份的影像品質，我們在未知顏色像素的不同方向以高階線性內插法計算出各自方向的估計值，然後根據這些估計值組合出最後的估計值。本論文使用兩種方法以組合這些不同方向的估計值，一種是中值濾波(median filtering)；另一種是線性組合，而線性組合的權重是決定於邊緣偵測的結果。為了評估本論文的結果，我們計算原始影像和內插影像之間的 PSNR (peak signal-to-noise ratio)值，實驗結果顯示，我們的方法對於大部分的測試影像都有改進。

關鍵字

彩色濾波器陣列

並列摘要

In this thesis, we propose a method of improving color interpolation for color filter array. Digital camera usually captures color images through the color filter array, and obtains one color for each pixel. The missing colors are estimated by color interpolation. The common color filter array used in practical is Bayer pattern. In this thesis, we uses high order linear interpolation to estimate the missing colors. For smooth regions, high order linear interpolation can obtain accurate estimate. However, for the regions containing edges, linear interpolation may blur edges and leads to false color.To improve image quality around edges, we calculate the directional estimates of missing color by linear interpolation, then combine these estimates to obtain the final one. In this thesis, we use two methods to combine the directional estimates. One is median filtering, and another is a linear combination whose weights are determined by the result of edge detection. To evaluate the performance of the proposed method, we use PSNR (peak signal to noise ratio) between the original image and interpolated one. The resulting PSNRs show that our approach improves the quality of most of our test images.

並列關鍵字

color filter array

參考文獻

[1] Li, J.S.J. and Randhawa, S., “Color filter array demosaicking using highorder interpolation techniques with a weighted median filter for sharp color edge preservation” , IEEE Trans. Image Process.,vol. 18, pp. 1946-1957, SEPTEMBER 2009.

[2] Pei, S.C. and Tam, I.K., “Effective color interpolation in CCD color filter arrays using signal correlation” , IEEE Trans. Circuits And Systems For Video Technology.,vol. 13, pp. 503-513, JUNE 2003.

[3] Hornberg, A., Handbook of machine vision , Wiley-VCH, 2006.

[6] Adams Jr, J.E., “Interactions between color plane interpolation and other image processing functions in electronic photography” , in Proc. SPIE,vol.2416, C. Anagnostopoulos and M. Lesser, Eds., Bellingham, WA, 1995, pp.144-151.

[8] W. Lu and Y.-P. Tan, “Color filter array demosaicking: New method and performance measures” , IEEE Trans. Image Process.,vol. 12, pp. 1194-1210,2003.