以暗黑頻道預測為基礎的水下影像修復技術

Underwater Image Restoration Based on Dark Channel Prior

陳泊芳(Po-Fang Chen)

指導教授 ： 張恆華

國立臺灣大學/工學院/工程科學及海洋工程學研究所/碩士(2012年)

https://doi.org/10.6342/NTU.2012.01122

摘要

水下影像受到水中濃密混濁介質與水中普遍存在的懸浮微粒影響，使得光在水中傳遞時產生衰減、吸收及散射問題，導致水下影像的對比度嚴重下降，並產生如被一層薄紗覆蓋般的霧化現象。而不同顏色的光因為波長不同，入水後會隨波長的長短依序消失在水中，造成水下影像普遍偏藍這種由單一顏色主宰整張影像的情形。另一方面，我們觀察到水下影像退化的現象與濃霧中拍攝得到的影像十分類似，影像都同樣呈現低對比度與色彩改變的情形。本文改良影像除霧技術之暗黑頻道預測方法來解決水下影像的霧化問題，修復影像的清晰度。接著分別對每個RGB色彩頻道做線性轉換，均化各個色彩頻道亮度平均值，以解決影像色偏問題。最後使用 CLAHE方法進一步強化影像對比度，得到細節更加清晰的最終修復影像。實驗結果顯示，本研究方法能有效地對各種不同的水下影像除霧，並成功地復原色彩和對比。

關鍵字

水下影像 ； 影像修復 ； 影像強化 ； 影像除霧 ； 暗黑頻道預測

並列摘要

Underwater images are usually affected by the turbid water medium and floating particles existed in the water. Thus attenuation, absorption, and scattering happen while light propagates in the water. These phenomena cause low contrast in underwater images, and make them look like covering by a veil. In addition, colors disappear sequentially according to the wavelength while light travels deeper in the water, which makes underwater images blue. On the other hand, we observe that underwater images are similar to haze images because they have the same problems of low contrast and color shifting. This has motivated our use of the haze removal technique, namely, dark channel prior, to dehaze underwater images. Subsequently, we equalize the color mean in each RGB channel to balance the color. Finally, we use CLAHE to enhance the contrast of the images. Experimental results indicate that the proposed method effectively removes the haze in a wide variety of underwater images and successfully recovers the color and contrast.

並列關鍵字

underwater image ； image enhancement ； image restoration ； haze removal ； dark channel prior

參考文獻

[1] G. L. Foresti, Visual inspection of sea bottom structures by an autonomous underwater vehicle. IEEE Trans Syst Man Cybern B Cybern, 31(5), p. 691-705, 2001.

[2] D. F. Coleman, J. B. Newman, and R. D. Ballard, Design and implementation of advanced underwater imaging systems for deep sea marine archaeological surveys. Oceans 2000 Mts/IEEE - Where Marine Science and Technology Meet, Vols 1-3, Conference Proceedings, p. 661-665, 2000.

[4] Y. Y. Schechner and N. Karpel, Clear underwater vision. Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Vol 1, p. 536-543, 2004.

[5] R. Schettini and S. Corchs, Underwater Image Processing: State of the Art of Restoration and Image Enhancement Methods. EURASIP Journal on Advances in Signal Processing, (1), p. 746052, 2010.

[8] J. S. Jaffe, K. D. Moore, J. McLean, and M. P. Strand, Underwater optical imaging: status and prospects. Oceanography, 14, p. 66-76, 2001.