單張影像除霧之相關研究議題於近年受到大量關注，且許多能有效消除影像中因霧氣所造成之散射影響的演算法陸續被提出。然而這些演算法直接應用於水下影像之結果並不理想。其中一個原因為，不同波長的光在水下之吸收衰減速度皆有不同，導致水下影像存在著明顯的色偏。因此，本論文提出一種能有效改善影像對比度並修正色偏之水下影像強化方法。所提出方法之第一個步驟為，利用Red Channel Prior計算出影像之水下背景光強度與透射率圖。為避免除霧後之影像產生光暈效應，利用WLS濾波器(weighted least square filter)對透射率圖進行修正，使其在原始影像景深變化劇烈處保有邊緣特徵。在消除影像之散射影響後，分別對RGB(red、green、blue)三個顏色通道進行吸收衰減補償，消除光線自物體表面傳輸至成像裝置過程中吸收衰減所造成之影響。接著利用白平衡方法，修正因光線自水面傳輸至物體表面這段距離之衰減所造成的色偏現象。最後，將除霧後影像進行直方圖拓寬，得到不同曝光程度之除霧後影像，並將這些影像融合為一張各局部區域曝光良好之除霧後影像。實驗結果顯示，經本論文所提出方法處理後之絕大部分測試影像的量化數據，優於多個近年來所提出之水下影像除霧方法之結果。本論文也利用SIFT特徵點配對進行測試，證明所提出之方法能使影像被找出更多特徵點。未經處理之水下影像原本僅能正確配對約四個特徵點，經所提出之方法處理後之影像可正確配對約四十個特徵點。

In recent years, the topic of single image dehazing has received a lot of attention, and many enhancing algorithms have been proposed to efficiently alleviate the effect of light scattering. However, these algorithms cannot be directly applied to the underwater images effectively. One of the main reasons is that the color cast caused by various degrees of absorption for different light wavelengths in the underwater environment cannot be neglected. Therefore, this thesis proposes a series of image processing steps in order to enhance the color contrast and correct the color cast of the underwater images. Our first step is to apply Red Channel Prior approach with modified pixel selection criteria in order to estimate the underwater background light intensity. In order to obtain the transmission map without halo artifacts, weighted least square filter has been used to refine the map to preserve the edge. After eliminating the scattering attenuation effect, our next step is to compensate the absorption loss along the propagation path between objects and camera. Then, the white balance algorithm is utilized to remove the color cast caused by the light attenuation along the propagation path between the water surface and the objects. Finally, multiple images obtained by histogram stretching of different scales will be fused to adjust the light intensity of the dehazed images. Our experimental results show that the proposed approach can achieve the best quantitative visual metric results for most of the test images compared with those recent state-of-the-art ones. We also conduct SIFT feature test in order to illustrate if our enhanced method can help increasing the image features. Our test results show that an average of 40 features can be found matched after applying our dehazing method compared with only four can be detected in the original images.

論文審定書 i
論文公開授權書 ii
摘要 iii
Abstract iv
目錄 v
圖次 vii
Chapter 1 概論 1
1.1 研究動機 1
1.2 論文大綱 2
Chapter 2 研究背景與相關研究 3
2.1 大氣散射原理 3
2.2 暗原色先驗影像除霧演算法 4
2.2.1 暗原色先驗 5
2.2.2 透射率估算 6
2.2.3 大氣光值計算 8
2.2.4 無霧影像還原 9
2.3 水下影像強化相關文獻 10
2.3.1 Galdran等人之方法 10
2.3.2 Chiang等人之方法 12
2.3.3 Ancuti等人之方法 17
Chapter 3 所提出之水下除霧方法 19
3.1 水下成像模型 19
3.2 水下背景光計算 21
3.3 透射率圖計算 22
3.4 透射率圖修正 23
3.5 消除散射影響 26
3.6 光線吸收衰減補償 26
3.7 色偏校正 29
3.8 對比度提升 32
3.8.1 直方圖拓寬 32
3.8.2 多重曝光融合 34
Chapter 4 實驗結果與分析 38
4.1 量化數據比較 38
4.2 視覺比較 40
4.3 應用測試比較 47
Chapter 5 結論與未來工作 52
5.1 結論 52
5.2 未來工作 52
參考文獻 53

[1] D.-M. He and G. G. L. Seet, "Divergent-beam Lidar imaging in turbid water," Optics and Lasers in Engineering, vol. 41, no. 1, pp. 217-231, 2004.
[2] T. Treibitz and Y. Y. Schechner, "Active Polarization Descattering," IEEE Trans Pattern Anal Mach Intell, vol. 31, no. 3, pp. 385-399, 2009.
[3] J. J. Chiang and Y.-C. Chen, "Underwater Image Enhancement by Wavelength Compensation and Dehazing," IEEE Transactions on Image Processing, vol. 21, no. 4, pp. 1756-1769, Apr. 2012.
[4] C. Ancuti, C. O. Ancuti, T. Haber and P. Bekaert, "Enhancing underwater images and videos by fusion," 2012 IEEE Conference on Computer Vision and Pattern Recognition, Providence, RI, 2012, pp. 81-88.
[5] K. He, J. Sun and X. Tang, "Single Image Haze Removal Using Dark Channel Prior," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 33, no. 12, pp. 2341-2353, Dec. 2011.
[6] S. G. Narasimhan and S. K. Nayar, " Vision and the Atmosphere," in International Journal of Computer Vision, vol. 48, pp. 233-254, 2002.
[7] A. Levin, D. Lischinski and Y. Weiss, "A Closed-Form Solution to Natural Image Matting," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 30, no. 2, pp. 228-242, Feb. 2008.
[8] K. He, J. Sun and X. Tang, "Guided Image Filtering," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 35, no. 6, pp. 1397-1409, Jun. 2013.
[9] R. T. Tan, "Visibility in bad weather from a single image," 2008 IEEE Conference on Computer Vision and Pattern Recognition, Anchorage, AK, 2008, pp. 1-8.
[10] R. Fattal, "Single Image Dehazing," in ACM Transaction on Graphics, vol. 27, no. 3, pp. 1-8, 2008.
[11] K. Zuiderveld, “Contrast limited adaptive histogram equalization,” in Graphics Gems IV, San Diego, CA, USA： Academic, 1994, pp. 474–485.
[12] A. Galdran, D. Pardo, A. Picon and A. Alvarez-Gila, "Automatic Red-Channel underwater image restoration," Journal of Visual Communication and Image Representation, vol. 26, pp. 132-145, Jan. 2015.
[13] G. Finlayson and E. Trezzi, “Shades of gray and colour constancy,” in IS&T/SID Twelfth Color Imaging Conference, 2004, pp. 37–41.
[14] C. Ancuti, C. O. Ancuti, C. De Vleeschouwer, R. Garcia and A. C. Bovik, "Multi-scale Underwater Descattering," in International Conference on Pattern Recognition (ICPR), 2016.
[15] N. Carlevaris-Bianco, A. Mohan, and R. M. Eustice, “Initial results in underwater single image dehazing,” in Proc. IEEE OCEANS, Sep. 2010, pp. 1–8.
[16] S. Emberton, L. Chittka and A. Cavallaro, "Hierarchical rank-based veiling light estimation for underwater dehazing," in British Machine Vision Conference(BMVC), 2015.
[17] S. Serikawa and H. Lu, "Underwater image dehazing using joint trilateral filter," Computers & Electrical Engineering, vol. 40, no. 1, pp. 41-50, 2014.
[18] H. Wen, Y. Tian, T. Huang and W. Gao, "Single underwater image enhancement with a new optical model," 2013 IEEE International Symposium on Circuits and Systems (ISCAS2013), Beijing, 2013, pp. 753-756.
[19] Z. Farbman, R. Fattal, D. Lischinski and R. Szeliski, "Edge-Preserving Decompositions for Multi-Scale Tone and Detail Manipulation," ACM Transaction on Graphics, vol. 27, no. 3, pp. 1-10, Aug. 2008.
[20] D. Park, D. K. Han and H. Ko, "Enhancing underwater color images of diving mask mounted digital camera via non-local means denoising," 2016 IEEE International Conference on Consumer Electronics (ICCE), Las Vegas, NV, 2016, pp. 441-442.
[21] A. A. Goshtasby, "Fusion of Multi-Exposure Images," Image and Vision Computing, vol. 23, pp. 611-618, 2005.
[22] Z. Xiang, "Color Image Quantization by Minimizing the Maximum Intercluster Distance," ACM Transactions on Graphics, vol. 16, no. 3, pp. 260-276, 1997.
[23] D. Zhengguo, "A New Visibility Metric For Haze Images," [Online]. Available： http：//www.mathworks.com/matlabcentral/fileexchange/33529-a-new-visibility-metric-for-haze-images.
[24] D. G. Lowe, "Distinctive Image Features from Scale-Invariant Keypoints," International Journal of Computer Vision, vol. 60, no. 2, pp. 91-110, 2004.
[25] C.-Y. Li, J.-C. Guo, R.-M. Cong, Y.-W. Pang and B. Wnag, "Underwater Image Enhancement by Dehazing With Minimum Information Loss and Histogram Distribution Prior," IEEE Transactions on Image Processing, vol. 25, no. 12, pp. 5664-5677, Dec. 2016.