在現今生活中，由於攝影裝置的日漸普及，各種影像處理演算法的技術顯得越發重要，例如：低光度影像之增強。低光度影像增強的技術已被應用在許多領域，如夜間監視系統、道路安全監控系統等，但在目前已被提出的低光度影像增強演算法中，大部份的演算法都缺少了可適性，例如：針對能見度極差的超低光度影像可以得到不錯的增強結果，而能見度稍加提升的影像，則可能發生過度增強的現象，導致極差的結果。本論文提出一個具備可適性的低光度影像增強演算法，無論是影像中的亮部或暗部，亦或是極度欠缺光源的超低光度影像、光源稍弱的低光度影像乃至一般夜間影像，都可以獲得理想的增強結果。
本論文所提出之低光度影像增強演算法中，首先對低光度影像取其負片，並根據其影像資訊進行一全自動化且具可適性的影像除霧演算法，以達到低光度增強之目的。此外，不同於白天影像，在低光度環境中所拍攝的影像，其雜訊含量會大幅提升，而夜間影像中由各種人造光源（如車燈、路燈等）產生之光暈，也可能影響到部分影像之能見度，以及攝影器材可能產生之影像色偏等等問題，所以本論文接著對增強後的影像依序進行去雜訊、消除光暈、色彩校正及對比度強化等運算，以提升增強結果影像之品質。如實驗結果所示，本論文所提出之低光度影像增強演算法，不僅能有效解決過度增強以及強化不足的問題，且對於夜間影像中的光暈、雜訊及色彩偏移等問題，也能有效達到消除與校正。

Low-light image enhancement technology has been applied in many areas, such as night surveillance systems, road safety monitoring systems, etc. However, most of the existing low-light image enhancement algorithms are lack of adaptability. For example, for very low-light level images with poor visibility, considerably good results can be obtained. But images with visibility slightly improved, poor results which are over-enhanced may be produced. Therefore, an adaptive low-light image enhancement algorithm is proposed in this paper. Whether these are very low-light level images with extreme lack of illumination, slightly weak luminosity images or even general night images, desirable enhancement results can be obtained.
The proposed low-light image enhancement algorithm first takes the inverse image from a low-light image, then according to its information, an automated and adaptive image dehazing operation is conducted to achieve the purpose of low-light enhancement. In addition, different from daytime images, images obtained in low-light environment, the portion of noise will be greatly increased, and the glow generated at night by a variety of artificial light sources (such as vehicular and street lamps, lights) can affect the visibility of some areas of the image, along with image color shift that may be generated by photographic equipment. Therefore, in this paper, the resultant images of low-light enhancement are subsequently processed by noise reduction, glow removal, and color calibration. As shown in the experimental results, the proposed low-light image enhancement algorithm can not only effectively solve the problem of over- and under-enhancement, it can also effectively remove glow, reduce noise, and correct color in nighttime images.

學位論文審定書 i
誌謝 ii
中文摘要 iii
英文摘要 iv
第一章　緒論 1
1.1大氣散射模型 1
1.2影像除霧演算法 2
1.3暗通道先驗 3
第二章　相關研究 5
2.1低光度影像增強 5
2.2低光度影像與含霧影像之相關性 5
2.3色彩校正 7
第三章　研究方法 11
3.1低光度影像增強 11
3.1.1透射率估計 12
3.1.2大氣光照估計 14
3.2消除雜訊 15
3.3消除光暈 16
3.4色彩校正 18
第四章　實驗結果 21
4.1低光度影像增強 21
4.2消除雜訊 22
4.3消除光暈 25
4.4色彩校正 27
4.5客觀評估 28
第五章　結論與未來展望 31
參考文獻 32

[1] R. Tan, “Visibility in Bad Weather from a Single Image,” IEEE Conf. on Computer Vision and Pattern Recognition, pp. 1-8, Jun. 2008.
[2] R. Fattal, “Single Image Dehazing,” ACM Trans. on Graphics, Vol. 27, No. 3, pp. 1–8, 2008.
[3] K. He, J. Sun, and X. Tang, “Single Image Haze Removal using Dark Channel Prior,” IEEE Conf. on Computer Vision and Pattern Recognition, pp. 1956-1963, Jun. 2009.
[4] H. Ngo, L. Tao, M. Zhang, A. Livingston, and V. Asari, “A Visibility Improvement System for Low Vision Drivers by Nonlinear Enhancement of Fused Visible and Infrared Video,” IEEE Computer Society Conf. on Computer Vision and Pattern Recognition Workshops, pp. 25-32, 2005.
[5] S.M. Pizer, et al., “Adaptive Histogram Equalization and Its Variations,” Computer Vision, Graphics, and Image Processing, Vol. 39, No. 3, pp. 355-368, Sep. 1987.
[6] Z. Rahman, D.J. Jobson, and G.A. Woodell, “Multi-scale Retinex for Color Image Enhancement,” IEEE Intl. Conf. on Image Processing, Vol. 3, pp. 1003-1006, 1996.
[7] X. Fu, D. Zeng, Y. Huang, X. Ding, and X.P. Zhang, “A Variational Framework for Single Low Light Image Enhancement using Bright Channel Prior,” IEEE Global Conf. on Signal and Information Processing, pp. 1085–1088, 2013.
[8] T. Arici, S. Dikbas, and Y. Altunbasak, “A Histogram Modification Framework and Its Application for Image Contrast Enhancement,” IEEE Trans. on Image Processing, pp. 1921-1935, 2009.
[9] T. Celik, “Two-dimensional Histogram Equalization and Contrast Enhancement,” Pattern Recognition, pp. 3810-3824, 2012.
[10] M.N. Do, and M. Vetterli, “The Contourlet Transform: An Efficient Directional Multi-resolution Image Representation,” IEEE Trans. on Image Processing, Vol. 14, No. 12, pp. 2091–2106, Dec. 2005.
[11] Y.B. Rao, and L. Chen, “An Efficient Contourlet Transform-based Algorithm for Video Enhancement,” Journal of Information Hiding and Multimedia Signal Processing, pp. 282-293, 2011.
[12] A. Saleem, A. Beghdadi, and B. Boashash, “Image Fusion-based Contrast Enhancement,” EURASIP Journal on Image and Video Processing, pp. 1-17, 2012.
[13] X. Dong, Y. Pang, and J. Wen, “Fast Efficient Algorithm for Enhancement of Low Lighting Video,” IEEE Intl. Conf. on Multimedia and Expo, pp. 1-6, 2011.
[14] X.S. Jiang, and H.X. Yao, “Night Video Enhancement using Improved Dark Channel Prior,” IEEE Intl. Conf. on Image Processing, pp. 553–557, Sep. 2013.
[15] X. Zhang, P. Shen, L. Luo, L. Zhang, and J. Song, “Enhancement and Noise Reduction of Very Low Light Level Images,” IEEE Intl. Conf. on Pattern Recognition, pp. 2034-2037, 2012.
[16] J. Im, D. Kim, J. Jung, T.C. Kim, and J. Paik, “Dark Channel Prior Based White Point Estimation for Automatic White Balance,” IEEE Intl. Conf. on Consumer Electronics, pp. 123-124, 10-13 Jan. 2014.
[17] L. Li, R. Wang, W. Wang, and W. Gao, “A Low-light Image Enhancement Method for Both Denoising and Contrast Enlarging,” IEEE Intl. Conf. on Image Processing, 2015, pp. 3730–3734, 2015.
[18] H. Malm, M. Oskarsson, E. Warrant, P. Clarberg, J. Hasselgren, and C. Lejdfors, “Adaptive Enhancement and Noise Reduction in Very Low Light-Level Video,” IEEE Intl. Conf. on Computer Vision, pp. 1–8, 2007.
[19] J. Zhang, Y. Cao, and Z. Wang, “Nighttime Haze Removal Based on a New Imaging Model,” IEEE Intl. Conf. on Image Processing, 2014.
[20] Y. Li, R.T. Tan, and M.S. Brown, “Nighttime Haze Removal with Glow and Multiple Light Colors,” IEEE Intl. Conf. on Computer Vision, pp. 226-234, Dec. 2015.