電腦視覺演算法像是物件偵測、影像辨識技術已經被導入許多日常生活的應用中，然而這些演算法通常沒有辦法在實際情形都運作的很完善，原因在於實際景象中，無法預測的影像品質損害情形時常會發生，例如：雜訊、曝光度，以及多變的氣候情況。曝光度不足的影像會因為影像品質下降而嚴重地影響許多電腦視覺演算法的表現。
本論文提出一個基於UNet的低光源影像增強演算法來強化影像中陰暗的部分，使其還原為正常光源的影像。一個重新設計過的深度可分離卷積將有助於提取更豐富的特徵圖，同時可大量減少模型的參數及計算量。一個結合通道注意機制的UNet可以在進行訓練時，針對特徵圖做進一步的強化，以提升訓練過程的準確度。實驗結果顯示，本論文的方法在比較的方法當中，得到了更自然的紋理與細節。

Computer vision algorithms like object detection, image recognition technology have changed our lives significantly in various applications. However, these algorithms cannot usually have good performance in practical applications since the unpredictable degradations often occur in a realistic scene, for instance, noise, illumination, and severe weather conditions. Underexposed images will seriously affect the performance of many computer vision algorithms due to the degradation of image quality.
In this Thesis, a low-light image enhanced algorithm based on UNet is proposed to enhance the dark part of the image and restore it to normal light. A proposed depthwise separable convolution will help extract richer feature maps and can greatly reduce model parameters and calculations at the same time. An UNet combined with the channel attention mechanism can further strengthen the feature map during training to improve the accuracy of the training process. Experimental results show that the proposed method gets more natural textures and details compared with available methods.

[1] Q. Wang and R. K. Ward, "Fast Image/Video Contrast Enhancement Based on Weighted Thresholded Histogram Equalization," in IEEE Transactions on Consumer Electronics, vol. 53, no. 2, pp. 757-764, May 2007.
[2] T. Celik and T. Tjahjadi, "Contextual and Variational Contrast Enhancement," in IEEE Transactions on Image Processing, vol. 20, no. 12, pp. 3431-3441, Dec. 2011.
[3] Chulwoo Lee, Chul Lee, and Chang-Su Kim, “Contrast enhancement based on layered difference representation,” in Proc. International Conference on Image Processing (ICIP), Orlando, USA, pp. 965-968, Sept.-Oct. 2012.
[4] Chulwoo Lee, Chul Lee, and Chang-Su Kim, “Contrast enhancement based on layered difference representation of 2D histograms,” IEEE Transactions on Image Processing, vol. 22, no. 12, pp. 5372-5384, Dec. 2013.
[5] E. H. Land, ``The Retinex,'' Amer. Sci., vol. 52, no. 2, pp. 247_264, 1964.
[6] E. H. Land, ``The Retinex theory of color vision,'' Sci. Amer., vol. 237, no. 6, pp. 108_128, Dec. 1977.
[7] S. Wang, J. Zheng, H. Hu, and B. Li, "Naturalness Preserved Enhancement Algorithm for Non-Uniform Illumination Images," in IEEE Transactions on Image Processing, vol. 22, no. 9, pp. 3538-3548, Sept. 2013.
[8] X. Fu, Y. Liao, D. Zeng, Y. Huang, X. Zhang, and X. Ding, "A Probabilistic Method for Image Enhancement with Simultaneous Illumination and Reflectance Estimation," in IEEE Transactions on Image Processing, vol. 24, no. 12, pp. 4965-4977, Dec. 2015.
[9] X. Fu, D. Zeng, Y. Huang, X.-P. Zhang, and X. Ding, “A Weighted Variational Model for Simultaneous Reflectance and Illumination Estimation,” in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, Jun. 2016.
[10] X. Guo, “LIME: A Method for Low-light IMage Enhancement,” in Proceedings of the 2016 ACM on Multimedia Conference - MM ’16, Amsterdam, The Netherlands, 2016.
[11] X. Guo, Y. Li, and H. Ling, “LIME: Low-Light Image Enhancement via Illumination Map Estimation,” IEEE Trans. on Image Process., vol. 26, no. 2, pp. 982–993, Feb. 2017.
[12] F. Lv, F. Lu, J. Wu, and C. Lim, ``MBLLEN: Low-light image/video enhancement using CNNs,'' in Proc. Brit. Mach. Vis. Conf., pp. 1_13, 2018.
[13] W. Wang, C. Wei, W. Yang, and J. Liu, "GLADNet: Low-Light Enhancement Network with Global Awareness," 2018 13th IEEE International Conference on Automatic Face & Gesture Recognition.
[14] Y. Jiang, X. Gong, D. Liu, Y. Cheng, C. Fang, X. Shen, J. Yang,P. Zhou, and Z. Wang, ``EnlightenGAN: Deep light enhancement without paired supervision,'' arXiv:1906.06972,2019.
[15] O. Ronneberger, P. Fisher, T. Brox, “U-Net: Convolutional Networks for Biomedical Image Segmentation”, in MICCAI, 2015.
[16] Chunle Guo, Chongyi Li, Jichang Guo, Chen Change Loy, Junhui Hou, Sam Kwong, Runmin Cong, “Zero-Reference Deep Curve Estimation for Low-Light Image Enhancement,” in 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA, pp. 1777–1786, Jun. 2020.
[17] Q. Mu, Y. Wei, and J. Li, "Research on the improved Retinex algorithm for low illumination image enhancement", J. Harbin Eng. Univ., vol. 39, no. 12, pp. 1-7, Jan. 2018.
[18] D. J. Jobson, Z. Rahman and G. A. Woodell, "Properties and performance of a center/surround retinex," in IEEE Transactions on Image Processing, vol. 6, no. 3, pp. 451-462, March 1997.
[19] Z. Rahman, D. J. Jobson, and G. A. Woodell, "A Multiscale Retinex for Colour Rendition and Dynamic Range Compression," SPIE International Symposium on Optical Science, Engineering and Instrumentation, Applications of Digital Image Processing XIX, Proceedings SPIE 2825, Andrew G. Tescher, ed., 1996.
[20] Z. Rahman, D. J. Jobson and G. A. Woodell, "Multi-scale retinex for color image enhancement," Proceedings of 3rd IEEE International Conference on Image Processing, Lausanne, Switzerland, 1996.
[21] D. J. Jobson, Z. Rahman and G. A. Woodell, "A multiscale retinex for bridging the gap between color images and the human observation of scenes," in IEEE Transactions on Image Processing, vol. 6, no. 7, pp. 965-976, July 1997.
[22] M. Yang, X. Nie, and R. W. Liu, ``Coarse-to-_ne luminance estimation for low-light image enhancement in maritime video surveillance,'' inProc. IEEE Intell. Transp. Syst. Conf. (ITSC), Auckland, New Zealand, pp. 299_304, Oct. 2019
[23] L. Tao, C. Zhu, G. Xiang, Y. Li, H. Jia, and X. Xie, "LLCNN: A convolutional neural network for low-light image enhancement," 2017 IEEE Visual Communications and Image Processing (VCIP), St. Petersburg, FL, 2017.
[24] C. Wei, W. Wang, W. Yang, and J. Liu, "Deep Retinex Decomposition for Low-Light Enhancement." BMVC 2018, Oral Presentation.
[25] Y. Zhang, J. Zhang, and X. Guo, “Kindling the Darkness: A Practical Low-light Image Enhancer,” arXiv preprint arXiv:1905.04161 [cs], May 2019.
[26] Zhang, Y., Guo, X., Ma, J. et al. “Beyond Brightening Low-light Images”. Int J Comput Vis 129, 1013–1037, 2021.
[27] Auto-association by multilayer perceptrons and singular value decomposition, Bourlard, etc, 1988.
[28] Reducing the dimensionality of data with neural networks, Geoffrey Hinton, etc, 2006.
[29] J. Long, E. Shelhamer and T. Darrell, “Fully convolutional networks for semantic segmentation,” 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, MA, pp. 3431-3440, 2015.
[30] Ozan Oktay, Jo Schlemper, Loic Le Folgoc, Mattew Lee, Mattias Heinrich, Kazunari Misawa, Kensaku Mori, Steven McDonagh, Nils Y Hammerla, Bernhard Kainz, Ben Glocker, Daniel Rueckert, “Attention U-Net: Learning Where to Look for the Pancres”, in MIDL, 2018.
[31] Jie Hu, Li Shen, and Gang Sun, “Squeeze-and-excitation networks,” arXiv preprint arXiv:1709.01507, 2017.
[32] Andrew G. Howard, Menglong Zhu, Bo Chen Dmitry Kalenichenko, Weijun Wang, Tobias Weyand, Marco Andreetto, Hartwig Adam. “MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications”, arXiv preprint arXiv:1704.04861v1, 2017.
[33] K. Simonyan and A. Zisserman. Very deep convolutional networks for large-scale image recognition. In ICLR, 2015.
[34] J. Johnson, A. Alahi, and L. Fei-Fei. Perceptual losses for real-time style transfer and super-resolution. 2016.
[35] K. Ma, K. Zeng, and Z. Wang, “Perceptual quality assessment for multi-exposure image fusion,” IEEE Trans. Image Process., vol. 24, no. 11, pp. 3345–3356, Nov. 2015.
[36] S. Wang, J. Zheng, H.-M. Hu, and B. Li, “Naturalness preserved enhancement algorithm for non-uniform illumination images,” IEEE Trans. Image Process., vol. 22, no. 9, pp. 3538–3548, Sep. 2013.
[37] C. Lee, C. Lee, and C.-S. Kim, “Contrast enhancement based on layered difference representation of 2D histograms,” IEEE Trans. Image Process., vol. 22, no. 12, pp. 5372–5384, Dec. 2013.
[38] Q. Chen and D. Wu, “Image denoising by bounded block matching and 3D filtering,” Signal Process., vol. 90, no. 9, pp. 2778–2783, Sep. 2010.
[39] D.-T. Dang-Nguyen, C. Pasquini, V. Conotter, and G. Boato, “RAISE: A raw images dataset for digital image forensics,” in Proc. 6th ACM Multimedia Syst. Conf., pp. 219–224, Mar. 2015.
[40] Mark Sandler, Andrew G. Howard, Menglong Zhu, Andrey Zhmoginov, Liang-Chieh Chen. “MobileNetV2: Inverted Residuals and Linear Bottlenecks”, arXiv:1801.04381.
[41] Andrew G. Howard, Mark Sandler, Grace Chu, Liang-Chieh Chen, Bo Chen, Mingxing Tan, Weijun Wang, Yukun Zhu, Ruoming Pang, Vijay Vasudevan, Quoc V. Le, Hartwing Adam, “Searching for MobileNetV3”, arXiv: 1905.02244.
[42] A. Mittal, R. Soundararajan and A. C. Bovik, "Making a “Completely Blind” Image Quality Analyzer," in IEEE Signal Processing Letters, vol. 20, no. 3, pp. 209-212, doi: 10.1109/LSP.2012.2227726, March 2013.
[43] Prajit Ramachandran, Barret Zoph, Quoc V. Le, “Searching for Activation Functions”, arXiv: 1710.05941.
[44] Diganta Misra, “Mish: A Self Regularized Non-Monotonic Activation Function”, arXiv: 1908.08681.
[45] Wenhan Yang, Wenjing Wang, Ahofeng Huang, Shiqi Wang, Jiaying Liu, “Sparse Gradient Regularized Deep Retinex Network for Robust Low-Light Image Enhancement”, IEEE Transactions on Image Processing., vol. 30, pp. 2072-2086, Jan. 2021.
[46] Li-Wen Wang, Zhi-Song Liu, Wan-Chi Siu, and Daniel P. K. Lun, “Lightening Network for Low-Light Image Enhancement”, IEEE Transactions on Image Processing., vol. 29, pp. 7984-7996, Jul. 2020.
[47] Wencheng Wang, Xiaojin Wu, Xiaohui Yuan, and Zairui Gao, “An Experiment-Based Review of Low-Light Image Enhancement Methods”, IEEE Access, vol. 8, pp. 87884 – 87917, May. 2020.
[48] K. Sharifi and A. Leon-Garcia, “Estimation of shape parameter for generalized Gaussian distributions in subband decompositions of video,” IEEE Trans. Circuits Syst. Video Technol., vol. 5, no. 1.pp. 52-56, 1995.
[49] N. E. Lasmar, Y. Stitou, and Y. Berthoumieu, “Multiscale skewed heavy tailed model for texture analysis,” in IEEE Int. Conf. Image Process, pp. 2281-2284, 2009.
[50] Land, E.H., McCann,J.J.: Lightness and retinex theory. J. Opt. Soc. Am. 61, 1-11(1971)
[51] McCann, J.J.: Lessons learned from Mondrians applied to real images and color gamuts. In: Proceedings of IS&T/SID Color Imaging Conference, vol. 7, pp. 1–8. Scottsdale (1999).
[52] McCann, J.J.: Capturing a black cat in shade: past and present of retinex color appearance models. J. Electron. Imaging 13, 36–47 (2004).
[53] McCann, J.J., McKee, S., Taylor, T.: Quantitative studies in retinex theory, a comparison between theoretical predictions and observer responses to color Mondrian experiments. Vision Res. 16, 445–458 (1976).
[54] McCann, J.J.: Simultaneous Contrast and Color Constancy: Signatures of Human Image Processing, Chapter 6 in Color Perception: Philosophical, Psychological, Artistic, and Computational Perspectives, Vancouver Studies in Cognitive Studies: Davis, A., Davis, S. (eds.) pp. 87–101. Oxford University Press, Vancouver 2000.
[55] Frankle, J., McCann, J.: Method and apparatus of lightness imaging. US Patent 4,384,336 (1983).
[56] McCann, J., Rizzi, A.: The Art and Science of HDR Imaging, pp. 221–375. Wiley, Chichester (2012).
[57] Qing Zhang, Yongwei Nie, and Wei-Shi Zheng, “Dual Illumination Estimation for Robust Exposure Correction” arXiv preprint arXiv:1910.13688 [cs], Oct. 2019.
[58] Xu L., Yan Q., Xia Y., Jia J.: Structure Extraction from Texture via Relative Total Variation. ACM Transactions on Graphics 31, 139, 6 (2012).
[59] Mertens T., Kautz J., Van Reeth F.: Exposure Fusion: A Simple and Practical Alternative to High Dynamic Range Photography. Computer Graphics Forum 28, 161-171, 1 (2009).
[60] V. Bychkovsky, S. Paris, E. Chan, and F. Durand, “Learning photographic global tonal adjustment with a database of input/output image pairs,” in Processings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.97-104, 2011.