近年來由於機器學習的盛行，許多電腦視覺的課題都有了重大的突破 特別是對於生活所需的一切，我們都可以透過先進的技術去做判別，譬如人臉檢測、車牌偵測、物件識別等等，除了有賴良好硬體設備取得品質良好的影像外，有時受限於困難，例如燈光、迷霧等，會影 響 照片的品質，甚至是其中的細節流失 ，對於判斷以及檢測都會有相當大的影響 因此近年來對於低照度 光源 的影像還原方法成為了重要的電腦視覺課題。
本文提出一個結合傳統影像處理以及機器學習的方法， 將冗長的影像處理過程，藉由模型的訓練，達成相同效果。影像處理方法基於色溫板在不同照度下的資料做統計模型，配合降噪處理，所輸出圖片能夠 改善 LOE以及 VIF值 約 5%和 2%，並能夠解決以往影像處理在低光源還原時容易 產生 亮區過度曝光的問題，使其結果能夠滿足人類視覺系統裡較為符合的常態 直 方圖分布狀況，與以往機器學習主要在模擬集上做訓練，本文提出方法直接在低光源數據集上做訓練， 從模型預測產生的結果可以發現對低光源處的亮度還原以及原圖的亮區的細節保留都有不錯結果 。

In recent years, due to the popularity of machine learning, many computer vision topics have made major breakthroughs. Especially for everything that is needed in life, we can use advanced technology to make judgments, such as face detection, license plate detection, and objects recognition, etc. Some awful conditions, such as light and fog, will affect the quality of the photo, some of the information in the image being loss and make the detection difficult to complete. To overcome this impact, except using expensive hardware, low light image restoration methods have become one of the main issues in computer vision.
This thesis proposes a method that combines traditional image process and deep learning, which reduces the computational time of lengthy image process by training model. Unlike the traditional deep learning methods, the training data of our method is based on the low light images, not the simulation darkness data.
The simulation results show improvement on LOE and VIF values of 5% and 2%. It also solves the over enhancement problem in bright areas. Moreover, the histogram distribution shows that the obtained result is more suitable for the human visualize system. The experiment results demonstrate that the proposed approach performs better in both subjective and objective qualities.

[1] Y. Mei and G. Qiu, “Recovering High Dynamic Range Radiance Maps from Photographs Revisited: A Simple and Important Fix,” 2013 Seventh International Conference on Image and Graphics, Qingdao, Oct. 2013, pp. 23-28.
[2] H. Heinz, S. Markus, “CMOS vs. CCD sensors in speckle interferometry,” Optics and Laser Technology - OPT LASER TECHNOL. vol. 35, Dec. 2003, pp.587-595.
[3] Z. Wu and X. Xiao, “Study on Histogram Equalization,” International Symposium on Intelligence Information Processing and Trusted Computing, Hubei, China, Oct. 2011, pp. 177-179.
[4] J. A. Stark, “Adaptive image contrast enhancement using generalizations of histogram equalization,” IEEE Transactions on Image Processing, vol. 9, no. 5, May 2000, pp. 889-896.
[5] J. Tang, N. Isa, “Bi-histogram equalization using modified histogram bins,” Applied Soft Computing. vol. 55, Jun. 2017, pp. 31-43.
[6] M. Abdullah-Al-Wadud, M. H. Kabir, M. A. A. Dewan and O. Chae, “A Dynamic Histogram Equalization for Image Contrast Enhancement,” Consumer Electronics, IEEE Transactions, vol. 53, Jun. 2017, pp. 593-600.
[7] C. S. Josephus and S. Remya, “Multilayered Contrast Limited Adaptive Histogram Equalization Using Frost Filter,” 2011 IEEE Recent Advances in Intelligent Computational Systems, Trivandrum, Kerala, Sep. 2011, pp. 638-641.
[8] S. Chen and A. R. Ramli, “Minimum mean brightness error bi-histogram equalization in contrast enhancement,” IEEE Transactions on Consumer Electronics, vol. 49, no. 4, Nov. 2003, pp. 1310-1319.
[9] E. H. Land, “The Retinex theory of color vision,” Scientic American, vol. 237, no. 6, Dec. 1997, pp. 108-129.
[10] D. H. Choi, L. H. Jang, M. H. Kim and N. C. Kim, “Color image enhancement using single-scale retinex based on an improved image formation model,” European Signal Processing Conference, Lausanne, Switzerland, Aug. 2008.
[11] A. B. Petro, C. Sbert and J. M. Morel, “Multiscale Retinex,” Image Processing On Line, vol. 4, Apr. 2014, pp.71-88.
[12] X. Dong, G. Wang, Y. Pang, W. Li, J. Wen, W. Meng, Y. Lu, “Fast efficient algorithm for enhancement of low lighting video,” 2011 IEEE International Conference on Multimedia and Expo, Barcelona, Sep. 2011, pp. 1-6.
[13] X. Fu, D. Zeng, Y. Huang, X. Zhang and X. Ding, “A Weighted Variational Model for Simultaneous Reflectance and Illumination Estimation,” 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, Jun. 2016, pp. 2782-2790.
[14] C. Chen, Q. Chen, J. Xu and V. Koltun, “Learning to See in the Dark,” 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, May 2018, pp. 3291-3300.
[15] L. Kin, A. Akintayo, S. Sarkar, “LLNet: A Deep Autoencoder Approach to Natural Low-light Image Enhancement,” Pattern Recognition. vol. 61, Jan. 2017, pp. 650-662.
[16] 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, Dec. 2017, pp. 1-4,
[17] C. Li, J. Guo, F. Porikli and Y. Peng, “LightenNet: A Convolutional Neural Network for weakly illuminated image enhancement,” Pattern Recognition Letters, vol. 104, Jan. 2018, pp. 15-22.
[18] J. Cai, S. Gu and L. Zhang, "Learning a Deep Single Image Contrast Enhancer from Multi-Exposure Images," IEEE Transactions on Image Processing, vol. 27, no. 4, Apr. 2018, pp. 2049-2062.
[19] L. Jiang, Y. Jing, S. Hu, B. Ge and W. Xiao, “Deep Refinement Network for Natural Low-Light Image Enhancement in Symmetric Pathways,” Symmetry, vol. 10, Oct. 2018, pp. 491-508.
[20] Y. Guo, X. Ke, J. Ma and J. Zhang, “A Pipeline Neural Network for Low-Light Image Enhancement,” IEEE Access, vol. 7, Jan. 2019, pp. 13737-13744.
[21] Y. P. Loh, X. Liang, C. S. Chan, “Low-light image enhancement using Gaussian Process for features retrieval,” Signal Processing: Image Communication. vol. 74, May 2019, pp. 175-190.
[22] S. Ma, H. Ma, Y. Xu, S. Li, C. Lv and M. Zhu, “A Low-Light Sensor Image Enhancement Algorithm Based on HSI Color Model,” Sensors, vol. 18, Oct. 2018 pp. 3583-3599.
[23] R. Liu, S. Cheng, L. Ma, X. Fan and Z. Luo, “Deep Proximal Unrolling: Algorithmic Framework,” Convergence Analysis and Applications. IEEE Transactions on Image Processing, vol. 28, no. 10, May 2019, pp. 5013-5026.
[24] W. Ren et al., “Low-Light Image Enhancement via a Deep Hybrid Network,” IEEE Transactions on Image Processing, vol. 28, no. 9, Sep. 2019, pp. 4364-4375.
[25] C. M. Tsai, Z. M. Yeh, “Contrast enhancement by automatic and parameter-free piecewise linear transformation for color images,” Consumer Electronics, IEEE Transactions, vol. 54, issue 2, Jul. 2008, pp. 213 - 219.
[26] S. Wang, J. Zheng, H. M. Hu, B. Li, “Naturalness Preserved Enhancement Algorithm for Non-Uniform Illumination Images”. IEEE transactions on image processing: a publication of the IEEE Signal Processing Society, vol 22, 2017, pp. 3538-3548.
[27] X. Guo, Y. Li, H. Ling, “LIME: Low-Light Image Enhancement via Illumination Map Estimation,” IEEE Transactions on Image Processing, vol. 26, issue 2, Dec. 2016, pp. 982-993.
[28] M. Li, J. Liu, W. Yang, X. Sun and Z. Guo, “Structure-Revealing Low-Light Image Enhancement Via Robust Retinex Model,” IEEE Transactions on Image Processing, vol. 27, no. 6, Jun. 2018, pp. 2828-2841.
[29] E. Ullah, R. Nawaz and J. Iqbal, “Single image haze removal using improved dark channel prior,” 2013 5th International Conference on Modelling, Identification and Control (ICMIC), Cairo, Oct. 2013, pp. 245-248.
[30] X. Zhang, P. Shen, L. Luo, L. Zhang and J. Song, “Enhancement and noise reduction of very low light level images,” Proceedings - International Conference on Pattern Recognition, Nov. 2012, pp. 2034-2037.
[31] S. Yu and H. Zhu, "Low-Illumination Image Enhancement Algorithm Based on a Physical Lighting Model," IEEE Transactions on Circuits and Systems for Video Technology, vol. 29, no. 1, Jan. 2019, pp. 28-37.
[32] Q. Xu, H. Jiang R. Scopigno and M. Sbert, “A novel approach for enhancing very dark image sequences,” Signal Processing, vol. 103, Oct. 2014, pp. 309-330.
[33] M. Kim, D. Park, D. K. Han and H. Ko, “A novel approach for denoising and enhancement of extremely low-light video,” IEEE Transactions on Consumer Electronics, vol. 61, no. 1, Feb. 2015, pp. 72-80.
[34] S. W. Lee, V. Maik, J. Jang, J. Shin and Paik, J. Paik, “Noise-adaptive spatio-temporal filter for real-time noise removal in low light level images,” IEEE Transactions on Consumer Electronics. vol. 51. May 2005, pp. 648 - 653.
[35] Y. Ren, Z. Ying, T. Li, and G. Li, “LECARM: Low-Light Image Enhancement using Camera Response Model,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 29, issue. 4, Apr. 2018, pp. 968-981.
[36] L. Florea, C. Florea and C. Ionascu, “Avoiding the Deconvolution: Framework Oriented Color Transfer for Enhancing Low-Light Images,” 2016 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Dec. 2016, pp.936-944.
[37] L. Florea and C. Florea, “Directed color transfer for low-light image enhancement,” Digital Signal Processing, vol. 93, Jul. 2019, pp. 1-12.
[38] H. Zhu, Francis H.Y. Chan and F.K. Lam, “Image contrast enhancement by constrained local histogram equalization,” Computer Vision and Image Understanding, vol. 73, Issue 2, Feb. 1999, pp. 281-290.
[39] K. Bernardin and R. Stiefelhagen, “Evaluating Multiple Object Tracking Performance: The CLEAR MOT Metrics,” Journal on Image Video Processing, vol. 24309, May. 2008, pp.1-10.
[40] L. Jin, S. Satoh and M. Sakauchi, “A novel adaptive image enhancement algorithm for face detection,” Proceedings of the 17th International Conference on Pattern Recognition, ICPR 2004, Cambridge, vol. 42004, Nov. 2004, pp. 843-848.
[41] L. Tommer, D. Cohen-Or, G. Dror and D. Lischinski, “Data-driven enhancement of facial attractiveness,” ACM Trans. Graph, vol. 27, Aug. 2008.
[42] R. Girshick, J. Donahue, T. Darrell and J. Malik, “Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation,” 2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, 2014, pp. 580-587.
[43] D. N. Diana Rexiline and D. J. Anusmina, “Fusion and restoration of multifocus image using sparse representation,” IEEE-International Conference On Advances In Engineering, Science And Management (ICAESM -2012), Nagapattinam, Tamil Nadu, Nov. 2012, pp. 291-296.
[44] J. Kim, J. K. Lee and K. M. Lee, “Accurate Image Super-Resolution Using Very Deep Convolutional Networks,” Proc. of IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, Jun. 2016.
[45] L. Zhuo, X. Hu, J. Li, J. Zhang and X. Li, “A Naturalness-Preserved Low-Light Enhancement Algorithm for Intelligent Analysis,” Chinese Journal of Electronics, vol. 28, no. 2, Mar. 2019, pp. 316-324.
[46] K. Leng, “An improved non-local means algorithm for image denoising,” 2017 IEEE 2nd International Conference on Signal and Image Processing (ICSIP), Singapore, Mar. 2017, pp. 149-153.
[47] T. H. Lee and B. C. Song, “De-noising algorithm using sparse 3D transform-domain collaborative filtering and adaptive soft thresholding,” 2011 IEEE 15th International Symposium on Consumer Electronics (ISCE), Singapore, 2011, pp. 128-131.
[48] D. Yang and J. Sun, “BM3D-Net: A Convolutional Neural Network for Transform-Domain Collaborative Filtering,” IEEE Signal Processing Letters, vol. 25, no. 1, Jan. 2018, pp. 55-59.
[49] K. Zhang, W. Zuo, Y. Chen, D. Meng and L. Zhang, “Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising,” IEEE Transactions on Image Processing, vol. 26, no. 7, Jul. 2017, pp. 3142-3155.
[50] X. Ji, Y. Lu and L. Guo, “Image Super-Resolution with Deep Convolutional Neural Network,” 2016 IEEE First International Conference on Data Science in Cyberspace (DSC), Changsha, Mar. 2016, pp. 626-630.
[51] P. Aswathy, Siddhartha and D. Mishra, “Deep GoogLeNet Features for Visual Object Tracking,” 2018 IEEE 13th International Conference on Industrial and Information Systems (ICIIS), Rupnagar, India, Nov. 2018, pp. 60-66.
[52] Y. Loh and C. Chan, “Getting to Know Low-light Images with The Exclusively Dark Dataset,” Computer Vision and Image Understanding. vol. 178, Jan. 2019, pp. 30-42.
[53] H. Zhao, O. Gallo, I. Frosio and J. Kautz, “Loss Functions for Image Restoration with Neural Networks,” IEEE Transactions on Computational Imaging, vol. 3, no. 1, Mar. 2017, pp. 47-57.
[54] Z. Ying, G. Li, Y. Ren, R. Wang, W. Wang. “A New Image Contrast Enhancement Algorithm Using Exposure Fusion Framework,” International Conference on Computer Analysis of Images and Patterns, vol. 10425, Jul. 2017, pp.36-46.