隨著機器學習及人工智慧的快速發展，機器視覺中關於影像處理的技術被廣泛運用於各項領域，成功解決醫學及工程上的挑戰。在這個自動化的時代中，影像處理在電腦視覺中為不可或缺的一環，利用電腦應用於各類影像進行目標辨識與物件偵測，因此影像強化或是影像調整的技術至關重要。大多數可用的電腦視覺應用都是基於可見光相機，因此只能在正常光源的條件下使用。在低光源條件下，由於被拍攝物本身色彩因素導致影像資訊的分布過於集中於某些強度範圍內，造成噪點與對比度損失，而影像的品質下降，產生的低對比度和低品質影像不僅會降低電腦視覺以及影像處理演算法的效果，還會降低影像的視覺美感。因此，對比度增強是增加影像品質並使細節更加清楚的重要步驟。
本文提出了一種基於卷積神經網路的方法，利用成對影像之間像素差值，並將其稱為強化移動矩陣，並使用此強化移動矩陣作為標註資料，經過神經網路模型訓練後，預測出圖片的移動向量，藉此來得到低對比強化後的圖片，並與Fuzzy Automatic Cluster Enhancement（FACE）、Multi-Scale Retinex with Color Restoration（MSRCR）方法及Low-Light Image Enhancement（LLNet）進行比較，並且針對不同參數的影像進行實驗，依照均方根差、均方根對比、峰值訊噪比、結構相似性、及運算時間作為比較的標準，來突顯本文所提出的方法，其強化效果及運算效率皆有一定的水準。

With the rapid development of machine learning and artificial intelligence, image processing technology in computer vision has been widely used in various fields, successfully solving medical and engineering challenges. In this era of automation, image processing is an indispensable part of computer vision. Computers are applied to various types of images for target recognition and object detection. Therefore, image enhancement or image adjustment technology is very important. Most of the available computer vision applications are based on visible light cameras and thus can only be used under normal lighting conditions. Under low light source conditions, the distribution of image information is too concentrated in certain intensity ranges due to the color factor of the subject itself, resulting in noise and contrast loss. The quality of the image is degraded, resulting in low contrast and low quality images that not only reduce the effectiveness of computer vision and image processing algorithms, but also reduce the visual aesthetics of the image. Therefore, contrast enhancement is an important step in increasing image quality and making details clearer.
In this study, a method based on convolutional neural network is proposed, which uses the pixel difference between paired images, which is called a motion matrix, and uses this motion matrix as an annotation for low-contrast images. After the neural network model is trained, the motion vector of the picture is predicted to obtain the low-contrast enhanced picture, and compared with the Fuzzy Automatic Cluster Enhancement (FACE)、Multi-Scale Retinex with Color Restoration (MSRCR) and Low-Light Image Enhancement（LLNet）methods, and experimented with images with different parameters, according to the peak signal-to-noise ratio, structural similarity, root-mean-squared-error, root-mean-squared-contrast, and operation time as the comparison standards, the method proposed in this paper is highlighted, and its enhancement effect and operation efficiency have a certain level.

[1] R. Maini, H. Aggarwal, A comprehensive review of image enhancement techniques, arXiv preprint arXiv:1003.4053, 2010.
[2] S. M. Pizer, E. P. Amburn, J. D. Austin, R. Cromartie, A. Geselowitz, T. Greer, B. ter Haar Romeny, J. B. Zimmerman, K. Zuiderveld, Adaptive histogram equalization and its variations, Computer vision, graphics, and image processing, 39(3), 355-368, 1987.
[3] M. Abdullah-Al-Wadud, M. H. Kabir, M. A. A. Dewan, O. Chae, A dynamic histogram equalization for image contrast enhancement, IEEE Transactions on Consumer Electronics, 53(2), 593-600, 2007.
[4] E. D. Pisano, S. Zong, B. M. Hemminger, M. DeLuca, R. E. Johnston, K. Muller, M. P. Braeuning, S. M. Pizer, Contrast limited adaptive histogram equalization image processing to improve the detection of simulated spiculations in dense mammograms, Journal of Digital imaging, 11(4), 193-200, 1998.
[5] happychu, 路思義教堂, 2015, URL: https://www.flickr.com/photos/125549858@N07/20936333843/in/photolist-xU5bDa-p6TsT4-oPE7xq-oPEwBf-qV7XuC-a3eTTR-oPEwG4-oPEwMq-p57H9u-p79P8B-p77Now-oPDKCP-e9sWTG-p77N81-hpLsPd-e9ngqB-p77NpJ-hpLLxs-n6qAhe-oPDKVc-HtsZFG-p77Nuo-oPEwqT-p79NYP-hpLbZr-p79P7e-n6qz4n-p77Nfq-p6TsYp-n6qyop-n6qyDe-n6qup4-p77N7Q-p6Tt9e-nj5FKD-p57GKy-vFMHAv-p6Ttar-p6TsFk-p6TsYV-PJ1jcV-PForSQ-oPEwPZ-oPEwKS-oPE7HA-oPE7UN-voQfHN-5r4RKk-oPEwvo-vFptTt.
[6] E. H. Land, J. J. McCann, Lightness and retinex theory, Josa, 61(1), 1-11, 1971.
[7] D. J. Jobson, Z.-u. Rahman, G. A. Woodell, Properties and performance of a center/surround retinex, IEEE transactions on image processing, 6(3), 451-462, 1997.
[8] X. Fu, D. Zeng, Y. Huang, X.-P. Zhang, X. Ding, A weighted variational model for simultaneous reflectance and illumination estimation, Proceedings of the IEEE conference on computer vision and pattern recognition, 2782-2790, 2016.
[9] X. Guo, Y. Li, H. Ling, LIME: Low-light image enhancement via illumination map estimation, IEEE Transactions on image processing, 26(2), 982-993, 2016.
[10] Z.-u. Rahman, D. J. Jobson, G. A. Woodell, Multi-scale retinex for color image enhancement, Proceedings of 3rd IEEE international conference on image processing, 1003-1006, 1996.
[11] C. Wei, W. Wang, W. Yang, J. Liu, Deep retinex decomposition for low-light enhancement, arXiv preprint arXiv:1808.04560, 2018.
[12] P. T. Lin, B. R. Lin, Fuzzy automatic contrast enhancement based on fuzzy C-means clustering in CIELAB color space, 2016 12th IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications (MESA), 1-10, 2016.
[13] J. Kim, J. K. Lee, K. M. Lee, Accurate image super-resolution using very deep convolutional networks, Proceedings of the IEEE conference on computer vision and pattern recognition, 1646-1654, 2016.
[14] K. Zhang, W. Zuo, Y. Chen, D. Meng, L. Zhang, Beyond a gaussian denoiser: Residual learning of deep cnn for image denoising, IEEE transactions on image processing, 26(7), 3142-3155, 2017.
[15] A. Wahab, M. Mashor, Z. Salleh, S. A. Shukor, N. A. Rahim, F. M. Idris, H. Hasan, S. Noor, A neural network approach for contrast enhancement image, 2008 International Conference on Electronic Design, 1-4, 2008.
[16] K. G. Lore, A. Akintayo, S. Sarkar, LLNet: A deep autoencoder approach to natural low-light image enhancement, Pattern Recognition, 61, 650-662, 2017.
[17] X. Sun, L. Liu, J. Dong, Underwater image enhancement with encoding-decoding deep CNN networks, 2017 IEEE SmartWorld, Ubiquitous Intelligence & Computing, Advanced & Trusted Computed, Scalable Computing & Communications, Cloud & Big Data Computing, Internet of People and Smart City Innovation (SmartWorld/SCALCOM/UIC/ATC/CBDCom/IOP/SCI), 1-6, 2017.
[18] A. Shihavuddin, N. Gracias, R. Garcia, J. Escartin, R. B. Pedersen, Automated classification and thematic mapping of bacterial mats in the north sea, 2013 MTS/IEEE OCEANS-Bergen, 1-8, 2013.
[19] V. Brandou, A.-G. Allais, M. Perrier, E. Malis, P. Rives, J. Sarrazin, P.-M. Sarradin, 3D reconstruction of natural underwater scenes using the stereovision system IRIS, OCEANS 2007-Europe, 1-6, 2007.
[20] M. D. Zeiler, G. W. Taylor, R. Fergus, Adaptive deconvolutional networks for mid and high level feature learning, 2011 international conference on computer vision, 2018-2025, 2011.
[21] H. Noh, S. Hong, B. Han, Learning deconvolution network for semantic segmentation, Proceedings of the IEEE international conference on computer vision, 1520-1528, 2015.
[22] R. Fattal, Single image dehazing, ACM transactions on graphics (TOG), 27(3), 1-9, 2008.
[23] L. Tao, C. Zhu, J. Song, T. Lu, H. Jia, X. Xie, Low-light image enhancement using CNN and bright channel prior, 2017 IEEE International Conference on Image Processing (ICIP), 3215-3219, 2017.
[24] K. Dabov, A. Foi, V. Katkovnik, K. Egiazarian, Image processing: Algorithms and systems, neural networks, and machine learning, Image noise reduction with block-matching and 3D filtering, 6064, 354-365, 2006.
[25] Y. Jia, E. Shelhamer, J. Donahue, S. Karayev, J. Long, R. Girshick, S. Guadarrama, T. Darrell, Caffe: Convolutional architecture for fast feature embedding, Proceedings of the 22nd ACM international conference on Multimedia, 675-678, 2014.
[26] L. Li, R. Wang, W. Wang, W. Gao, A low-light image enhancement method for both denoising and contrast enlarging, 2015 IEEE international conference on image processing (ICIP), 3730-3734, 2015.
[27] W. Zhou, Image quality assessment: from error measurement to structural similarity, IEEE transactions on image processing, 13, 600-613, 2004.
[28] H. Yeganeh, Z. Wang, Objective quality assessment of tone-mapped images, IEEE Transactions on Image processing, 22(2), 657-667, 2012.
[29] Lenna, URL: https://zh.m.wikipedia.org/zh-tw/%E8%90%8A%E5%A8%9C%E5%9C%96.
[30] Y. James, [資料分析&機器學習] 第5.1講: 卷積神經網絡介紹(Convolutional Neural Network), 2017, URL: https://medium.com/jameslearningnote/%E8%B3%87%E6%96%99%E5%88%86%E6%9E%90-%E6%A9%9F%E5%99%A8%E5%AD%B8%E7%BF%92-%E7%AC%AC5-1%E8%AC%9B-%E5%8D%B7%E7%A9%8D%E7%A5%9E%E7%B6%93%E7%B6%B2%E7%B5%A1%E4%BB%8B%E7%B4%B9-convolutional-neural-network-4f8249d65d4f.
[31] S. Ruder, An overview of gradient descent optimization algorithms, arXiv preprint arXiv:1609.04747, 2016.
[32] W. S. McCulloch, W. Pitts, A logical calculus of the ideas immanent in nervous activity, The bulletin of mathematical biophysics, 5(4), 115-133, 1943.
[33] M.-C. Su, N. DeClaris, T.-K. Liu, Application of neural networks in cluster analysis, 1997 IEEE International Conference on Systems, Man, and Cybernetics. Computational Cybernetics and Simulation, 1-6, 1997.
[34] Y. LeCun, L. Bottou, Y. Bengio, P. Haffner, Gradient-based learning applied to document recognition, Proceedings of the IEEE, 86(11), 2278-2324, 1998.
[35] C.-H. Wei, C.-S. Fahn, The multisynapse neural network and its application to fuzzy clustering, IEEE transactions on neural networks, 13(3), 600-618, 2002.
[36] M. J. Christ, R. Parvathi, Fuzzy c-means algorithm for medical image segmentation, 2011 3rd international conference on electronics computer technology, 33-36, 2011.
[37] F. Taher, R. Sammouda, Lung cancer detection by using artificial neural network and fuzzy clustering methods, 2011 IEEE GCC conference and exhibition (GCC), 295-298, 2011.
[38] D. Raj, P. Mamoria, Comparative analysis of contrast enhancement techniques on different images, 2015 International Conference on Green Computing and Internet of Things (ICGCIoT), 27-31, 2015.
[39] M. Kaur, K. Verma, A novel hybrid technique for low exposure image enhancement using sub-imge histogram equilization and artificial neural network, 2016 International Conference on Inventive Computation Technologies (ICICT), 1-5, 2016.
[40] R. Kaur, S. Kaur, Comparison of contrast enhancement techniques for medical image, 2016 conference on emerging devices and smart systems (ICEDSS), 155-159, 2016.
[41] K. B. Singh, T. V. Mahendra, R. S. Kurmvanshi, C. R. Rao, Image enhancement with the application of local and global enhancement methods for dark images, 2017 International Conference on Innovations in Electronics, Signal Processing and Communication (IESC), 199-202, 2017.
[42] L. Tao, C. Zhu, G. Xiang, Y. Li, H. Jia, X. Xie, LLCNN: A convolutional neural network for low-light image enhancement, 2017 IEEE Visual Communications and Image Processing (VCIP), 1-4, 2017.
[43] E. S. Yelmanova, Y. M. Romanyshyn, Automatic histogram-based contrast enhancement for low-contrast images with small-sized objects, 2017 IEEE First Ukraine Conference on Electrical and Computer Engineering (UKRCON), 965-970, 2017.
[44] B. G. Han, H. S. Yang, H. G. Lee, Y. S. Moon, Low contrast image enhancement using convolutional neural network with simple reflection model, Advances in Science Technology and Engineering Systems Journal, 4(1), 159-164, 2019.
[45] C. Guo, C. Li, J. Guo, C. C. Loy, J. Hou, S. Kwong, R. Cong, Zero-reference deep curve estimation for low-light image enhancement, Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 1780-1789, 2020.
[46] J. Kalyani, M. Chakraborty, Contrast enhancement of MRI images using histogram equalization techniques, 2020 International Conference on Computer, Electrical & Communication Engineering (ICCECE), 1-5, 2020.
[47] H. Lee, K. Sohn, D. Min, Unsupervised low-light image enhancement using bright channel prior, IEEE Signal Processing Letters, 27, 251-255, 2020.
[48] J. Nalepa, M. Myller, Y. Imai, K.-i. Honda, T. Takeda, M. Antoniak, Unsupervised segmentation of hyperspectral images using 3-D convolutional autoencoders, IEEE Geoscience and Remote Sensing Letters, 17(11), 1948-1952, 2020.
[49] G. Li, Y. Yang, X. Qu, D. Cao, K. Li, A deep learning based image enhancement approach for autonomous driving at night, Knowledge-Based Systems, 213, 106617, 2021.