現今攝影設備的普及與各種影像處理演算法的技術不斷發展，為影像處理領域帶來了許多革新性的進展。其中，針對低光照環境下的影像處理技術，日益受到廣泛關注。然而，在現有的低光照影像增強演算法中，大部分都缺乏自適性，這是一個嚴重的問題。這些演算法可能在處理能見度極差的超低光照影像時表現良好，但對於能見度稍有改善或背光的影像，可能會產生過度增強或增強不夠的現象，從而導致影像質量的下降。這些問題對於一些關鍵性的應用場景，如夜間監控系統和影像識別等，尤其需要解決。
為了解決這些問題，本文提出了一種具備自適性 的低光照影像增強 演算法，該演算法可以為影像中的亮區和暗區以及各種低光照環境下的影像產生理想的增強效果 ，並將提出的演算法運用在 FPGA(Field Programmable Gate Array)，對低光照影像做實時的增強。該演算法 以大氣散射模型為基礎，首先將低光照影像取負片，採用負片的平均 飽和度和像素飽和度，自動調整增強參數，避免過度增強或增強不夠等問題，從而提高影像質量和影像細節的還原度。該演算法不僅在夜間監控和影像識別等領域有著廣泛的應用，同時也可以應用於其他低光照影像處理領域，為影像處理技術的發展提供了新的思路和方法。
實驗結果表明此方法 相較於其他的方法能得到更好的低光照增強影像，並且可以成功在FPGA上實時實現。從攝影鏡頭輸入並將低光照增強後的影像顯示輸出至螢幕上，最後的結果顯示本系統使用了 6,968個邏輯元件和 119,816 bits內部記憶體，功耗為 446.75mW 且 處理速度達到每秒 351張影像 (NTSC Input)。

Low-light image processing techniques have been developed with the increasing popularity of photography equipment, but many algorithms lack adaptability. This can result in over or insufficient enhancement for images with slightly improved visibility, causing a decrease in image quality. This is especially problematic for nighttime monitoring and image recognition applications.
This thesis proposes an adaptive low-light image enhancement algorithm that can produce ideal enhancement effects for bright and dark areas in images and various low-light environments. The proposed algorithm is implemented on an FPGA (Field Programmable Gate Array) for real-time enhancement of low-light images. Based on the atmospheric scattering model, the algorithm first takes the inverse image of the low-light image and uses the average saturation and saturation of the inverse image to automatically adjust the enhancement parameters, avoiding problems such as over-enhancement or insufficient enhancement, thereby improving the image quality and detail restoration. This algorithm has a wide range of applications not only in nighttime monitoring and image recognition but also in other low-light image processing fields, providing new ideas and methods for the development of image processing technology.
The experimental results demonstrate that this method can achieve better low-light image enhancement compared to other methods, and can be successfully implemented in real-time on an FPGA. It takes input from a camera lens and displays the enhanced low-light image on a screen. The final results show that the system uses 6968 logic elements and 119,816 bits of internal memory, with a power consumption of 446.75mW. Furthermore, it achieves a processing speed of 351 frames per second for NTSC input.

[1] X. Dong, Y. Pang, and J. Wen, "Fast efficient algorithm for enhancement of low lighting video," ACM SIGGRAPH 2010 Posters, pp. 1-1, 2010.
[2] S.M. Pizer, E.P. Amburn, J.D. Austin, R. Cromartie, A. Geselowitz, T. Greer, B.H. Romeny, J.B. Zimmerman, and K. Zuiderveld, "Adaptive histogram equalization and its variations.," Computer vision, graphics, and image processing, vol. 39, no. 3, pp. 355-368, 1987.
[3] E.H. Land, "The retinex theory of color vision.," Scientific american, vol. 237, no. 6, pp. 108-129, 1977.
[4] D.J. Jobson, Z.U. Rahman, and G.A. Woodell, "Properties and performance of a center/surround retinex.," IEEE transactions on image processing, vol. 6, no. 3, pp. 451-462, 1997.
[5] D.J. Jobson, Z.U. Rahman, and G.A. Woodell, "A multiscale retinex for bridging the gap between color images and the human observation of scenes.," IEEE Transactions on Image processing, vol. 6, no. 7, pp. 965-976, 1997.
[6] Park, W. Jin, et al., "A low-cost and high-throughput fpga implementation of the retinex algorithm for real-time video enhancement.," IEEE Transactions on Very Large Scale Integration (VLSI) Systems, vol. 28, no. 1, pp. 101-114, 2019.
[7] C. Wei, W. Wang, W. Yang, and J. Liu., “Deep retinex decomposition for low-light enhancement.,” Mach. Vis. Conf, pp. 127-136, 2018.
[8] F. Lv, Y. Li, and F. Lu, "Attention guided low-light image enhancement with a large scale low-light simulation dataset," International Journal of Computer Vision, vol. 129, no. 7, pp. 2175-2193, 2021.
[9] 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, 2009.
[10] Y.F. Wang, H.M. Liu, and Z.W. Fu, "Low-light image enhancement via the absorption light scattering model," IEEE Transactions on Image Processing, vol. 28, no. 11, pp. 5679-5690, 2019.
[11] Y. Li, Y. Zhang, A. Geng, L. Cao, and J. Chen, "Infrared image enhancement based on atmospheric scattering model and histogram equalization.," Optics & laser technology, pp. 99-107, 2016.
[12] S.E. Kim, T.H. Park, and I.K. Eom, "Fast single image dehazing using saturation based transmission map estimation.," IEEE Transactions on Image Processing, pp. 1985-1998, 2019.
[13] W. Wang, C. Wei, W. Yang, and J. Liu, "Gladnet: Low-light enhancement network with global awareness.," IEEE international conference on automatic face & gesture recognition, pp. 751-755, 2018.
[14] C. Lee, C. Lee, and C.S. Kim, "Contrast enhancement based on layered difference representation of 2D histograms. IEEE transactions on image processing," IEEE transactions on image processing, vol. 22, no. 12, pp. 5372-5384, 2013.
[15] X. Fu, D. Zeng, Y. Huang, Y. Liao, Y. Liao, X. Ding, and J. Paisley, "A fusion-based enhancing method for weakly illuminated images.," Signal Processing, pp. 82-96., 2016.
[16] X. Guo, Y. Li, and H. Ling, "Low-light image enhancement via illumination map estimation.," IEEE Transactions on image processing, vol. 26, no. 2, pp. 982-993, 2016.
[17] S.G. Narasimhan, S.K. Nayar, "Contrast restoration of weather degraded images.," IEEE transactions on pattern analysis and machine intelligence, vol. 25, no. 6, pp. 713-724, 2003.
[18] H. Yue, J. Yang, X. Sun, F. Wu, and C. Hou, "Contrast enhancement based on intrinsic image decomposition.," IEEE Transactions on Image Processing, vol. 26, no. 8, pp. 3981-3994, 2017.
[19] "DE2 Development and Education Board User Manual," [Online]. Available: https://www.terasic.com.tw/attachment/archive/226/DE2_70_User_manual_v105.pdf.
[20] "Sony EVI-D70 技術手冊," [Online]. Available: https://pro.sony/en_GR/products/ptz-network-cameras/evi-d70-d70p-pal-.