臺灣博碩士論文加值系統

上色(Colorization)為將色彩付予灰階的圖片或影片的一種技術，近年來，灰階影像如何自動上色為彩色影像成為電腦繪圖中的一項熱門議題，為了將灰階影像彩色化且在過程中減少人為的干預，許多論文都以自動化上色為研究目標。例如Welsh、Levin、Yatziv所提供的灰階影像上色方法，皆可成功地將灰階影像轉變成合適的彩色影像，並且減少人力的參予。其中Welsh的方法採用參考彩色圖片來取得灰階圖片上的顏色；雖然方法簡單易實現，不過需要的時間成本較高，並可能會因為自動比對失誤，而導致顏色上的誤差。而Levin和Yatziv的方法，則在灰階圖片上色時需要人為先做粗略描繪(Scribble)，但人為彩繪時所選擇的位置，將可能影響一張圖片上色結果的好壞。因此，針對以上提到的問題，本篇論文提出了一個灰階影像著色位置之提示方法來解決上述提到的問題。在我們設計的實驗上，除了利用肉眼的主觀判斷以外，並使用PSNR(Peak Signal to Noise Ratio)和UIQI(Universal Image Quality Index)等測量值來客觀地評估圖片好壞；實驗結果證實，我們產生的提示加快了參考彩色圖片自動化上色的過程，提高自動化上色系統之效率，亦可使人為描繪上色更加便利，此確實改善了先前論文中的幾項缺失。

Colorization is a process of adding color to a monochrome image or movie.In order to color the grayscale images and minimize the amount of human labor required,Welsh is to transfer color to a grayscale image from a color image.Though the method is simple and easy to realize, the shortcoming is that the speed is relatively slow,and may get the wrong color because of the wrong compare.The methods of Levin and Yatziv color a image only need　people to annotate the image with a few color scribbles,however the position of the scribbles influence the quality of a colored image.We provide a method to produce the hint position of a grayscale image to solve above problems.In our experiments,besides utilizing visual subjective judgement,use PSNR(Peak Signal to Noise Ratio) and UIQI(Universal Image Quality Index) to evaluate objectively.The results can demonstrate our method that is convenient and practical.

1 緒論 7
1.1 背景 7
1.2 研究動機 7
1.3 研究目標 8
1.4 論文架構 9
2 文獻探討 10
2.1 色彩空間模型 (Color Model)10
2.2 灰階影像上色方法 13
2.3 影像區域分割 (Image Segmentation) 16
2.4 影像評估方法 20
3 灰階圖提示產生方法 23
3.1 方法流程 24
3.2 增強對比 (Constrast Enhancement) 26
3.3 測邊法 (Edge detection)27
3.4 改善測邊效果 30
33.5 邊線連結 32
3.6 細線化 33
3.7 毛邊雜訊去除 35
3.8 提示線上色 37
3.9 上色演算法 39
4 實驗 41
4.1 實驗簡介 41
4.2 實驗一: 對比增強對上色的影響 42
4.3 實驗二: 人為提示線上色 43
4.4 實驗三: 利用原彩色圖片將提示位置上色及比較半自動上色法成效 45
4.5 實驗四: 模擬人工上色行為 49
4.6 實驗五: 參考彩色圖片自動化上色 51
4.7 小結 55
5 結論和未來工作 56

[1] 鍾國亮. 影像處理與電腦視覺. 東華書局, 2003.
[2] TomihisaWelsh,MichaelAshikhmin, andKlausMueller. Transferring color to greyscale images. ACM Trans. Graph., 21(3):277–280, 2002.
[3] Anat Levin, Dani Lischinski, and Yair Weiss. Colorization using optimization. In SIGGRAPH ’04: ACMSIGGRAPH 2004 Papers, pages 689–694, New York, NY, USA, 2004. ACM Press.
[4] Liron Yatziv andGuillermo Sapiro. Fast image and video colorization using chrominance blending. IEEE Transactions on Image Processing, 15(5):1120–1129, 2006.
[5] Daniel L. Ruderman, ThomasW. Cronin, and Chuan-Chin Chiao. Statistics of cone responses to natural images: implications for visual coding. J. Opt. Soc. Am. A, 15(8):2036–2045, 1998.
[6] Erik Reinhard, Michael Ashikhmin, Bruce Gooch, and Peter Shirley. Color transfer between images. IEEE Comput. Graph. Appl., 21(5):34–41, 2001.
[7] Rafael C. Gonzalez and Richard E. Woods. Digital Image Processing. Addison-Wesley Longman Publishing Co., Inc., Boston, MA, USA, 2001.
[8] R. Adams and L. Bischof. Seeded region growing. IEEE Trans. Pattern Anal. Mach. Intell., 16(6):641–647, 1994.
[9] Michael Kass, Andrew Witkin, and Demetri Terzopoulos. Snakes: Active contour models. International Journal of Computer Vision, V1(4):321–331, January 1988.
[10] Luc Vincent and Pierre Soille. Watersheds in digital spaces: An efficient algorithmbased on immersion simulations. IEEE Trans. Pattern Anal.Mach. Intell., 13(6):583–598, 1991.
[11] J. A. Hartigan and M. A. Wong. A k-means clustering algorithm. JSTOR: Applied Statistics, 28(1):100–108, 1979.
[12] Jianbo Shi and Jitendra Malik. Normalized cuts and image segmentation. IEEE Transactions on Pattern Analysis andMachine Intelligence, 22(8):888–905, 2000.
[13] Timothee Cour, Florence Benezit, and Jianbo Shi. Spectral segmentation with multiscale graph decomposition. In CVPR ’05: Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’05) - Volume 2, pages 1124–1131, Washington, DC, USA, 2005. IEEE Computer Society.
[14] Z. Wang and A.C. Bovik. A universal image quality index. IEEE Signal Processing Letters, 9(3):81–84, March 2002.
[15] A classified and comparative study of edge detection algorithms. In ITCC ’02: Proceedings of the International Conference on Information Technology: Coding and Computing, page 117, Washington, DC, USA, 2002. IEEE Computer Society.
[16] Clausi D Ali M. Using the canny edge detector for feature extraction and enhancement of remote sensing images. In IGARSS-01, pages 2298–2300 vol.5, Sydney, NSW, Australia, 2001. IEEE International Symposium on Geoscience and Remote Sensing.
[17] J. Canny. A computational approach to edge detection. IEEE Trans. Pattern Anal. Mach. Intell., 8(6):679–698, November 1986.
[18] T. Y. Zhang and C. Y. Suen. A fast parallel algorithm for thinning digital patterns. Commun. ACM, 27(3):236–239, 1984.
[19] J.R. Parker. Algorithm for Image Processing and Computer Vision. Wiley Computer Publishing, 1999.
[20] Luiz Filipe M. Vieira, Erickson R. do Nascimento, Fernando A. Fernandes Jr., Rodrigo L. Carceroni, Rafael D. Vilela, and Arnaldo de Albuquerque Ara ´ ujo. Fully automatic coloring of grayscale images. Image Vision Comput., 25(1):50–60, 2007.
[21] J.R. Parker. Algorithms for Image Processing and Computer Vision. Wiley, 1996.
[22] The USC-SIPI Image Database. http://sipi.usc.edu/database/.
[23] Standard Image Database. http://sampl.ece.ohiostate.edu.
[24] T. Chen, Y. Wang, V. Schillings, and C. Meinel. Grayscale image matting and colorization, 2004.
[25] Revital Ironi, Daniel Cohen-Or, and Dani Lischinski. Colorization by example. In Rendering Techniques, pages 201–210, 2005.
[26] Qing Luan, Fang Wen, Daniel Cohen-Or, Lin Liang, Ying-Qing Xu, and Heung-Yeung Shum. Natural image colorization. In Jan Kautz and Sumanta Pattanaik, editors, Rendering Techniques 2007 (Proceedings Eurographics Symposium on Rendering). Eurographics, June 2007.