系統識別號 | U0002-1807200712274300 |
---|---|
DOI | 10.6846/TKU.2007.00532 |
論文名稱(中文) | 灰階影像之最佳彩色化 |
論文名稱(英文) | Optimal Colorization for Grayscale Images |
第三語言論文名稱 | |
校院名稱 | 淡江大學 |
系所名稱(中文) | 資訊工程學系碩士班 |
系所名稱(英文) | Department of Computer Science and Information Engineering |
外國學位學校名稱 | |
外國學位學院名稱 | |
外國學位研究所名稱 | |
學年度 | 95 |
學期 | 2 |
出版年 | 96 |
研究生(中文) | 吳軍逸 |
研究生(英文) | Chun-Yi Wu |
學號 | 694190546 |
學位類別 | 碩士 |
語言別 | 繁體中文 |
第二語言別 | 英文 |
口試日期 | 2007-06-28 |
論文頁數 | 73頁 |
口試委員 |
指導教授
-
林慧珍(hjlin@cs.tku.edu.tw)
委員 - 林慧珍(hjlin@cs.tku.edu.tw) 委員 - 顏淑惠(shyen@cs.tku.edu.tw) 委員 - 徐道義(taoi@gc.shu.edu.tw) |
關鍵字(中) |
彩色化 YCbCr顏色空間 擴散 線性規劃 線性內插 互動式 |
關鍵字(英) |
Colorization YCbCr color space Expansion Linear programming Linear interpolation Interactive |
第三語言關鍵字 | |
學科別分類 | |
中文摘要 |
灰階影像彩色化是一門相當重要的領域。然而由於灰階影像只擁有一維的資訊,也就是只有亮度的資訊,因此若想要將灰階影像轉換成擁有三維資訊,如紅、綠、藍的彩色影像,必須藉由使用者或者是參考的圖形所提供的顏色資訊,配合灰階影像的亮度來進行彩色化的工作。然而直接以RGB顏色空間進行運算,會因為三個維度的關連性過高導致運算上的困難,所以在此我們是利用YCbCr的顏色空間作為彩色化的基礎。彩色化的領域中,最常被討論的議題就是邊緣判斷的問題。而本論文分兩階段來解決區域判定的問題,第一階段以擴散的方式,猶如水波漣漪一般,初步決定顏色的區域;第二階段針對未決定所屬顏色的部份,利用計算其與顏色區域的關係資訊,來完成顏色區域的判斷。接著以線性規劃求出每個顏色區域內,原灰階影像最亮與最暗的像素點之最佳顏色資訊,再利用線性內插法,將整個顏色區域完成彩色化的工作。本論文是使用者為導向的著色方法,屬於互動式的著色系統,藉由提供顏色、亮度、區域關係判定的選擇,讓使用者彈性調整彩色化的結果。 |
英文摘要 |
This paper proposes a user friendly interactive colorization system that utilizes simple regional judgment and linear programming to accomplish the work of colorization. The system provides several colorization modes for users to select by themselves. User only needs to simply scribble on the given gray-level image and select a coloring mode they want. Experimental results show that our system gives natural and satisfactory coloring results. |
第三語言摘要 | |
論文目次 |
第1章 緒論 1 1.1 研究背景 1 1.2 研究內容 4 1.3 論文架構 6 第2章 相關研究與理論基礎 7 2.1 彩色化相關研究 7 2.2 理論基礎 15 2.2.1 線性規劃 (Linear Programming) 15 2.2.1 線性內插(Linear interpolation) 22 2.2.1 顏色空間YCbCr(YCbCr Color Space) 24 第3章 彩色化 25 3.1 邊緣及區域判斷 26 3.1.1 前處理(Pre-process) 27 3.1.2 後處理(Post-proces) 34 3.1.3 Meta Search API & Related Page preprocess 37 3.2 最佳化 38 3.3 彩色化 47 第4章 實驗結果與探討 49 4.1 實驗結果 49 4.2 實驗結果探討 52 4.3 實驗結果比較 54 第5章 結論與未來研究方向 60 5.1 結論 60 5.2 未來研究目標與方向 61 參考文獻: 62 附錄—英文論文 65 圖目錄 圖1.1 灰階影像的彩色化 1 圖1.2 灰階影像彩色化的應用 2 圖1.3 互動式的著色系統 (a) 一張灰階影像,上面塗畫(scribbles)為使用者給的顏色,(b) 圖(a)經過彩色化都的結果 3 圖1.4 自動著色系統 (a) 為參考影像,(b)為欲著色灰階影像,(c)為彩色化後的結果 3 圖1.5 彩色化系統流程圖 5 圖2.1 Levin et al.方法的著色效果 (a) 使用者於灰階影像上給予顏色資訊,(b) 彩色化後的結果,(c) 原圖彩色影像 10 圖2.2 Liron Yatziv et al.方法的彩色化效果 (a) 為輸入灰階影像,(b) 使用者做區域顏色的定義,(c) 彩色化後的結果 11 圖2.3 Guoping Qiu et al.方法的著色結果 (a) 為使用者定義顏色,(b) 為著色後的影像 12 圖2.4 Hedlena et al.的著色系統: 最左為上一個畫面的彩色影像,中間為欲著色的畫面影像,右方為著色後的結果 13 圖2.5 Daniel Sykora et al.的著色系統 (a) 前一張彩色卡通畫面,(b) 需要上色的灰階卡通畫面,(c) 彩色化後的結果 13 圖2.6 Daniel Sykora et al.的著色系統,利用資料庫比對,取得顏色資訊 14 圖2.7 二維座標中的(x0,y0)與(x1,y1)22 圖3.1 使用者於灰階影像上經由塗畫(scribbles)給予顏色資訊 26 圖3.2 Sobel邊緣偵測 (a) 原影像,(b)影像經過邊緣偵測後的結果(未執行二值化) 27 圖3.3 方向i=0~3,佇列Qi儲存將進行方向 的擴散的點 29 圖3.4 (Cr',Cb')與(Cr",Cb")於二維座標系統表示方式 47 圖4.1 彩色化實驗例子 (a)與(d) 原灰階影像,(b)與(e) 使用者塗色,(c)與(f) 彩色化結果 (主要權重為灰階變化,亮度為接近原圖灰階) 50 圖4.2 使用者各種不同選擇與設定所產生的彩色化結果 (a)與(b)分別為原灰階影像以及使用者定義區域顏色,(c)與(d)亮度分別為接近原影像灰階及接近使用者顏色,主要權重為灰階變化,(e) 的權重判定依據為距離,(f)的權重判定依據為灰階差(兩者亮度接近原圖) 51 圖4.3 本方法與Liron et al.的方法之著色結果比較 (a) 灰階影像,加上少許使用者的塗畫,(b) Liron et al.方法之著色結果,(c) 本論文的方法之著色結果 56 圖4.4 本方法與Liron et al.的方法之著色結果比較 (a) 灰階影像,加上少許使用者的塗畫,(b) Liron et al.方法之著色結果,(c) 本論文的方法之著色結果 57 圖4.5 本方法與Liron et al.的方法之著色結果比較 (a) 灰階影像,加上少許使用者的塗畫,(b) Liron et al.方法之著色結果,(c) 本論文的方法之著色結果 58 圖4.6 本篇論文的著色系統 (a) 原灰階影像,(b) 使用者定義顏色,(c) 彩色化的結果 59 表目錄 表3.1 視窗中心為初始點,初始點座標加上Si表示其四個方向的鄰居 29 |
參考文獻 |
[1] Simon Premože and William B. Thompson, “Automated Coloring of Panchromatic Orthoimagery”, in Proc. of the ICA Mountain Cartography Workshop, (Mt. Hood, Oregon), May 2002. [2] T. Horiuchi and S. Hirano, “Colorization Algorithm for Grayscale Image by Propagating Seed Pixels”, in Proc. of the IEEE Int. Conf. Pattern Recognition, 2003, pp. 457–460. [3] A. Levin, D. Lischinski, and Y. Weiss, “Colorization Using Optimization,” in Proc. of the ACM SIGGRAPH Conf., 2004, pp. 689–694. [4] E. Reinhard, M. Ashikhmin, B. Gooch, and P. Shirley, “Color transfer between images”, IEEE Comput. Graph. Appl., vol. 21, no. 5, May 2001, pp. 34–41. [5] T. Welsh, M. Ashikhmin, and K. Mueller, “Transferring Color to Greyscale Images”, in Proc. of the ACM SIGGRAPH 2002, July 2002, pp. 277-280. [6] Liron Yatziv and Guillermo Sapiro, “Fast Image and Video Colorization Using Chrominance Blending,” IEEE Transactions on Image Processing, vol. 15, May 2006, pp. 1120 – 1129. [7] Paul R. Thie, “An Introduction to Linear Programming and Game Theory,” second edition, Publisher: John Wiley & Sons, 1988, Chapter 3. [8] Jianbo Shi and Jitendra Malik, “Normalized Cuts and Image Segmentation,” in Proc. of the IEEE Conf. of Computer Vision and Pattern Recognition, 1997, pp.731–737. [9] Yair Weiss, “Segmentation Using Eigenvectors: A Unifying View,” in Proc. of the IEEE International Conference on Computer Vision, 1999, pp. 975-982. [10] G. Sapiro, “Inpainting the colors,” IMA Preprint Series 1979: Inst. Math. Appl., Univ. Minnesota, Minneapolis, May 2004. [11] T. Horiuchi, “Estimation of color for gray-level image by probabilistic relaxation,” in Proc. of the IEEE Int. Conf. Pattern Recognition, 2002, pp 867–870. [12] Guoping Qiu and Jian Guan, “Color by Linear Neighborhood Embedding,” in Proc. of the IEEE International Conference on Image Processing (ICIP2005), vol. 3, 11-14, Sept. 2005, pp. 988-991. [13] http://mountains.ece.umn.edu/~liron/colorization/ [14] S. T. Roweis and L. K. Saul, “Nonlinear dimensionality reduction by locally linear embedding”, Science 290, 2000, pp. 2323 – 2326. [15] L. K. Saul and S. T. Roweis, “An Introduction to Locally Linear Embedding”, http://www.cs.toronto.edu/~roweis/lle/ [16] Hedlena Bezerra, Bruno Feijo, and Luiz Velho, “A Computer-Assisted Colorization Algorithm based on Topological Difference”, Manaus, 19th Brazilian Symposium on Computer Graphics and Image Processing, 2006, Los Alamitos, CA: IEEE Press, 2006, pp. 71–77. [17] D. Sykora, J. Burianek, and J. Zara, “Unsupervised Colorization of Black-and-White Cartoons,” in Proc. of the 3rd Int. Symp. NPAR, Annecy, France, June 2004, pp. 121–127. |
論文全文使用權限 |
如有問題,歡迎洽詢!
圖書館數位資訊組 (02)2621-5656 轉 2487 或 來信