臺灣博碩士論文加值系統

在電腦圖學( Computer Graphics )的領域中，對於色彩的研究一直以來都是個重要的課題且值得探討的課題。在2001學者Reinhard等人提出的色彩轉換演算法(Color Transfer)，由使用者輸入來源影像(Source Image)與目標影像(Target Im- age)後，演算法自動轉換原始影像之色彩，轉換完成之結果影像(Result Image)其色彩分佈類似於目標影像的色彩。其演算法具有簡單易懂、實做容易、處理速度快等優點，但是當來源影像及目標影像在色彩上匹配度不佳時，容易產生不符合視覺經驗之結果影像。

本篇論文針對Reinhard所提出之色彩轉換演算法之缺失，提出具體改進之道。我們以Reinhard的演算法為基礎，透過使用統計學上的轉換技巧，對來源影像進行轉換，使其色彩資訊分布趨於常態分布，藉此來避免產生不符合視覺效果的結果影像，而提高色彩轉換演算法的效率。

本文使用的轉換技巧分為冪轉換及模轉換兩種，兩者皆有藉著降低目標分佈偏態係數，使其分布趨於常態的特性，也因此我們可以藉著此兩種轉換技巧，達成我們的目的。

經過實驗結果顯示，我們提出的利用轉換技巧達成有效之色彩轉換演算法，結果明顯優於Reinhard的方法，也大幅減少了結果影像產生不符合視覺經驗的情形，成功的提升色彩轉換演算法的效率。

除了改善不符合視覺經驗的結果影像外，我們所使用的轉換技巧，也可以用在使用Reinhard的方法就得到良好結果的影像，使其增加多樣化的風格，而使色彩轉換演算法更加的有彈性。

另外轉換技巧也可使用在普通亮度分布不均勻的影像中，修正其亮度分布，達到亮度校正的功能。

總結此研究，我們認為我們所提出之利用轉換技巧達成有效之色彩轉換之演算法，成功的保留了Reinhard方法的優點，包括處理速度快及簡單易懂等，也提升了色彩轉換演算法的效率，除了減少不符合視覺經驗之結果影像的產生，更提供了更多樣化的色彩轉換，我們認為所提出之演算法已對色彩轉換作出具體之貢獻。

Colour theory has been the subject of intensive research these years in computer graphics community. Recently, researchers presented a color transfer algorithm in which needs two images, source and target images, inputted by users. The algorithm automatically delivers colors of target image onto source image. The resulting image’s color distributions will be similar to target’s. Reinhard’s algorithm has three advantages: 1) easy to understand, 2) easy to practice, and 3) less computational time. However, it has a distinct drawback, that is, as the source image’s color distributions does not match with target’s, the resulting image may not conform to our visual experiences.

In this thesis, we followed Reinhard’s color transfer algorithm and employed a transformation technique to improve the drawback. The main concept is that we attempted to vary source image’s color distributions . Firstly, we let images be fit in with target’s distribution, and then we execute the color transfer operator. We used statistical transformation techniques, power and modulus transformations, to change the distributions of source image to avoid yielding an unnatural resultant image.

The experimental results show that our algorithm produces a better result than Reinhard’s. This is due to the fact that we solve the unnatural problem so that the result can be adopted to our visual experiences, always the algorithm’s effects. In spite of solving the drawback efficiently, our algorithm can also employ successful images shown in Reinhard’s algorithm, providing more varied-style results. In addition, the transformation techniques can also be used to images to balance their luminance distributions. Overall, our proposed algorithm retains the advantages of Reinhard’s, and eliminates the unnatural effects. The benefit of our algorithm is to provide an easy way to simulate varied scenes with different applications such as movie, photography, and astronomy.

第一章 緒論…………………………………………………………………………..1
1.1 研究動機與目標…………………………………………………………….1
1.2 論文架構…………………………………………………………………….3

第二章 相關文獻回顧………………………………………………………………..4
2.1 色彩轉換演算法之相關文獻回顧……..………………………………..….4
2.2 灰階轉彩色演算法之相關文獻回顧…………………..………………….22
2.3 轉換技巧之相關文獻回顧…………………….………………………..…26

第三章 使用轉換技巧達成有效之色彩轉換……………………………………....30
3.1 傳統的轉換技巧……………….…………………………………………..30
3.2 新的轉換技巧……………………………………………………………...33
3.3 使用轉換技巧達成有效之色彩轉換演算法……………………………...37

第四章 測試結果與分析……………………………………………………………48
4.1 測試環境…………………………………………………………………...48
4.2 系統實現…………………………………………………………………...48
4.3 測試結果與分析…………………………...................................................49

第五章 結論與未來工作……………………………………………………………65
5.1 結論………………………………………………………………………...65
5.2 未來工作…………………………………………………………………...66

參考文獻……………………………………………………………………………..68
中英對照表…………………………………………………………………………..70
英中對照表…………………………………………………………………………..72

[Berl1969]B. BERLIN, and P. KAY, “Basic Color Terms: Their Universality
and Evolution,” University of California Press, 1969
[Boxg1964]G. E. P. Box & D. R. Cox, “An Analysis of Transformations (with Discussion),” . J. R. Statist. Soc. B 26, 211-52 , 1964
[Chan2003]Y. Chang, S. Saito, K. Uchikawa, and M. Nakajima, “A Framework for Transfer Colors Based on the Basic Color Categories,” in Proceedings of Computer Graphics International, pp. 176-181, 2003.
[Chan2004]Y. Chang, S. Saito, K. Uchikawa, and M. Nakajima, “Example-Based Color Stylization Based on Categorical Perception,” in First Symposium on Applied Perception in Graphics and Visualization, 2004.
[Hert2001]A. Hertzmann, C. E. Jacobs, N. Oliver, B. Curless, and D. H. Salesin, “Image Analogies,” ACM Transactions on Graphics (in Proceedings of SIGGRAPH ‘01), pp. 327-340, 2001.
[Hunt2004]R. W. G. Hunt, “The Reproduction of Colour,” John Wiley & Sons, 6 edition, November 1, 2004
[Ichi1997]N. ICHIMURA, “Region Segmentation Method Using both Feature Space and Image Space Based on Robust Clustering.” in IEICE Transaction on Information and Systems, vol. J80-D2, pp. 1752-1763, 1997
[Pero1990]P. PERONA, and J. MALIK, “Scale-Space and Edge Detection Using Anisotropic Diffusion.” in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 12, pp. 629-639, 1990
[Rein2001]E. Reinhard, M. Ashikhmin, B. Gooch, and P. Shirley, “Color Transfer between Images,” IEEE Computer Graphics and Applications, Vol. 21, No. 5, pp. 34-41, 2001.
[Rude1998]D. J. Ruderman, T. W. Cronin, and C. C. Chiao, “Statistics of Cone Responses to Natural Images: Implications for Visual Coding,” Journal of Optical Society of America, Vol. 15, No. 8, pp. 2036-2045, 1998.
[Wang2004]C. M. Wang and Y. H. Huang, “A Novel Color Transfer Algorithm for Image Sequences,” Journal of Information Science and Engineering, 2004 (accepted).
[Wels2002]T. Welsh, M. Ashikhmin, and K. Mueller, “Transferring Color to Greyscale Images,” ACM Transactions on Graphics, (in Proceedings of SIGGRAPH 2002), pp. 277-280, 2002.
[Yeoi2000]I. K. Yeo and R. A. Johnson, “A New Family of Power Transformations　to　Improve　Normality　or　Symmetry,”　 Biometrika, 87, 4, pp 954-959, 2000.
[Zhan2004]M. Zhang, N. D. Georganas, “Fast Color Correction Using Principal Regions Mapping in Different Color Spaces,” Real-Time Imaging, Vol. 10, pp. 23-30, 2004.
[Zhao2004]G. Zhao, S. Xiang, and H. Li, ”Application of Higher Moments in Color Transfer Between Images,” Journal of Computer　Aided Design & Computer Graphics Beijing, 2004.