透過您的圖書館登入
IP:18.116.42.108
  • 學位論文

基於可調式邊界偵測的塗色演算法及其應用

Adaptive Edge Dectection based Colorization Algorithms and Its Applications

指導教授 : 吳家麟

摘要


塗色通常是指將顏色填入黑白相片或是電影的一種需要電腦幫忙的程序,而這個程序通常會需要在影像中分割區域或是在一段影片中去追縱這些區域。這些方法都無法完全自動地達到正確塗色的目的,因此,塗色通常是需要一個人花費很多的時間,很多的力氣,以及所有的耐心去完成為影片上色,而這通常是需要很昂貴的費用。 面對這一個非常有挑戰性的問題,我們利用邊界偵測來發展出一套可以正確塗色的演算法。一般的塗色方法可以看成是一個將三維的資訊(色差模組或是點陣圖的模組)轉到一張灰階圖的所有像素上。最後所得到的色彩資訊將會隨著亮度的變動而變化,然而,塗色方法在某些特別的情況將不可避免地失敗,這時候就需要人來幫忙做一些特定的指示,這些方法的目的都是為了要降低人力的消耗。在我們的實驗中,完全地證明了我們的方法在影像或影片上均有相當好的品質。 接著,我們根據所提出的塗色演算法發展了一套新的彩度壓縮系統,我們將原本輸入進塗色系統的彩度資訊改為色彩種子,這些色彩種子將會在編碼端被當做彩度資訊而被壓縮,並傳送到解碼處,而在解碼處,這些種子將會被解壓縮出來並被拿來產生出整張影像的顏色。 這篇論文的貢獻將不只在發展出一個塗色演算法,並根據這個演算法發展了一套全新的彩度壓縮方法,在未來,更可以將我們所發展出的壓縮方法加入新的壓縮標準。我們也將整個論文發展成一個半自動的塗色系統。

關鍵字

演算法 邊界偵測 塗色 應用 彩度 壓縮

並列摘要


Colorization is a computer-assisted process for adding colors to a grayscale image or movie. The process typically involves segmenting images into regions and tracking these regions across image sequences. Neither of these tasks can be performed reliably in practice; consequently, colorization requires considerable user intervention and remains a tedious, time-consuming and expensive task. Facing such a challenging issue, we introduce a general colorization methodology for images and videos with the aid of edge detection. The colorization is treated as a process for assigning three-dimensional vector (YUV or RGB) to all pixels in a grayscale image. The resultant chrominance will vary with Luminance or Intensity. However, it is inevitable that colorization will fail in some cases. It then needs human-assistance, and the goal is to reduce the necessary labor. In our experiments, the proposed method performs quite well for color images and videos. Next, we also propose a new chrominance coding scheme based on the presented colorization process. We refine the colorization algorithm to input with a set of seeding colors, in which the seeds are the symbols to be coded for representing the chrominance information. At the decoder side, based on those seeds and the predefined colorization scheme, a color video can then be reconstructed. In summary, the contributions of this thesis are not only limited to provide a new colorization algorithm but also a new chrominance coding methodology, which can be used to enhance the existing video coding standards.

參考文獻


[3] Welsh, T., Ashikhmin, M., and Mueller, K. 2002. “Transferring color to greyscale images,” ACM Transactions on Graphics 21, 3 (July), 277-280.
[4] Efros, A. A.and Freeman, W.T., 2001. “Image Quilting for Texture Synthesis and Transfer,” In Proceedings of ACM SIGGRAPH 2001, 341-346.
[6] Tang, B., Sapiro, G., and Casseles, V. 2001. “Color image enhancement via chromaticity diffusion,” IEEE Transactions on Image Processing 10, 5, 701-708
[7] Andreas, K., and Mongi, A. 2005. “Detection and classification of edges in color images.” IEEE Signal Processing Magazine (Jan), 64-73.
[8] Dorin, C., and Peter, M., 1997. Robust analysis of feature spaces: “color image segmentation.” Proceedings of CVPR..

延伸閱讀