研究生: |
官晉宇 Kuan, Chin-Yu |
---|---|
論文名稱: |
基於文字影像高對比灰階性質的去模糊方法 High-contrast Intensity Prior for Text Images Deblurring |
指導教授: |
張隆紋
Chang, Long-Wen |
口試委員: |
張寶基
Chan,Pao-Chi 杭學鳴 Hang, Hsueh-Ming |
學位類別: |
碩士 Master |
系所名稱: |
電機資訊學院 - 資訊工程學系 Computer Science |
論文出版年: | 2016 |
畢業學年度: | 104 |
語文別: | 英文 |
論文頁數: | 38 |
中文關鍵詞: | 文字影像 、去模糊 |
外文關鍵詞: | text images, deblurring |
相關次數: | 點閱:2 下載:0 |
分享至: |
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
在這篇論文中,我們提出了一個簡單且有效的高對比灰階性質用來進行文字影像的去模糊化,而此性質是觀察自清晰文字影像與模糊影像之間性質差異而來。我們會根據所提出的性質來設計我們的目標函式,並且演繹簡單的方法來對此目標函式進行最佳化以得到使影像模糊的模糊核。我們將目標函式變成兩個子問題來解決,並且改寫這兩個子問題。我們在不失去原本的意思之下改寫其中一個子問題以化簡計算所需的步驟,而另一個原本是在灰階空間中的子問題則是被改寫成在梯度空間以使得此問題的解能變得更穩定。在論文的最後,我們會與其他傑出的方法來比較我們的實驗結果,並且我們會展示我們的方法不只對文字影像有用甚至是真實世界場景的影像也有效。
In this paper, we propose a simple effective high-contrast intensity prior for text image deblurring based on the distinct properties within sharp text images and blurred text images. We develop an iterative and efficient optimization method to solve the objective function based on the high-contrast intensity prior for the blur kernel. We split the objective function into two sub-problems. We simplify one sub-problem to make its computation more efficient. We rewrite the other sub-problem based on the intensity space by changing it to the gradient space to make its solution stable. The proposed method can be applied to text images and even real-world images with very good results compared to some state-of-the art methods.
[1] T.-H. Li and K.-S. Lii. A joint estimation approach for two-tone image deblurring by blind deconvolution. IEEE Transactions on Image Processing, 11(8):847–858, 2002.
[2] A. Levin. Blind motion deblurring using image statistics. In NIPS, 2006.
[3] R. Fergus, B. Singh, A. Hertzmann, S. T. Roweis, and W. T. Freeman. Removing camera shake from a single photograph. ACM Trans. Graph., 25(3):787–794, 2006.
[4] J. Jia. Single image motion deblurring using transparency. In CVPR, 2007.
[5] Q. Shan, J. Jia, and A. Agarwala. High-quality motion deblurring from a single image. ACM Trans. Graph., 27(3):73, 2008.
[6] D. Krishnan, R. Fergus, Fast image deconvolution using hyper-laplacian priors, in NIPS 2009.
[7] S. Cho and S. Lee. Fast motion deblurring. SIGGRAPH ASIA, 2009.
[8] L. Xu and J. Jia. Two-phase kernel estimation for robust motion deblurring. In ECCV, pages 157–170, 2010.
[9] T. S. Cho. Removing motion blur from photographs. PhD thesis, Massachusetts Institute of Technology, 2010.
[10] X. Chen, X. He, J. Yang, and Q. Wu. An effective document image deblurring algorithm. In CVPR, pages 369–376, 2011.
[11] H. Cho, J. Wang, and S. Lee. Text image deblurring using text-specific properties. In ECCV, pages 524–537, 2012.
[12] L. Xu, S. Zheng, and J. Jia. Unnatural l0 sparse representation for natural image deblurring. In CVPR, pages 1107–1114, 2013.
[13] L. Zhong, S. Cho, D. Metaxas, S. Paris, and J. Wang. Handling noise in single image deblurring using directional filters. In CVPR, pages 612–619, 2013.
[14] J. Kotera, F. Šroubek, P. Milanfar. Blind deconvolution using alternating maximum a posteriori estimation with heavy-tailed priors. R. Wilson, E. Hancock, A. Bors, W. Smith, Eds., Springer: Berlin/Heidelberg, Germany, Volume 8048, In CAIP, pages 59–66, 2013.
[15] J. Pan, Z. Hu, Z. Su, and M.-H. Yang. Deblurring Text Images via L_0-Regularized Intensity and Gradient Prior. In CVPR 2014, pages 2901-2908,