在這篇論文中，我們提出了簡單有效的雙色灰階性質跟梯度性質用來對文字影像的去模糊化，而此性質是根據清晰文字影像跟模糊文字影像的特性來提出。而我們提出的方法可以很簡易的進行最佳化且計算時間相當快。此外，也不需要額外的前處理作業跟額外的工具。而反摺積最常發生的振鈴效應，我們也提出了一個有效的局部平滑方法來解決。根據以上的敘述，我們設計一個目標函式來進行最佳化，得到使清晰影像以及模糊核。實驗結果顯示我們的方法在跟幾篇傑出的論文比較中表現得較好。最後，我們展示了我們的方法可以擴展到複雜背景的文字影像、低光源的影像以及自然影像也有效。

In this thesis, we propose a simple effective two-tone intensity prior and gradient prior for text image deblurring. It can be optimized easily and its computational time is quite fast. We also present a local smooth prior to solve the ringing artifacts caused by deconvolution. We propose an objective function to generate intermediate results and the blur kernel. Experimental results demonstrate that our method performs well against the state-of-the-art methods. Finally, we show our method can be extended for image deblurring in text images of complex background, low illumination images and natural images.

[1] R. Fergus, B. Singh, A. Hertzman, S. T. Roweis, and W. T. Freeman, “Removing camera shake from a single photograph, ”ACM Transactions on Graphics, vol. 25, no. 3, pp. 787–794, 2006.
[2] Q. Shan, J. Jia, and A. Agarwala, “High-quality motion deblurring from a single image,” ACM Transactions on Graphics, vol. 27, no. 3, p. 73, 2008.
[3] S. Cho and S. Lee, “Fast motion deblurring,” ACM Transactions on Graphics, vol. 28, no. 5, p. 145, 2009.
[4] L. Xu and J. Jia, “Two-phase kernel estimation for robust motion deblurring,” in ECCV, 2010, pp. 157–170.
[5] D. Krishnan, T. Tay, and R. Fergus, “Blind deconvolution using a normalized sparsity measure,” in CVPR, 2011, pp. 2657–2664.
[6] L. Xu, S. Zheng, and J. Jia, “Unnatural L0 sparse representation for natural image deblurring,” in CVPR, 2013, pp. 1107–1114.
[7] H. Cho, J. Wang, and S. Lee, “Text image deblurring using text-specific properties,” in ECCV, 2012, pp. 524–537.
[8] X. Chen, X. He, J. Yang, and Q. Wu, “An effective document image deblurring algorithm,” in CVPR, 2011, pp. 369–376.
[9] D. Krishnan and R. Fergus, “Fast image deconvolution using Hyper-Laplacian priors,” in NIPS, 2009, pp. 1033–1041.
[10] J. Pan and Z. Su, “Fast l0-regularized kernel estimation for robust motion deblurring,” IEEE Signal Processing Letters, vol. 20, no. 9, pp. 841–844, 2013.
[11] L. Xu, C. Lu, Y. Xu, and J. Jia, “Image smoothing via L0 gradient minimization,” ACM Transactions on Graphics, vol. 30, no. 6, p. 174, 2011.
[12] L. Sun, S. Cho, J. Wang, and J. Hays, “Edge-based blur kernel estimation using patch priors,” in ICCP, 2013.
[13] J. Pan, Z. Hu, Z. Su, and M.-H. Yang, “Deblurring text images via L0-regularized intensity and gradient prior,” in CVPR, 2014, pp. 2901–2908.
[14] S. Fang, Y.-D. Liu, and Y. Cao, “Blur kernel estimation using blurry structure,”, in ICIP, 2016,pp.2703-2707
[15] Y. Zhou and N. Komodakis, “A MAP-estimation Framework for Blind Deblurring Using High-level Edge Priors”, in ECCV,2014, pp 142-157
[16] J. Pan and D. Sun, “Blind Image Deblurring Using Dark Channel Prior”, in CVPR,2016,pp.1628-1636
[17] J. Pan, R. Liu, Z. Su, G. Liu, “Motion blur kernel estimation via salient edges and low rank prior”, in ICME,2014
[18] J. Pan, Z. Lin, Z. Su, and M.-H. Yang, “Robust Kernel Estimation with Outliers Handling for Image Deblurring”,in CVPR,2016,pp.2800-2808
[19] L. Mai, F. Liu, “Kernel Fusion for Better Image Deblurring”,in CVPR,2015,pp.371-380
[20] Y.-W. Tai, St. Lin, “Motion-aware noise filtering for deblurring of noisy and blurry images”, in CVPR,2012,pp.17-24
[21] N. Joshi, C. L. Zitnick, R. Szeliski, and D. Kriegman“ Image Deblurring and Denoising using Color Priors”, in CVPR,2009,pp.1550-1557
[22] L. Xiao, J. Wang, W. Heidrich, and M. Hirsch, “Learning High-Order Filters for Efficient Blind Deconvolution of Document Photographs”, in ECCV,2016
[23] W.-S. Lai, J.-J. Ding, Y.-Y. Lin, and Y.-Y. Chuang, “Blur Kernel Estimation using Normalized Color-Line Priors, in CVPR,2015”