臺灣博碩士論文加值系統

在這篇論文中，我們提出了從單張模糊影像中利用二階區塊的影像還原演算法，我們的方法可以還原運動模糊之數位影像並且此模糊點擴散函數不需具有任何特定的參數型式。首先，我們先手動選取多個二階區塊，希望藉由二階區塊的幫助來估算出點擴散函數。接著我們提出一個搜尋演算法來尋找最好的初始估計，此方法包含藉由我們定義的能量函數來尋找最好的門檻值以及利用多項式光罩模型處理些微光影變化所照成的影響。我們在此論文中提出一個機率模型將動態模糊估計及影像還原結合成ㄧ個統計估測的形式。針對此模型，我們利用了交替演算法反覆地調整動點擴散函數來得到更好的影像還原結果。在估計出動態模糊之後，我們再利用Richardson-Lucy影像還原演算法來對整張影像進行影像還原。經由實驗結果的呈現，包含對模擬模糊影像及真實模糊影像的影像還原結果，我們驗證了本篇論文所提出之動態模糊影像還原演算法的功效。

In this thesis, we propose a novel image restoration framework for restoring images degraded by unknown motion blurs from a single image. Our approach takes advantage of the bi-level image patches to estimate the blur kernel. The patches which contain only two-grayscale regions are first selected. Then we propose a method to find an initial guess for our algorithm. This method includes searching a threshold value based on a cost function and utilizing an illumination model to account for small illumination variations. Afterwards, we propose a probabilistic model which combines both blur kernel estimation and non-blind bi-level image deconvolution into a single maximum a posteriori (MAP) formulation. An alternating minimization algorithm is developed to iteratively refine both the blur kernel and the bi-level blurred patches. Finally, we apply Richardson-Lucy (RL) deconvolution to restore the entire image by using the estimated blur kernel. Some experimental results on the deblurring of simulated and real blurred images are given to demonstrate the performance of the proposed blind motion deblurring algorithm.

Contents

Chapter 1 Introduction 1
1.1 Problem Background 1
1.2 Problem Description 2
1.3 Previous Works 3
1.4 Main Contribution 8
1.5 Thesis Organization 9
Chapter 2 Proposed Motion Deblurring Algorithm using Bi-level Image Patches 10
2.1 System Overview 10
2.2 Blur Kernel Estimation using Bi-level patches 12
2.2.1 Definition of the Likelihood Term 13
2.2.2 Definition of the Prior on the Ideal Unblurred Image 14
2.2.3 Definition of the Prior on the Blur Kernel 19
2.3 Energy Minimization 21
2.3.1 Optimization for Latent image F 22
2.3.2 Optimization for Blur Kernel H 24
2.4 Initial Image Patch Thresholding 25
2.4.1 Image Patch Adjustment 26
2.4.2 Image Patch Thresholding 26
2.4.3 Image Patch Illumination Modeling 28
Chapter 3 Experimental Results 31
3.1 Experiments on Simulated Blurred Images 31
3.2 Experiments on Real Blurred Images 38
Chapter 4 Discussion and Conclusion 43
References 46





List of Figures

Fig. 1. Flowchart of the proposed bi-level-based blind motion deblurring algorithm. 12
Fig. 2. Overview of our proposed blur kernel estimation algorithm. 12
Fig. 3. Ill-posed problem for blind image deconvolution. 15
Fig. 4. The distribution of image gradients. 17
Fig. 5. Effect of the sparse prior Ps(F). 18
Fig. 6. Effect of the locally smooth prior Pm(F). 19
Fig. 7. The heavy-tailed curve of kernel values. 21
Fig. 8. Flowchart of our proposed image patch thresholding algorithm. 26
Fig. 9. The effect of the illumination model. 29
Fig. 10. Poster photo deblurring experiment. 34
Fig. 11. Map photo deblurring experiment. 35
Fig. 12. Results of different iterations in our deconvolution algorithm. 36
Fig. 13. Stadium photo deblurring experiment. 37
Fig. 14. Garage photo deblurring experiment. 37
Fig. 15. Deblurring on a real blurred stadium photo. 40
Fig. 16. A deblurring experiment on a real blurred image. 41
Fig. 17. A real blurred image. 42
Fig. 18. A failed deblurring example. 45

List of Tables

Table 1. Comparison of the restoration results. 38

References

[1] M. R. Banham and A. K. Katsaggelos, Digital image restoration, IEEE Signal Processing Magazine, 42(2):24–41, 1997.
[2] N. Wiener, Extrapolation, interpolation, and smoothing of stationary time series, MIT Press, 1964.
[3] W. H. Richardson, Bayesian-based iterative method of image restoration, Journal of the Optical Society of America, 62(1):55–59, 1972.
[4] A. Levin, R. Fergus, F. Durand, and W. T. Freeman, Image and depth from a conventional camera with a coded aperture, ACM Trans. on Graphics, 26(3):70, 2007.
[5] L. Yuan, J. Sun, L. Quan, and H.-Y, Shum, Progressive inter-scale and intra-scale non-blind image deconvolution, ACM Trans. on Graphics, 2008.
[6] D. Kundur and D. Hatzinakos, Blind image deconvolution, IEEE Signal Processing Magazine, 13(3):43–64, 1996.
[7] M. Chang, A. Tekalp, and A. Erdem, Blur identification using the bispectrum, IEEE Transactions on Signal Processing, 39(10):2323–2325, 1991.
[8] R. Lagendijk, J. Biemond, and D. Boekee, Identification and restoration of noisy blurred images using theexpectation-maximization algorithm, IEEE Transactions on Acoustics, Speech, and Signal Processing, 38(7):1180–1191, 1990.
[9] R. Lagendijk, A. Tekalp, and J. Biemond, Maximum likelihood image and blur identi_cation: a unifying approach, Opt. Eng., 29(5):422–435, 1990.
[10] Y. You and M. Kaveh, A regularization approach to joint blur identification and image restoration, IEEE Transactions on Image Processing, vol. 5, pp. 416–428, Mar. 1996.
[11] T. Chan and C. Wong, Total Variation Blind Deconvolution, IEEE Transactions on Image Processing, Vol. 7, pp. 370-375, 1998.
[12] R. Fergus, B. Singh, A. Hertzmann, S. T. Roweis and W. T. Freeman, Removing camera shake from a single photograph, ACM Trans. on Graphics, 25:787–794, 2006.
[13] J. Jia, Single image motion deblurring using transparency, In Proc. Computer Vision and Pattern Recognition, 1-8, 2007.
[14] X. Liu and A. Gamal, Simultaneous image formation and motion blur restoration via multiple capture, In Proc. Int. Conf. Acoustics, Speech, Signal Processing, (3):1841-1844, 2001.
[15] M. Ben-Ezra and S. K. Nayar, Motion-based motion deblurring, IEEE Trans. on Pattern Analysis and Machine Intelligence, 26(6):689-698, 2004.
[16] R. Raskar, A. Agrawal and J. Tumblin, Coded-exposure photography: motion deblurring using fluttered shutter, ACM Trans. on Graphics, 25(3):795-804, 2006.
[17] A. Agrawal and R. Raskar, Resolving objects at higher resolution from a single motion-blurred image, In Proc. Computer Vision and Pattern Recognition, 1-8, 2007.
[18] B. Bascle, A. Blake and A. Zisserman, Motion deblurring and superresolution from an image sequence, In Proc. Of European Conf. on Computer Vision, 2:573–582, 1996.
[19] A. Rav-Acha and S. Peleg, Two motion-blurred images are better than one, Pattern Recognition Letter 26(3):311–317, 2005.
[20] L. Yuan, J. Sun, L. Quan, and H.-Y, Shum, Image deblurring with blurred/noisy image pairs, ACM Trans. on Graphics, 26(3):1–10, 2007.
[21] J. Jia, J. Sun, C.-K. Tang and H.-Y, Shum, Bayesian correction of image intensity with spatial consideration, In Proc. of European Conf. on Computer Vision, 3:342-354, 2004.
[22] J. Chen, L. Yuan, C.-K Tang, L. Quan, Robust dual motion deblurring, In Proc. Of Computer Vision and Pattern Recognition, 2008.
[23] A. Levin, Blind motion deblurring using image statistics, in Advances in Neural Information Processing Systems, 2006, 2.
[24] J. Bardsley, S. Jefferies, J. Nagy, and R. Plemmons, Blind iterative restoration of images with spatially varying blur, Optics Express, vol. 14, pp. 1767-1782, 2006, 2.
[25] S. Cho, Y. Matsushita, and S. Lee, Removing non-uniform motion blur from images, In Proc. Of International Conference on Computer Vision, 2007.
[26] M. F. Tappen, B. C. Russell, and W. T. Freeman, Exploiting the sparse derivative prior for super-resolution and image demosaicing, In 3rd Int'l Workshop on Statistical and Computational Theories of Vision, 2003.
[27] S. Roth and M. Black, Fields of experts: A framework for learning image priors, In Proc. Of Computer Vision and Pattern Recognition, 2005.
[28] L. Yuan, J. Sun, L. Quan, and H.-Y, Blurred/non-blurred image alignment using sparseness prior, In Proc. Of International Conference on Computer Vision, 2007.
[29] S. Negahdaripour, Revised definition of optical flow: Integration of radiometric and geometric cues for dynamic scene analysis, IEEE Trans. on Pattern Analysis and Machine Intelligence, 20(9):961-979, 1998.
[30] Q. Shan, J. Jia, and A. Agarwala, High-quality motion deblurring from a single image, ACM Trans. on Graphics, 2008.
[31] Final Report From the Video Quality Experts Group on the Validation of Objective Models of Video Quality Assessment VQEG, Mar. 2000
[Online]. Available: http://www.vqeg.org/
[32] Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, Image quality assessment: From error measurement to structural similarity, IEEE Transactios on Image Processing, vol. 13, no. 4, Apr. 2004.
[33] W. H. Lee, S.H. Lai, and C.L. Chen, Iterative blind image motion deblurring via learning a no-reference image quality measure, In Proc. Of International Conference on Image Processing. 4:405-408, 2007.