Designing a stable feedback control system for blind image deconvolution. (English) Zbl 1458.94021
Summary: Blind image deconvolution is one of the main low-level vision problems with wide applications. Many previous works manually design regularization to simultaneously estimate the latent sharp image and the blur kernel under maximum a posterior framework. However, it has been demonstrated that such joint estimation strategies may lead to the undesired trivial solution. In this paper, we present a novel perspective, using a stable feedback control system, to simulate the latent sharp image propagation. The controller of our system consists of regularization and guidance, which decide the sparsity and sharp features of latent image, respectively. Furthermore, the formational model of blind image is introduced into the feedback process to avoid the image restoration deviating from the stable point. The stability analysis of the system indicates the latent image propagation in blind deconvolution task can be efficiently estimated and controlled by cues and priors. Thus the kernel estimation used for image restoration becomes more precision. Experimental results show that our system is effective on image propagation, and can perform favorably against the state-of-the-art blind image deconvolution methods on different benchmark image sets and special blurred images.
MSC:
| 94A08 | Image processing (compression, reconstruction, etc.) in information and communication theory |
| 93B52 | Feedback control |
Keywords:
image restoration; blind deconvolution; kernel estimation; feedback control system; latent sharp imageReferences:
| [1] | Bhaya, A., & Kaszkurewicz, E. (2004). Iterative methods as dynamical systems with feedback control. In IEEE conference on decision and control, Vol. 3; Bhaya, A., & Kaszkurewicz, E. (2004). Iterative methods as dynamical systems with feedback control. In IEEE conference on decision and control, Vol. 3 |
| [2] | Chan, T. F.; Wong, C.-K., Total variation blind deconvolution, IEEE TIP, 7, 3, 370-375 (1998) |
| [3] | Chen, Y., Yu, W., & Pock, T. (2015). On learning optimized reaction diffusion processes for effective image restoration. In CVPR; Chen, Y., Yu, W., & Pock, T. (2015). On learning optimized reaction diffusion processes for effective image restoration. In CVPR |
| [4] | Cho, S.; Lee, S., Fast motion deblurring, ACM ToG, 28, 5, 145 (2009) |
| [5] | Daniel, Z., & Yair, W. (2011). From learning models of natural image patches to whole image restoration. In ICCV; Daniel, Z., & Yair, W. (2011). From learning models of natural image patches to whole image restoration. In ICCV |
| [6] | Fergus, R.; Singh, B.; Hertzmann, A.; Roweis, S.; Freeman, W., Removing camera shake from a single photograph, ACM ToG, 25, 3, 787-794 (2006) · Zbl 1371.94125 |
| [7] | Gong, D., Tan, M., Zhang, Y., van den Hengel, A., & Shi, Q. (2016). Blind image deconvolution by automatic gradient activation. In CVPR; Gong, D., Tan, M., Zhang, Y., van den Hengel, A., & Shi, Q. (2016). Blind image deconvolution by automatic gradient activation. In CVPR |
| [8] | Gross, R.; Matthews, I.; Cohn, J.; Kanade, T.; Baker, S., Multi-pie, Image and Vision Computing, 28, 5, 807-813 (2010) |
| [9] | Gu, B.; Sheng, V. S.; Tay, K. Y.; Romano, W.; Li, S., Incremental support vector learning for ordinal regression, TNNLS, 26, 7, 1403-1416 (2015) |
| [10] | Gu, B.; Sun, X.; Sheng, V. S., Structural minimax probability machine, TNNLS (2017) |
| [11] | Han, Z. X.; Yan, C. H.; Zhang, Z., Study on robust control system of boiler steam temperature and analysis on its stability, Proceedings of the Csee, 30, 8, 101-109 (2010) |
| [12] | Hu, Z., Cho, S., Wang, J., & Yang, M.-H. (2014). Deblurring low-light images with light streaks. In IEEE CVPR; Hu, Z., Cho, S., Wang, J., & Yang, M.-H. (2014). Deblurring low-light images with light streaks. In IEEE CVPR |
| [13] | Hu, Z., & Yang, M.-H. (2012). Good regions to deblur. In ECCV; Hu, Z., & Yang, M.-H. (2012). Good regions to deblur. In ECCV |
| [14] | Krishnan, D., & Fergus, R. (2009). Fast image deconvolution using hyper-laplacian priors. In NIPS; Krishnan, D., & Fergus, R. (2009). Fast image deconvolution using hyper-laplacian priors. In NIPS |
| [15] | Krishnan, D., Tay, T., & Fergus, R. (2011). Blind deconvolution using a normalized sparsity measure. In CVPR; Krishnan, D., Tay, T., & Fergus, R. (2011). Blind deconvolution using a normalized sparsity measure. In CVPR |
| [16] | Levin, A.; Fergus, R.; Durand, F.; Freeman, W. T., Image and depth from a conventional camera with a coded aperture, ACM ToG, 26, 3, 70 (2007) |
| [17] | Levin, A., Weiss, Y., Durand, F., & Freeman, W. T. (2009). Understanding and evaluating blind deconvolution algorithms. In CVPR; Levin, A., Weiss, Y., Durand, F., & Freeman, W. T. (2009). Understanding and evaluating blind deconvolution algorithms. In CVPR |
| [18] | Levin, A., Weiss, Y., Durand, F., & Freeman, W. T. (2011). Efficient marginal likelihood optimization in blind deconvolution. In CVPR; Levin, A., Weiss, Y., Durand, F., & Freeman, W. T. (2011). Efficient marginal likelihood optimization in blind deconvolution. In CVPR |
| [19] | Liu, D. C.; Nocedal, J., On the limited memory bfgs method for large scale optimization, Mathematical Programming, 45, 1-3, 503-528 (1989) · Zbl 0696.90048 |
| [20] | Martin, D., Fowlkes, C., Tal, D., & Malik, J. (2001). A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics. In ICCV, Vol. 2; Martin, D., Fowlkes, C., Tal, D., & Malik, J. (2001). A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics. In ICCV, Vol. 2 |
| [21] | Nah, S., Kim, T. H., & Lee, K. M. (2017). Deep multi-scale convolutional neural network for dynamic scene deblurring. In IEEE CVPR; Nah, S., Kim, T. H., & Lee, K. M. (2017). Deep multi-scale convolutional neural network for dynamic scene deblurring. In IEEE CVPR |
| [22] | Pan, J., Hu, Z., Su, Z., & Yang, M.-H. (2014a). Deblurring face images with exemplars. In ECCV; Pan, J., Hu, Z., Su, Z., & Yang, M.-H. (2014a). Deblurring face images with exemplars. In ECCV |
| [23] | Pan, J., Hu, Z., Su, Z., & Yang, M.-H. (2014b). Deblurring text images via l0-regularized intensity and gradient prior. In CVPR; Pan, J., Hu, Z., Su, Z., & Yang, M.-H. (2014b). Deblurring text images via l0-regularized intensity and gradient prior. In CVPR |
| [24] | Perrone, D., & Favaro, P. (2014). Total variation blind deconvolution: The devil is in the details. In CVPR; Perrone, D., & Favaro, P. (2014). Total variation blind deconvolution: The devil is in the details. In CVPR |
| [25] | Schmidt, U., & Roth, S. (2014). Shrinkage fields for effective image restoration. In CVPR; Schmidt, U., & Roth, S. (2014). Shrinkage fields for effective image restoration. In CVPR |
| [26] | Schuler, C. J., Christopher Burger, H., Harmeling, S., & Scholkopf, B. (2013). A machine learning approach for non-blind image deconvolution. In CVPR; Schuler, C. J., Christopher Burger, H., Harmeling, S., & Scholkopf, B. (2013). A machine learning approach for non-blind image deconvolution. In CVPR |
| [27] | Shan, Q.; Jia, J.; Agarwala, A., High-quality motion deblurring from a single image, ACM ToG, 27, 3, 73 (2008) |
| [28] | Sun, L., Cho, S., Wang, J., & Hays, J. (2013). Edge-based blur kernel estimation using patch priors. In ICCP; Sun, L., Cho, S., Wang, J., & Hays, J. (2013). Edge-based blur kernel estimation using patch priors. In ICCP |
| [29] | SUN Guo-fa, T. Y.; Su-zhen, W., Adaptive neural output feedback control for strict feedback nonlinear system, Control Theory & Applications (2017) · Zbl 1389.93113 |
| [30] | Turner, J. D.; Chun, H. M.; Juang, J. N., An analytic solution for the state trajectories of a feedback control system, Journal of Guidance, Control, and Dynamics, 8, 1, 147-148 (2015) |
| [31] | Wipf, D.; Zhang, H., Revisiting bayesian blind deconvolution, Machine Learning Research, 15, 1, 3595-3634 (2014) · Zbl 1312.94012 |
| [32] | Xu, L., & Jia, J. (2010). Two-phase kernel estimation for robust motion deblurring. In ECCV; Xu, L., & Jia, J. (2010). Two-phase kernel estimation for robust motion deblurring. In ECCV |
| [33] | Xu, L., Zheng, S., & Jia, J. (2013). Unnatural l0 sparse representation for natural image deblurring. In CVPR; Xu, L., Zheng, S., & Jia, J. (2013). Unnatural l0 sparse representation for natural image deblurring. In CVPR |
| [34] | Yang, J.; Zhang, Y.; Yin, W., An efficient tvl1 algorithm for deblurring multichannel images corrupted by impulsive noise, SIAM Journal on Scientific Computing, 31, 4, 2842-2865 (2009) · Zbl 1195.68110 |
| [35] | Yousaf, S.; Qin, S., Closed-loop restoration approach to blurry images based on machine learning and feedback optimization, TIP, 24, 12, 5928-5941 (2015) · Zbl 1408.94777 |
| [36] | Zhang, H., Wipf, D., & Zhang, Y. (2013). Multi-image blind deblurring using a coupled adaptive sparse prior. In CVPR; Zhang, H., Wipf, D., & Zhang, Y. (2013). Multi-image blind deblurring using a coupled adaptive sparse prior. In CVPR |
| [37] | Zuo, W.; Ren, D.; Zhang, D.; Gu, S.; Zhang, L., Learning iteration-wise generalized shrinkage-thresholding operators for blind deconvolution, IEEE TIP, 25, 4, 1751-1764 (2016) · Zbl 1408.94856 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
