The Restoration of Motion Blurred Images Based on the Background Modeling

Abstract:

For the mutual effects of camera shake and subject movement, the image generation space varying motion blur. In order to achieve image restoration, firstly dividing the image area using the Gaussian background modeling, and updated model adaptive to improve the speed and convergence accuracy. Then use the total variation (TV) of the L1 model to estimate the regional point spread function (PSF), and adopted the edge density weight to reduce small edge’s interference for the PSF estimates. Eventually to restored image by Wiener filter. Through experimental analysis, compared with other algorithms, our algorithms get better results in the space varies motion-blurred image.

• Periodical: Applied Mechanics and Materials (Volumes 687-691)
• Pages: 3591-3595
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.687-691.3591
• Online since: November 2014
• Authors: Jiang Yang Chen*, Xi Ling Luo
• Keywords: Background Modeling, Image Restoration, Motion Blur, TV-L1 Model

References

[1] Wiener. N. Extrapolation, Interpolation, and Smoothing of Stationary Time Series. U.S.: The MIT Press, (1964).

[2] Lucy. L. B. An Iterative Technique for the Rectification of Observed Distributions. Astronomical Journal, 1974, 79: 745-754.

DOI: 10.1086/111605

[3] Fergus. R. Removing Camera Shake from a Single Photograph. ACM Trans. Graphics, 2006, 25: 787-794.

DOI: 10.1145/1141911.1141956

[4] Shan. Q. High-quality Motion Deblurring from a Single Image. ACM Trans. Graphics, 2008, 27: 73.

[5] Jia. J. Single Image Motion Deblurring using Transparency[A]. in: Bir Bhanu, Nalini Ratha. Computer Vision and Pattern Recognition. Minneapolis, MN: IEEE 2007, 1-8.

DOI: 10.1109/cvpr.2007.383029

[6] Hee Seok Lee. Dense 3D Reconstruction from Severely Blurred Images using a Single Moving Camera. in: Patrick Kellenberger, Computer Vision and Pattern Recognition. Portland, OR: IEEE 2013, 273-280.

DOI: 10.1109/cvpr.2013.42

[7] Hui Ji. A Two-stage Approach to Blind Spatially-varying Motion Deblurring. Computer Vision and Pattern Recognition Providence, RI: IEEE 2012, 73-80.

DOI: 10.1109/cvpr.2012.6247660

[8] Paramanand, C. Non-uniform Motion Deblurring for Bilayer Scenes[A]. in: Patrick Kellenberger. Computer Vision and Pattern Recognition. Portland, OR: IEEE 2013, 1115-1122.

DOI: 10.1109/cvpr.2013.148

[9] P. KaewTraKulPong. An Improved Adaptive Background Mixture Model for Realtime Tracking with Shadow Detection. Video-Based Surveillance Systems: Springer US 2002, 135-144.

DOI: 10.1007/978-1-4615-0913-4_11

[10] C. Stauffer, W. Grimson. Adaptive Background Mixture Models for Real-time Tracking. Computer Vision and Pattern Recognition Fort Collins, CO: IEEE 1999, 246-252.

DOI: 10.1109/cvpr.1999.784637

[11] Li Xu, Jiaya Jia, Two-Phase Kernel Estimation for Robust Motion Deblurring[A]. in: K. Daniilidis, P. Maragos, N. Paragios (Eds), ECCV 2010, Part I, LNCS[C]. Berlin Heidelberg: Spring-Verlag, 2010. 157-170.

DOI: 10.1007/978-3-642-15549-9_12