A Modern Image Quality Measurement Method for Blind Image Restoration
-
摘要: 盲图像复原过程中,图像质量评价至关重要. 通过分析重构图像质量与其总变分值之间的关系, 提出了用于图像复原的一种基于总变分(Total bounded variation, TBV)的图像质量评估方法, 并构建关系模型, 证明了原始清晰图像的总变分值在所有模糊图像中具有极大值, 且在所有重构图像的变分值中具有极小值. 通过分析, 得出结论: 当总变分取极值时, 基于所提度量方法, 可以获得更好的盲图像重构效果. 最后, 比较了原始清晰图像、模糊图像和重构图像之间的变分值, 计算机仿真验证了该方法的有效性和准确性.Abstract: In the process of blind image restoration, image quality assessment is of paramount importance. In this paper, A novel image quality assessment method is presented by analyzing the relation between reconstructed image quality and its total bounded variation (TBV), on this basis, the relationship model is constructed, that is, the original clear image's TBV is maximum in all the blurring image, and it is minimal in all the reconstructed image. Further, based on the metric method proposed, the better blind image reconstruction effect is obtained when the TBV is extremal. Finally, the TBV of original clear image, blurred images and blind restored images are compared, the simulation results shows the validation and veracity of the method proposed.
-
[1] Liu X W, Huang L H. Split Bregman iteration algorithm for total bounded variation regularization based image deblurring. Journal of Mathematical Analysis and Applications, 2010, 372(2): 486-495[2] Lieu L H, Vese L A. Image restoration and decomposition via bounded total variation and negative Hilbert-Sobolev spaces. Applied Mathematics and Optimization, 2008, 58(2): 167-193[3] Kirova Y M, Pena P C, Hijal T, Fournier-Bidoz N, Laki F, Sigal-Zafrani B. Improving the definition of tumor bed boost with the use of surgical clips and image registration in breast cancer patients. International Journal of Radiation Oncology Biology Physics, 2010, 78(5): 1352-1355[4] Rudin L I, Osher S, Fatemi E. Nonlinear total vriation based noise removal algorithms. Physica D, 1992, 60(4): 259-268[5] Blomgren P V. Total variation method for restoration of vector valued images [Ph.D. dissertation], University of California, USA, 1998[6] Maleki M, Latifi M, Amani-Tehran M. Definition of structural features of nano coated webs by image processing methods. International Journal of Nanotechnology, 2009, 6(12): 1131-1154[7] Quarello E, Trabbia A. High-definition flow combined with spatiotemporal image correlation in the diagnosis of fetal coarctation of the aorta. Ultrasound in Obstetrics and Gynecology, 2009, 33(3): 365-367[8] Cheng X G, An M W, Chen Q M. Image distortion metric based on total bounded variation. China Communications, 2012, 9(2): 79-85[9] Cheng Xiao-Gang, Chen Qi-Mei, Liu Guo-Qing. The relation between total bounded variation and image definition detection. Journal of Beijing University of Posts and Telecommunications, 2009, 32(S1): 120-122, 139(成孝刚, 陈启美, 刘国庆. 总有界变差与图像清晰度之间的关系. 北京邮电大学学报, 2007, 32(S1): 120-122, 139)[10] Awwal A A S, Rice K, Taha T. Fast implementation of matched-filter-based automatic alignment image processing. Optics and Laser Technology, 2009, 41(2): 193-197[11] Saad M A, Bovik A C, Charrier C. A DCT statistics-based blind image quality index. IEEE Signal Processing Letters, 2010, 17(6): 583-586[12] Li Bo, Su Zhi-Xun, Liu Xiu-Ping. An adaptive PDE image processing method based on Lp norm. Acta Automatica Sinica, 2008, 34(8): 849-853(李波, 苏志勋, 刘秀平. 基于Lp范数的局部自适应偏微分方程图像恢复. 自动化学报, 2008, 34(8): 849-853)[13] Moorthy A, Bovik A. Blind image quality assessment: from natural scene statistics to perceptual quality. IEEE Transactions on Image Processing, 2011, 20(12): 3350-3364
点击查看大图
计量
- 文章访问数: 1692
- HTML全文浏览量: 70
- PDF下载量: 1132
- 被引次数: 0