|
|
- 2018
基于SSIM的自适应样本块图像修复算法
|
Abstract:
现有基于样本块的图像修复算法, 大多通过人工设定样本块大小来达到最佳修复效果, 缺乏自适应性; 此外, 对图像不同纹理和结构区域采用相同大小的样本块, 也不利于获得整体最优修复效果．为解决上述问题, 本文提出一种基于改进结构相似性的自适应样本块大小选取算法, 在传统的SSIM算法的基础上增加了梯度信息, 并通过结合样本块亮度、对比度和结构3个模块来衡量结构差异, 以此确定不同结构和纹理区域的最优样本块大小, 提高算法适应性, 改善修复效果．仿真实验结果表明, 当图像存在复杂的结构和纹理信息时, 本文算法仍然能够获得理想的修复效果．
The current exemplar-based algorithms lack adaptability due to manually determining the size of block. In addition，using a patch with the constant size is not suitable to obtain the optimal effect in different structure and texture regions. To address this problem，this paper puts forward an adaptive patch size selection method using improved structure similarity(SSIM). The gradient information is added to the traditional SSIM and is combined with the brightness，contrast ratio，and structure of patch to measure the structural difference. On this basis，the optimal size of patch in different structure or texture regions is determined，thus improving the adaptability，as well as the inpainting effect. The simulation results demonstrate the effectiveness of the proposed method even when complex structure or texture exists
| [1] | Cao F, Gousseau Y, Masnou S, et al. Geometrically guided exemplar-based inpainting[J]. <i>SIAM Journal on Imaging Science</i>, 2009, 4(4):1143-1179. |
| [2] | Smith L N, Elad M. Improve dictionary learning:Multiple dictionary updates and coefficient reuse[J]. <i>IEEE Signal Processing Letter</i>, 2013, 20(1):9-28. |
| [3] | Huang J B, Kang S B. Image completion using planar structure guidance[J]. <i>ACM Transactions on Graphics</i>, 2014, 33(4):129-1-129-10. |
| [4] | He Kaiming, Sun Jian. Computing nearest-neighbor fields via propagation-assisted KD-trees[C]//<i>Proceedings of<i> 2012 <i>IEEE Conference on Computer Vision and Pattern Recognition</i>. Providence, USA, 2012:111-118. |
| [5] | 谢琼, 张红英, 彭博. 基于模式相似性的图像修复算法[J]. 现代电子技术, 2013, 36(2):94-96. |
| [6] | Vadim F, Gabriele F, Pablo A, et al. Variational framework for non-local inpainting[J]. <i>Image Processing on Line</i>, 2015, 5:362-386. |
| [7] | Zhou Yatong, Li Lin. Research on weighted priority of exemplar-based image inpainting[J]. <i>Journal of Electronics</i>, 2012, 29(1):166-170. |
| [8] | Xie Qiong, Zhang Hongying, Peng Bo. Image inpainting algorithm based on pattern similarity[J]. <i>Modern Electronics Technique</i>, 2013, 36(2):94-96(in Chinese). |
| [9] | Meng Chunzhi, He Kai, Jiao Qinglan. Image completion method with adaptive patch size[J]. <i>Journal of Image and Graphics</i>, 2012, 17(3):337-341(in Chinese). |
| [10] | Wang Z, Bovik A C, Sheikh H T, et al. Image quality assessment:From error cisbility to structural similarity [J]. <i>IEEE Transactions on Image Processing</i>, 2004, 13(4):600-612. |
| [11] | Nan A, Xi X. An improved Criminisi algorithm based on a new priority function and updating confidence[C]// <i>IEEE<i> 2014 7<i>th International Conference on Biomedical Engineering and Informatics</i>. Dalian, China, 2014:885-889.</i></i></i></i></i></i></i></i></i></i></i></i> |
| [12] | Luo K, Li D, Feng Y. Depth-aided inpainting for disocclusion restoration of multi-view images using depth-image-based rendering[J]. <i>Journal of Zhenjiang University</i>:<i>Science A</i>, 2009, 10(12):1738-1749. |
| [13] | Criminisi A, Perez P, Toyama K. Object removal by exemplar-based inpainting[C]// <i>Proceedings of<i> 2003 <i>IEEE Computer Society Conference on Computer Vision and Pattern Recognition</i>. Wisconsin, USA, 2003:721-728. |
| [14] | Xu Zongben, Sun Jian. Image inpainting by patch propagation using patch sparsity[J]. <i>IEEE Transactions on Image Processing</i>, 2010, 13(9):1200-1212. |
| [15] | Anamandra S H, Chandrasckaran V. Exemplar-based color image inpainting using a simple and effective gradient function[C]//<i>International Conference on Image Processing and Computer Vision</i>. Las Vegas, USA, 2010:140-145. |
| [16] | Li S, Zhao M. Image inpainting with salient structure completion and texture propagation[J]. <i>Patter Recognition Letters</i>, 2011, 32(9):1256-1266. |
| [17] | Sairam V, Sarma R R, Balasubramanian S, et al. A unified framework for geometry and exemplar based image inpainting[C]//<i>IEEE<i> 2<i>nd International Conference on Image Information Processing</i>. Paris, France, 2014:511-515. |
| [18] | Li Zhangming, Hu Wenjin. A novel method for exemplar-based image inpainting[J]. <i>Journal of Information & Computational Science</i>, 2012, 9(3):761-769. |
| [19] | Rodriguez-Sanche Z, Garcfa J A, Fdez-Valdivia J. Image inpainting with nonsubsampled Contourlet transform[J]. <i>Pattern Recognition Letters</i>, 2013, 34(13):1508-1518. |
| [20] | Zhou H L, Zheng J M. Adaptive patch size determination for patch-based image completion [C] //<i>Proceedings of<i> 2010 <i>IEEE<i> 17<i>th International Conference on Image Processing</i>. Hong Kong, China, 2010:421-424. |
| [21] | 孟春芝, 何凯, 焦青兰. 自适应样本块大小的图像修复方法[J]. 中国图像图形学报, 2012, 17(3):337-341. |
