Moon Rover Image Super-Resolution Reconstruction Algorithm
-
摘要: 为了更好地满足嫦娥探月工程二期中月球车导航和探测规划任务对图像数据的要求，提出了一种基于压缩感知的超分辨率图像重建方法，利用经过模糊处理并加入噪声的低分辨率图像重建原始的高分辨率图像，实现了月球车图像的超分辨率重建。算法采用局部Sparse\|Land模型，从美国阿波罗计划获取的月面图像、嫦娥二期工程实验中获取的图像以及随机选取的自然图像中提取了大量训练图块，采用K-SVD算法完成了高、低分辨率过完备字典Ah和Al的学习，在对待重建图像进行有效分割的基础上，通过求解优化问题获得待处理低分辨率图块的稀疏表示，并将表示系数用于Ah以生成对应的高分辨率图块。最后，运用最小二乘算法，得到满足重构约束的高分辨率图像。实验结果表明，此算法在视觉效果及PSNR指标上均优于插值方法和Yang的方法。Abstract: For moon rover navigation and exploration mission during the 2nd stage of Chang’e project, high-resolution images are necessary. So a moon rover image super-resolution reconstruction algorithm via using compressed sensing was presented. The target is to reconstruct an original image from its blurred and down-scaled noisy version. The algorithm assumed a local Sparse-Land model on image patches, serving as regularization. The images from Apollo Project, tests in the 2nd stage of Chang’e project and natural image database were used in extracting patches for building two dictionaries. The K-SVD algorithm was used in training the dictionaries. Then the effective segmentation was implemented on low-resolution image. Through solving optimization problem via orthogonal matching pursuit algorithm, the sparse representation for each low-resolution image patch with respect to Al was obtained, and the representation coefficients were applied to Ah in order to generate the corresponding high-resolution image patch. At the end of experiment the high-resolution image which satisfied the reconstruction constraint was achieved by using least squares algorithm. Numerical experiments about moon rover images from tests in the 2nd stage of Chang’e project demonstrated the effectiveness of the proposed algorithm. Moreover, the proposed algorithm outperforms bicubic interpolation based method and the algorithm via Yang in terms of visual quality and the peak signal to noise ratio.
-
-
[1] 刘经南,魏二虎,黄劲松,等.月球测绘在月球探测中的应用[J]. 武汉大学学报.信息科学版, 2005,30(2):95-100 [2] Thevenaz P, Blu T, Unser M. Interpolation Revisited[J]. IEEE Trans Medieal Imaging, 2000,19(7):739-758 [3] Elad M, Aharon M. Image Denoising via Sparse and Redundant Tepresentations over Learned Dictionaries[J]. IEEE Trans on Image Processing,2006, 15(12): 3 736-3 745 [4] Bruckstein A M, Donoho D L, Elad M. From Sparse Solutions of Systems of Equations to Sparse Modeling of Signals and Images[J]. SIAM Review, 2009, 51(1): 34-81 [5] Elad M. Sparse and Redundant Representations: From Theory to Applications in Signal and Image Processing
[M]. Heidelberg: Springer, 2010[6] Schultz R R, Stevenson R L. A Bayesian Approach to Image Expansion for Improved Definition[J]. IEEE Transactions on Image Processing,1994,3(3): 233-242 [7] Aharon M, Elad M, Bruckstein A M. The K-SVD: An Algorithm for Designing of over Complete Dictionaries for Sparse Representation[J]. IEEE Trans on Signal Processing, 2006, 54(11):4 311-4 322 [8] Yang J, Wright J, Huang T,et al. Image Super-resolution Via Sparse Representation[J]. IEEE Trans on Image Processing, 2010,19(11):2 861-2 873
计量
- 文章访问数: 1525
- HTML全文浏览量: 77
- PDF下载量: 931
下载:
