嫦娥工程二期要求嫦娥3号的安全降落是最为关键的任务.因此，提出了一种基于压缩感知的超分辨率图像重建方法，根据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，实现了月球探测器着陆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球影像、嫦娥1、2号卫星影像和嫦娥工程二期试验中获取的月球探测器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Al和Ah的学习，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.实验验证了算法的有效性，表明其在视觉效果及PSNR和RMSE指标上均优于插值方法和Yang的方法.

Because the landing security of Chang'E-3 is the most critical requirements during the second stage of Chang'E project, the high-resolution landing image is necessary. The super-resolution reconstruction problem for the single Lunar Rover landing image was solved using compressed sensing theory. A super-resolution reconstruction algorithm for sparse representation by using over-complete dictionary was presented. The goal was 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, CE-1, CE-2 and tests of the second stage of Chang'E project were applied to extract patches for building two dictionaries. The K-SVD algorithm was adopted for training the dictionaries. Through solving optimization problem via Orthogonal Matching Pursuit algorithm, the sparse representation for each low-resolution landing image patch with respect to Al was obtained. The representation coefficients were applied to Ah in order to generate the corresponding high-resolution landing image patch. At the end of the experiment the high-resolution image which satisfied the reconstruction constraint was obtained by using least squares algorithm. Numerical experiments for Lunar Rover landing images from the tests of the second 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, the Peak Signal to Noise Ratio (PSNR) and Root Mean Square Error (RMSE).