基于稀疏表示的超分辨率图像重建

Image Super-resolution Reconstruction Based on Sparse Representation

针对单幅低分辨率图像的超分辨率重建,提出一种基于稀疏表示的改进算法。通过联合输入低分辨率图像块和对应生成的高分辨率图像块,求解其在高低分辨率字典对上的稀疏表示系数,再将系数与高分辨率字典结合,修正输出的高分辨率图像块。仿真实验表明,文中提出的算法有效提升了重建图像的质量。

A super-resolution reconstruction approach to single low-resolution image based on sparse representa- tion is presented. We solve the sparse linear combination of the dictionary and generated through jointing input low-resolution image high-resolution image patches, and then use the coefficients to adjust the output high-resolution image patches. The results of simulation experiments show that the proposed algorithm is effective to improve the quality of the reconstruction of high-resolution image.