提出一种基于过完备字典稀疏表示的云图超分辨率算法。首先,联合训练针对低分辨率与高分辨率云图块的两个字典D_l和D_h,保证对应的低分辨率与高分辨率云图块关于各自的字典具有相似的稀疏表示;其次,通过求解优化问题,获得待处理云图每个低分辨率云图块关于D_l的稀疏表示,并将表示系数用于D_h以生成对应的高分辨率云图块;最后,运用最速下降算法,得到满足重构约束的高分辨率云图。红外与可见光云图的数值实验验证了本文算法的有效性,表明本文算法在视觉效果及PSNR指标上均优于插值方法。

Motivated by the fact that image patch can be sparse represented using a suitable over-complete dictionary, a nephogramsuper-resolution algorithm via sparse representation using over-complete dictionary is presented. During the experimenttwo dictionaries D_l and D_h for the low-resolution and high-resolution nephogram patches were trained jointly in order to guaranteethat the low-resolution and high-resolution patch pair possesses similar sparse representations as to their own dictionary.Through solving optimization problem, the sparse representation for each low-resolution nephogram patch with respect to D_l wasobtained, and the representation coefficients were applied to D_h in order to generate the corresponding high-resolution nephogrampatch. At the end of experiment the high-resolution nephogram which satisfies the reconstruction constraint was achieved by usinggradient descent algorithm. Numerical experiments for infrared and visual nephogram demonstrate the effectiveness of theproposed algorithm. Moreover, the proposed algorithm outperforms interpolation based methods in terms of visual quality andthe Peak Signal to Noise Ratio (PSNR).