基于时空域三维约束的多帧图像超分辨算法

Multi-frame Image Super-resolution Algorithm Based on Space-time Three-dimensional Constraints

对基于时空域三维正则约束的多帧图像超分辨算法进行了研究,首先在基于马尔科夫随机场的自适应正则约束的基础上,进一步引入多帧图像间的时域约束,建立了时空域三维正则约束模型,并提出了相应的超分辨方法;之后通过仿真实验对本文方法进行了仿真测试,结果表明本文方法能够取得更好的超分辨效果。

The multi-frame image of super-resolution algorithm based on time-space three-dimensional regular constraint is studied. Firstly, based on the adaptive regular constraint of the Markov random field, the time-domain constraint between multi-frame images is further introduced, the time-space three-dimensional regular constraint model is established, and the corresponding super-resolution method is proposed. Afterwards, the simulation test is carried out through simulation experiments. The results show that the proposed method can better achieve super-resolution effect.