A noise-suppressing and edge-preserving multiframe super-resolution image reconstruction method

作者：

Highlights：

• We have proposed a diffusion-based regularizing functional to address the ill-posedness of the multiframe super-resolution problem.

• The new regularizer can effectively suppress noise, preserve critical image features, and enhance edges.

• The proposed regularizer contains a shape-defining parameter that is automatically updated to ensure that the corresponding energy potential is convex.

• The new super-resolution method can achieve higher resolution factors while maintaining stability and appealing results.

摘要

Highlights•We have proposed a diffusion-based regularizing functional to address the ill-posedness of the multiframe super-resolution problem.•The new regularizer can effectively suppress noise, preserve critical image features, and enhance edges.•The proposed regularizer contains a shape-defining parameter that is automatically updated to ensure that the corresponding energy potential is convex.•The new super-resolution method can achieve higher resolution factors while maintaining stability and appealing results.

论文关键词：Convex optimization,Super-resolution,Backward diffusion,Regularization

论文评审过程：Received 25 November 2014, Revised 1 March 2015, Accepted 2 March 2015, Available online 12 March 2015.

论文官网地址：https://doi.org/10.1016/j.image.2015.03.001