Multimodal image fusion based on point-wise mutual information

作者：

Highlights：

• We introduce PMI into image fusion framework to measure the similarities between features of the source and fused images.

• A novel feature selection method is proposed to represent the image-wise operations.

• The proposed fusion method is archived the state-of-the-art performance by using the most fundamental gradient features.

摘要

•We introduce PMI into image fusion framework to measure the similarities between features of the source and fused images.•A novel feature selection method is proposed to represent the image-wise operations.•The proposed fusion method is archived the state-of-the-art performance by using the most fundamental gradient features.

论文关键词：Image fusion,Multimodal,Point-wise mutual information,Markov random field model,Gradient domain

论文评审过程：Received 16 July 2020, Revised 24 September 2020, Accepted 14 October 2020, Available online 19 October 2020, Version of Record 12 January 2021.

论文官网地址：https://doi.org/10.1016/j.imavis.2020.104047