Medical image fusion based on convolutional neural networks and non-subsampled contourlet transform

作者：

Highlights：

• Solved the problem that CNN cannot be directly used for medical image fusion.

• Using CNN to replace artificially designed fusion rules.

• Proposing a new CNN (PHF-CNN) to select high frequency subband coefficients.

摘要

•Solved the problem that CNN cannot be directly used for medical image fusion.•Using CNN to replace artificially designed fusion rules.•Proposing a new CNN (PHF-CNN) to select high frequency subband coefficients.

论文关键词：Medical image fusion,Convolutional neural network,Non-subsampled contourlet transform

论文评审过程：Received 18 October 2019, Revised 14 July 2020, Accepted 2 January 2021, Available online 8 January 2021, Version of Record 20 January 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.114574