MLDNet: Multi-level dense network for multi-focus image fusion

作者：

Highlights：

• We propose a new variation in CNN to address the well-known problem of image fusion.

• We introduce a multi-level dense network for fusing multi-focus images.

• The proposed model uses both shallow and deep feature extraction to make a robust model.

• Structural similarity is utilized to monitor image quality loss in the network.

摘要

•We propose a new variation in CNN to address the well-known problem of image fusion.•We introduce a multi-level dense network for fusing multi-focus images.•The proposed model uses both shallow and deep feature extraction to make a robust model.•Structural similarity is utilized to monitor image quality loss in the network.

论文关键词：Multi-focus image fusion,Deep learning,Feature extraction,Feature fusion

论文评审过程：Received 10 August 2019, Revised 26 December 2019, Accepted 16 April 2020, Available online 21 April 2020, Version of Record 29 April 2020.

论文官网地址：https://doi.org/10.1016/j.image.2020.115864