Semantic segmentation of clouds in satellite images based on U-Net++ architecture and attention mechanism

作者：

Highlights：

• Proposed U-Net++ based network with attention mechanism for cloud segmentation.

• Proposed SEUNet++ achieves an IoU score of 91.8 %.

• Transfer learning helps to improve the segmentation results.

• SEUNet++ performs better than the original U-Net++ architecture.

• SEUNet++ improves the state-of-the-art by lifting the IoU score by 0.23 %.

摘要

•Proposed U-Net++ based network with attention mechanism for cloud segmentation.•Proposed SEUNet++ achieves an IoU score of 91.8 %.•Transfer learning helps to improve the segmentation results.•SEUNet++ performs better than the original U-Net++ architecture.•SEUNet++ improves the state-of-the-art by lifting the IoU score by 0.23 %.

论文关键词：U-Net++,ResNet,95-Cloud,Semantic segmentation,Cloud segmentation,Multispectral satellite data

论文评审过程：Received 2 February 2022, Revised 27 June 2022, Accepted 1 August 2022, Available online 5 August 2022, Version of Record 8 August 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118380