PLE-Net: Automatic power line extraction method using deep learning from aerial images

作者：

Highlights：

• A novel power line extraction network (PLE-Net) is proposed to construct an endto-end segmentation method.

• A feature enhance block is proposed to capture rich contextual relationships from local feature maps.

• A self-attention block is proposed to embed into the proposed PLE-Net network to emphasize the regions about power lines.

摘要

•A novel power line extraction network (PLE-Net) is proposed to construct an endto-end segmentation method.•A feature enhance block is proposed to capture rich contextual relationships from local feature maps.•A self-attention block is proposed to embed into the proposed PLE-Net network to emphasize the regions about power lines.

论文关键词：Power line extraction,Deep architecture,Image segmentation,Multi-scale attention module

论文评审过程：Received 4 June 2021, Revised 6 January 2022, Accepted 25 February 2022, Available online 12 March 2022, Version of Record 18 March 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.116771