Click:

The Last Update Time:--

赵鹏程
+
Personal Information
  • Supervisor of Master's Candidates
  • Name (Pinyin):Zhao Pengcheng
  • Date of Birth:1993-09-05
  • E-Mail:
  • Date of Employment:2019-12-07
  • Administrative Position:高级实验师
  • Education Level:With Certificate of Graduation for Doctorate Study
  • Business Address:武汉大学信息学部遥感信息工程学院（5号楼）315办公室
  • Gender:Male
  • Contact Information:+86 15972003670
  • Status:Employed
  • Alma Mater:武汉大学
  • Teacher College:School of Remote Sensing and Information Engineering
Other Contact Information
  • PostalAddress:
  • OfficePhone:
  • Telephone:
  • Email:
Paper Publications
Current position: Home   >   Scientific Research   >   Paper Publications

Category attention guided network for semantic segmentation of Fine-Resolution remote sensing images

  • Date of Publication:2025-01-07
  • Hits:

  • DOI number: 

    10.1016/j.jag.2024.103661

  • Affiliation of Author(s): 

    School of Remote Sensing and Information Engineering, Wuhan University, China

  • Journal: 

    INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION

  • Key Words: 

    Category attention,Semantic segmentation,Remote sensing images,CNN,Transformer

  • Abstract: 

    The semantic segmentation task is an essential issue in various fields, including land cover classification and cultural heritage investigation. The CNN and Transformer have been widely utilized in semantic segmentation tasks due to notable advancements in deep learning technologies. However, these methodologies may not fully account for remote sensing images' distinctive attributes, including the large intra-class variation and the small inter-class variation. Driven by it, we propose a category attention guided network (CAGNet). Initially, a local feature extraction module is devised to cater to striped objects and features at different scales. Then, we propose a novel concept of category attention for remote sensing images as a feature representation of category differences between pixels. Meanwhile, we designed the Transformer-based and CNN-based category attention guided modules to integrate the proposed category attention into the global scoring functions and local category feature weights, respectively. The network is designed to give more attention to the category features by updating these weights during the training process. Finally, a feature fusion module is developed to integrate global, local, and category multi-scale features and contextual information. A series of extensive experiments along with ablation studies on the UAVid, Vaihingen, and Potsdam datasets indicate that our network outperforms existing methods, including those based on CNN and Transformer.

  • Co-author: 

    Li, Pengcheng,Zhao, Shaohua, Qingwu, Jiayuan,Wang,Hu, Mingyao,Ai, Shunli

  • Indexed by: 

    Journal paper

  • Discipline: 

    Engineering

  • Document Type: 

    J

  • Volume: 

    127

  • ISSN No.: 

    1569-8432

  • Translation or Not: 

    no

  • CN No.: 

    Scopus:2-s2.0-85183550122,WOS:001171572600001

  • Date of Publication: 

    2024-03-01

Copyright @ 2017 Wuhan University