北京邮电大学学报

  • EI核心期刊

北京邮电大学学报 ›› 2023, Vol. 46 ›› Issue (2): 84-90.

• 模式识别与图像处理 • 上一篇    下一篇

基于道路中心线的分阶段弱监督遥感图像道路提取

王薇1,谢沈惟2,闫浩田2,张闯3,吴铭2   

  1. 1. 应急管理部国家减灾中心
    2. 北京邮电大学
    3. 北京邮电大学人工智能学院模式识别与智能系统实验室
  • 收稿日期:2022-03-15 修回日期:2022-04-19 出版日期:2023-04-28 发布日期:2023-05-14
  • 通讯作者: 张闯 E-mail:zhangchuang@bupt.edu.cn

Stagewise Weakly Supervised Satellite Imagery Road Extraction Based on Road Centerline Scribbles

  • Received:2022-03-15 Revised:2022-04-19 Online:2023-04-28 Published:2023-05-14
  • Contact: Chuang ZHANG E-mail:zhangchuang@bupt.edu.cn

摘要: 通过语义分割算法从卫星图像中提取道路已经成为道路遥感监测任务的主流解决方案.但由于不同地理环境导致的卫星图像中道路纹理复杂多变等特点以及道路的像素级标注成本昂贵等现实情况,利用大量道路的像素级标注用于训练语义分割模型是不实际的.针对上述问题,提出了一种基于道路中心线涂鸦的分阶段弱监督道路提取算法,以弱监督的方式学习道路中心线涂鸦的特征并分阶段地训练道路分割模型.此外,还提出了伪掩码更新策略和混合训练策略,设计了分别针对道路前景和道路背景的损失函数.对比实验结果表明,新算法在道路分割任务中比其他基于道路中心线的弱监督方法取得了更优的表现,消融实验的结果也验证了所提出训练策略的有效性.

关键词: 道路提取, 图像分割, 弱监督学习, 遥感图像

Abstract: Extracting roads from satellite images through semantic segmentation algorithm has become the mainstream solution for RS-based road monitoring tasks. However, due to complex features and changeable textures of roads in satellite imagery which derive from various geographical environments and the high cost of pixel-level road labeling, it is unaffordable to acquire a substantial dataset with pixel-level road annotation to train semantic segmentation models. To solve the above problems, a stagewise weakly supervised road extraction algorithm based on road centerline scribbles is proposed. The feature of road centerline scribbles is learned in a weakly supervised way, and the road segmentation model is trained by stages. In addition, the pseudo mask update strategy and the hybrid training strategy are proposed, and the loss functions for road foreground and road background are designed. The results show that compared with other weak supervision methods based on road centerline, the proposed algorithm achieves superior performance in road segmentation task. Ablation studies are also conducted to verify the effectiveness of the proposed training strategy.

Key words: road extraction, image segmentation, weakly supervised learning, remote sensing imagery

中图分类号: