DiResNet: Direction-aware Residual Network for Road Extraction in VHR Remote Sensing Images
arXiv:2005.07232 · doi:10.1109/TGRS.2020.3034011
Abstract
The binary segmentation of roads in very high resolution (VHR) remote sensing images (RSIs) has always been a challenging task due to factors such as occlusions (caused by shadows, trees, buildings, etc.) and the intra-class variances of road surfaces. The wide use of convolutional neural networks (CNNs) has greatly improved the segmentation accuracy and made the task end-to-end trainable. However, there are still margins to improve in terms of the completeness and connectivity of the results. In this paper, we consider the specific context of road extraction and present a direction-aware residual network (DiResNet) that includes three main contributions: 1) An asymmetric residual segmentation network with deconvolutional layers and a structural supervision to enhance the learning of road topology (DiResSeg); 2) A pixel-level supervision of local directions to enhance the embedding of linear features; 3) A refinement network to optimize the segmentation results (DiResRef). Ablation studies on two benchmark datasets (the Massachusetts dataset and the DeepGlobe dataset) have confirmed the effectiveness of the presented designs. Comparative experiments with other approaches show that the proposed method has advantages in both overall accuracy and F1-score. The code is available at: https://github.com/ggsDing/DiResNet.
12 pages, 13 figures. IEEE Transactions on Geoscience and Remote Sensing, 2020
References in corpus (1)
Cited by in corpus (5)
- Bi-Temporal Semantic Reasoning for the Semantic Change Detection in HR Remote Sensing Images
- Adapting Segment Anything Model for Change Detection in HR Remote Sensing Images
- Stagewise Unsupervised Domain Adaptation with Adversarial Self-Training for Road Segmentation of Remote Sensing Images
- Looking Outside the Window: Wide-Context Transformer for the Semantic Segmentation of High-Resolution Remote Sensing Images
- Adversarial Shape Learning for Building Extraction in VHR Remote Sensing Images