DA-RoadNet: A Dual-Attention Network for Road Extraction From High Resolution Satellite Imagery

Abstract

Recent advances in deep-learning methods have shown extraordinary performance in road extraction from high resolution satellite imagery. However, most existing deep-learning network models yield discontinuous and incomplete results because of shadows and occlusions. To address this problem, a Dual-Attention Road extraction Network (DA-RoadNet) with a certain semantic reasoning ability is proposed. Firstly, DA-RoadNet is designed based on a shallow encoder-to-decoder network with densely connected blocks, which can minimize the loss of road structure information caused by multiple downsampling operations. Moreover, by constructing a novel attention mechanism module, the proposed network is able to explore and integrate the invisible correlations among road features with their global dependency in spatial and channel dimension respectively. Finally, considering that the proportion of road samples is small in the satellite imagery, a hybrid loss function is appended to handle class imbalance, which enables the network model to train stablely and avoid local optimum. The validation experiments using two open road datasets demonstrate that the proposed DA-RoadNet can effectively solve discontinuous problems and preserve integrity of the extracted roads, thus resulting in a higher accuracy of road extraction compared with other developed stateof-the-arts. The considerable performance on the two challenging benchmarks also proves the powerful generation ability of our method.

Semantic Segmentation and Edge Detection—Approach to Road Detection in Very High Resolution Satellite Images

“…Traditional road extraction methods are primarily based on the assumption that the grayscale value inside the road is relatively consistent and contrasts with the surrounding objects, such as trees and buildings, ensuring road area distinguishability and severability. In contrast to traditional methods, deep learning-based methods rely on advances in feature learning and parameter sharing, which can be used to achieve automatic and efficient road extraction [11,12]. Deep learning techniques have made significant advancements in image object segmentation.…”

Discriminative Context-Aware Network for Target Extraction in Remote Sensing Imagery

“…The purpose of RCNN unit is to generate rich abstractions of visual features, while the purpose of the "refine block" in the RD module is to reduce the semantic gap between the low-level features and the high-level features. DA-RoadNet [65] designs dual-attention module to capture road features with their global dependencies. The main difference between DA-RoadNet and RD module is that DA-RoadNet lacks the consideration of reducing the semantic gap between low-level features and high-level features.…”

A Residual Attention and Local Context-Aware Network for Road Extraction from High-Resolution Remote Sensing Imagery

“…Xu et al [53] introduced the attention unit into the deep convolutional neural network to extract local and global information of the road in the remote sensing image and improve the accuracy of road network extraction. Wan et al [54] used a shallow encoder framework to construct a Dual-Attention Network (DA-RoadNet) to explore and analyze the correlation of road features in spatial dimensions and channel dimensions and then extract road information from a complex environment. Ren et al [55] designed a Dual-Attention Capsule U-Net (DA-CapsUNet) to extract and fuse multi-scale context information of road networks by constructing a feature attention module.…”