Abstract
Automatic extraction of road from multi-source remote sensing data has always been a challenging task. Factors such as shadow occlusion and multi-source data alignment errors prevent current deep learning-based road extraction methods from acquiring road features with high complementarity, redundancy, and crossover. Unlike previous works that capture contexts by multi-scale feature fusion, we propose a dual attention dilated-LinkNet (DAD-LinkNet) to adaptively integrate local road features with their global dependencies by joint using satellite image and floating vehicle trajectory data. Firstly, a joint least-squares feature matching-based floating vehicle trajectory correction model is used to correct the floating vehicle trajectory; then a convolutional network model DAD-LinkNet based on a dual-attention mechanism is proposed, and road features are extracted from the channel domain and spatial domain of the target image in turn by constructing a dual-attention module in the dilated convolutional layer and adopting a cascade connection; a weighted hyperparameter loss function is used as the loss function of the model; finally, the road extraction is completed based on the proposed DAD-LinkNet model. Experiments on three datasets show that the proposed DAD-LinkNet model outperforms the state-of-the-art methods in terms of accuracy and connectivity.
Highlights
R OADS are an important public infrastructure and play an important role in the urbanization process and smart city construction
Studies show that the length and extent of global roads will expand dramatically in the 21st century, for example: by 2050, there are expected to be at least 25 million kilometers of new roads worldwide; a 60% increase in total road length compared to 2010. 90% of road construction is taking place in developing countries, including many areas with outstanding biodiversity and important ecosystem services, and the roads that penetrate these areas are a major driver of habitat loss and fragmentation, wildfires, overhunting, and other environmental degradation [3]
This study proposes a new road extraction method that uses high-resolution remote sensing imagery and floating vehicle trajectory data in a near real-time manner
Summary
R OADS are an important public infrastructure and play an important role in the urbanization process and smart city construction. 90% of road construction is taking place in developing countries, including many areas with outstanding biodiversity and important ecosystem services, and the roads that penetrate these areas are a major driver of habitat loss and fragmentation, wildfires, overhunting, and other environmental degradation [3]. These issues have received considerable attention from scientists and policy makers, and focused and sustainable efforts have been made to understand the complex environmental changes that occur in areas where roads are changing rapidly [4]. A variety of satellite remote sensing images (e.g., luminescent remote sensing data, MODIS, Landsat, Quickbird, SPOT, IKONOS, Gaofen, Worldview, etc.) are used to extract road information and assess urban conditions, which are critical for both natural and socio-economic issues [5]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
