Automatic Road Extraction from High-Resolution Remote Sensing Images Using a Method Based on Densely Connected Spatial Feature-Enhanced Pyramid
Open Access
- 7 December 2020
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 14 (19391404) , 3-17
Abstract
Road extraction is an important task in remote sensing image information extraction. Recently, deep learning semantic segmentation has become an important method of road extraction. Due to the impact of the loss of multiscale spatial features, the results of road extraction still contain incomplete or fractured results. In this paper, we proposed a deep learning model, which is called the dense-global-residual network, that reduces the loss of spatial information and enhances context awareness. In the dense-global-residual network, the residual network is used to extract the features at different levels. To obtain more abundant multiscale features, a dense and global spatial pyramid pooling module based on Atrous Spatial Pyramid Pooling is built to perceive and aggregate the contextual information. The proposed method obtains better results on the GF-2 road dataset and public Massachusetts road dataset of aerial imagery. In order to prove the effectiveness of our method, we compared with four methods, such as DeepLabV3+, U-net, D-LinkNet and coord-dense-global model, the accuracy of our method is considerably better. Moreover, the dense-global-residual network can also effectively extract roads, especially trees and building shadows occlude the road. In addition, our method can successfully extract roads in regions of different development levels in universality experiments. This indicates that the proposed method can effectively maintain the completeness and continuity of roads and improve the accuracy of road segmentation from high-resolution remote sensing images.Keywords
Funding Information
- National Natural Science Foundation of China (41971311, 41901282)
- Major Projects of Science and Technology of Anhui (18030801111)
- Natural Science Foundation of Anhui Province (2008085QD188)
This publication has 47 references indexed in Scilit:
- Fully convolutional networks for semantic segmentation2015
- A Semi-Automatic Method for Road Centerline Extraction From VHR ImagesIEEE Geoscience and Remote Sensing Letters, 2014
- An Integrated Method for Urban Main-Road Centerline Extraction From Optical Remotely Sensed ImageryIEEE Transactions on Geoscience and Remote Sensing, 2013
- Automatic road extraction using high resolution satellite image based on texture progressive analysis and normalized cut methodJournal of the Indian Society of Remote Sensing, 2009
- Robust Higher Order Potentials for Enforcing Label ConsistencyInternational Journal of Computer Vision, 2009
- TextonBoost for Image Understanding: Multi-Class Object Recognition and Segmentation by Jointly Modeling Texture, Layout, and ContextInternational Journal of Computer Vision, 2007
- A Review of Techniques for Extracting Linear Features from ImageryPhotogrammetric Engineering & Remote Sensing, 2004
- Support Vector Machines for Road Extraction from Remotely Sensed Images2003
- Modeling contextual knowledge for controlling road extraction in urban areas2002
- Recognition of roads and bridges in SAR imagesPattern Recognition, 1998