Abstract
Automatic road extraction from remote sensing images plays an important role for navigation, intelligent transportation, and road network update, etc. Convolutional neural network (CNN)-based methods have presented many achievements for road extraction from remote sensing images. CNN-based methods require a large dataset with high quality labels for model training. However, there is still few standard and large dataset, which is specially designed for road extraction from optical remote sensing images. Besides, the existing end-to-end CNN models for road extraction from remote sensing images are usually with symmetric structure, studying on asymmetric structure between encoding and decoding is rare. To address the above problems, this article first provides a publicly available dataset LRSNY for road extraction from optical remote sensing images with manually labelled labels. Second, we propose a reconstruction bias U-Net for road extraction from remote sensing images. In our model, we increase the decoding branches to obtain multiple semantic information from different upsamplings. Experimental results show that our method achieves better performance compared with other six state-of-the-art segmentation models when testing on our LRSNY dataset. We also test on Massachusetts and Shaoshan datasets. The good performances on the two datasets further prove the effectiveness of our method.
Highlights
B ENEFITING from the prosperity and development of navigation, automatic driving, smart city and intelligent transportation, etc., road network information plays a more and Manuscript received November 12, 2020; revised December 27, 2020; accepted January 19, 2021
We provide a publicly available road extraction dataset from high-resolution remote sensing images
We proposed a reconstruction bias U-Net for road extraction from high-resolution optical remote sensing images
Summary
B ENEFITING from the prosperity and development of navigation, automatic driving, smart city and intelligent transportation, etc., road network information plays a more and Manuscript received November 12, 2020; revised December 27, 2020; accepted January 19, 2021. Due to the new road construction, road network information update is always necessary. There are many kinds of methods for road network information update, such as manually labeling, tracking the changes of cars’ driving traces, and automatic road extraction from remote sensing images, etc. Tracking cars’ driving traces may miss the appearance information of roads, such as width, border, and so on. Automatic road extraction from optical remote sensing images is a more economic and more time saving way compared with the traditional manual road areas labeling [1]. Because the attracting of high research values about road extraction from remote sensing images, much research has focused on automatic road extraction from remote sensing images [1]–[8]
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
More From: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.