Abstract:
Human Progress in the past centuries was fueled and enhanced by the expansion and growth of global transportation networks, mainly roads. Road networks are continuously e...Show MoreMetadata
Abstract:
Human Progress in the past centuries was fueled and enhanced by the expansion and growth of global transportation networks, mainly roads. Road networks are continuously evolving and expanding. Many services and industries rely on that these roads are well known so that they can be used to plan and organize their activities. This has led many researchers to attempt to extract road networks from remotely sensed images. This can be challenging due to the complexity of road networks and the obstacles present in observing them from satellite images. In this study we present a road segmentation method using a dual-network approach with dilated convolution and enhancement modules. Additional tests were conducted on a multitude of datasets and multiple networks were reproduced and introduced in the tests to generate comprehensive test results.
Published in: 2021 Tenth International Conference on Intelligent Computing and Information Systems (ICICIS)
Date of Conference: 05-07 December 2021
Date Added to IEEE Xplore: 03 February 2022
ISBN Information: