Publication: Road Extraction from UAV Images Using a Deep ResDCLnet Architecture [Extraction de routes d’images de drones au moyen d’une architecture de réseau profond ResDCLnet]
0
0
Issued Date
2021
Resource Type
Language
eng
File Type
application/pdf
ISSN
7038992
Other identifier(s)
2-s2.0-85106318869
Rights Holder(s)
มหาวิทยาลัยศรีนครินทรวิโรฒ
Bibliographic Citation
Canadian Journal of Remote Sensing. Vol 47, No.3 (2021), p.450-464
Suggested Citation
Boonpook W., Tan Y., Bai B., Xu B. Road Extraction from UAV Images Using a Deep ResDCLnet Architecture [Extraction de routes d’images de drones au moyen d’une architecture de réseau profond ResDCLnet]. Canadian Journal of Remote Sensing. Vol 47, No.3 (2021), p.450-464. doi:10.1080/07038992.2021.1913046 Retrieved from: https://hdl.handle.net/20.500.14740/8049
Author(s)
Abstract
Obtaining near real-time road features is very important in emergent situations like flood and geological disaster cases. Remote sensing images with very high spatial resolution usually have many details in land use and land cover, which complicate the detection and extraction of road features. In this paper, we propose a deep residual deconvolutional network (Deep ResDCLnet), to extract road features from unmanned aerial vehicle (UAV) images. This proposed network is based on the deep neural network from SegNet architecture, the rich skip connection in a residual bottleneck, and the direct relationship among intermediate feature maps from the pixel deconvolution algorithm. It can improve the performance of a supervised learning model by differentiating and extracting complex road features on aerial photographs and UAV imagery. The proposed network is evaluated with the standard public Massachusetts road dataset and the UAV dataset collected alongside Yangtze River, and is compared with four state-of-art network architectures. The results show that the Deep ResDCLnet outperforms all four networks in terms of extraction accuracy, which demonstrates the effectiveness of the network in road extraction from very high spatial resolution imagery. ©, Copyright © CASI.
Subject(s)
Antennas
Arts computing
Deconvolution
Deep neural networks
Drones
Extraction
Feature extraction
Image enhancement
Image resolution
Land use
Remote sensing
Roads and streets
Aerial Photographs
Deconvolution algorithm
Extraction accuracy
Geological disaster
Land use and land cover
Remote sensing images
Road extraction
Very high spatial resolutions
Network architecture
Arts computing
Deconvolution
Deep neural networks
Drones
Extraction
Feature extraction
Image enhancement
Image resolution
Land use
Remote sensing
Roads and streets
Aerial Photographs
Deconvolution algorithm
Extraction accuracy
Geological disaster
Land use and land cover
Remote sensing images
Road extraction
Very high spatial resolutions
Network architecture