Abstract | ||
---|---|---|
The road network plays an important role for traffic management, GPS navigation and many other applications. Extracting the road from a high remote sensing (RS) imagery has been a hot research topic in recent years. The road structure always changing as the terrain, thus, how to extract the features of road network and identify the roads from RS imagery efficiently still a challenging. In this paper, we propose a road extraction method for RS imagery using the deep convolutional neural network, which is designed based on the deep residual networks and take full advantages of the U-net. Road network data form Las Vegas, America, are used to validate the method, and experiments show that the proposed model of deep convolutional neural network can extract road network accurately and effectively. |
Year | DOI | Venue |
---|---|---|
2018 | 10.1109/GEOINFORMATICS.2018.8557042 | 2018 26TH INTERNATIONAL CONFERENCE ON GEOINFORMATICS (GEOINFORMATICS 2018) |
Keywords | Field | DocType |
Road network extraction, deep learning, remote sensing imagery, convolutional neural network | Residual,Data mining,Convolutional neural network,Computer science,Remote sensing,Terrain,Gps navigation,Network data,Artificial intelligence,Deep learning | Conference |
ISSN | Citations | PageRank |
2161-024X | 0 | 0.34 |
References | Authors | |
0 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yongyang Xu | 1 | 33 | 2.50 |
Yaxing Feng | 2 | 6 | 0.82 |
Zhong Xie | 3 | 34 | 12.55 |
Anna Hu | 4 | 0 | 0.34 |
Xueman Zhang | 5 | 0 | 0.68 |