Researchers at the Massachusetts Institute of Technology (MIT) and Qatar Computing Research Institute have developed an artificial intelligence model that uses satellite imagery to tag road features in digital maps.
The RoadTagger model combines a convolutional neural network (CNN) and a graph neural network (GNN) to automatically predict the number of lanes and road types hidden by obstructions.
The CNN digests raw satellite imagery while the GNN segments the road into 20-meter tiles or graph nodes linked by lines; the CNN extracts road features and shares that data with its immediate neighbors.
RoadTagger analyzed occluded roads from digital maps of 20 U.S. cities, tallying lane numbers with 77% accuracy and deducing road types with 93% accuracy.
MIT's Sam Madden said, "Our goal is to automate the process of generating high-quality digital maps, so they can be available in any country."
From MIT News
View Full Article
Abstracts Copyright © 2020 SmithBucklin, Washington, DC, USA
No entries found