Road layers are useful in preparing basemaps and analysis workflows for urban planning and development, change detection, infrastructure planning, and a variety of other applications. Digitizing roads from imagery is a time-consuming task and is commonly done by digitizing features manually. Deep learning models are highly capable of learning these complex semantics and can produce better results. Use this deep learning model to automate the task of road extraction and reduce the time and effort required.
License requirements
To complete this workflow, the following are the license requirements:
- ArcGIS Desktop—ArcGIS Image Analyst
- ArcGIS Enterprise—ArcGIS Image Server with raster analytics configured
- ArcGIS Online—ArcGIS Image for ArcGIS Online
Model details
This model has the following characteristics:
- Input—Raster, mosaic dataset, or image service (1-meter spatial resolution).
- Output—Binary raster representing road centerlines.
- Compute—This workflow is compute intensive and a GPU with compute capability of 6.0 or higher is recommended.
- Applicable geographies—This model is designed to work well globally.
- Architecture—This implementation is based on the paper by Songtao He et al.
- Accuracy metrics—This model has an F1 score of 0.76 on a city-scale dataset, precision of 0.80, and recall of 0.72.
Access and download the model
Download the Road Extraction—Global pretrained model from ArcGIS Living Atlas of the World. Alternatively, access the model directly from ArcGIS Pro using the Detect Objects Using Deep Learning tool, or consume it in ArcGIS Image for ArcGIS Online.
- Browse to ArcGIS Living Atlas of the World.
- Sign in with your ArcGIS Online credentials.
- Search for Road Extraction—Global and open the item page from the search results.
- Click the Download button to download the model.
You can use the downloaded .dlpk file directly in ArcGIS Pro, or upload and use it in ArcGIS Enterprise. Additionally, you can fine-tune the pretrained model if necessary.
Release notes
The following are the release notes:
| Date | Description |
|---|---|
| January 2022 |
|