← Marketplace
Large Vehicle Detection
Detect trucks, buses, tractor-trailers and other land-based large vehicles.
Description
Large Vehicle Detection detects large vehicles (trucks, tractor-trailer, buses) in satellite images in both urban and rural settings. The block can detect vehicles in images with ground sampling distance (GSD) around 0.5 m.
Output is provided as images with detection bounding boxes overlayed on the vehicles and additional GeoJSON file. The algorithm is trained on a data-set obtained from South-Eastern Asia.
Supported workflows
Data platform:
- Catalog -> Pléiades Display -> Order and get the asset from storage of UP42
- In Projects workflows, Processing from Storage -> DIMAP GeoTIFF conversion -> Large Vehicle Detection
Data blocks:
- Pléiades Display (Streaming) -> Building Bounding Box Detector
Note: This block needs to be used without Raster Tiling.
Technical description
The algorithm uses deep learning techniques and RetinaNet detection architecture to detect large vehicles such as trucks, buses, and tractor-trailers. Most of the setting are factories, ports, highways, parking lots, industrial parks, etc.
Use cases
Counting large vehicles in parking lots, highway design, port design, determine retail and industrial activity, determining large vehicle sales/export, smart city applications, asset tracking.
| General Information | Description |
|---|---|
| Block Type | Processing |
| Supported input data | The GeoTIFF input image is expected to have Ground Sampling Distance (GSD) less than 0.55 m. PNG and JPEG formats are also supported. |
| Output data format | The output is the resultant image with the detected bounding boxes in a GeoJSON file. |
| Algorithm Training Data Details | The algorithm is trained on a dataset obtained from South Asia. Most of the settings are dense urban areas, neighborhoods, parking lots, industrial parks, etc. |
For more information, visit the provider website here.
Capabilities
Terms & Conditions
View the End User License Agreement conditions.