Marketplace

Sports Facilities Detection

Detects baseball fields, tennis courts, soccer fields, and stadiums in cities and towns.

Description

Sports Detection is a Processing Block for identifying sports facilities such as baseball parks, tennis courts, soccer fields, and stadiums in satellite images in .png or .jpeg or .jpg or .tiff format. Block can detect sports facilities in images with ground sampling distance (GSD) of 0.55m or less. Output is provided as images with detection bounding boxes overlayed on the sports facilites and detection details in a .json file.The block is trained on data-set obtained from South-Eastern Asia.

The algorithm uses generative deep learning techniques and CNN-based Artificial Neural Network architecture to achieve the computer vision objective. The solution is built in Python and uses Tensorflow at backend as deep learning framework. The algorithm processes satellite images in GeoTIFF, TIFF, PNG, JPG or JPEG formats, with no limitations on image dimensions.

The use cases for this block are infrastructure monitoring, urban planning and construction.

General InformationDescription
Block TypeProcessing
Supported input dataInput is required as a set of PNG, JPEG or TIFF images. The image is expected to have Ground Sampling Distance (GSD) less than 0.55 m
Output data formatThe output is resultant image with detection bounding boxes overlayed onto the input images
Algorithm Training Data DetailsThe algorithm has been trained using custom built data sets from satellite images captured over South Eastern Asian region.
Performance0.5 IoU and has detection of 0.5mAP on satellite images with GSD 0.55m

Fore more information, visit the provider website here.

Capabilities

Input Capabilities

raster
up42_standard
format
{
  "or": [
    "GTiff",
    "PNG",
    "JPEG"
  ]
}
resolution
0.5

Output Capabilities

vector
custom
object_type
sports
up42_standard
format
GeoJSON
geometry_type
Polygon

End User License Agreement

View the End User License Agreement conditions in a new tab.