This is a processing block that detects changes on 512x512 GeoTIFF images
Infrastructure Change Detection, based on SPOT or Pléiades imagery, detects changes in man-made infrastructures such as roads, buildings and houses, earthworks, etc. It is intended for urban development monitoring.
Assuming tiled SPOT or Pléiades images as an input, the Change Detection algorithm uses models based on deep learning methods to automatically provide a binary change / no change report as an output.
The main objective of change detection is to help image analyst to focus on specific location and thus prioritize their activities and deliver faster reports. It can also be used to measure the surface that has changed and thus provide statistics at the level of a county or city. Possible use cases: urban development, maps/databases update, conflict areas monitoring, illegal activities monitoring, etc.
Note: This block currently fails if less or more than two images are given. Set the following parameters to run it successfully with Pléiades / SPOT:
- limit set to 2
- contains filter instead of intersects or bbox
- match_image_extents (tiling) set to True
|Supported input data||Input tile size between 256x256 and 1024x1024 pixels. Compatibility with jpg, png, 3 or 4 bands, tif 3 or 4 bands (8 bits RGB), Resolution 2.4 m, Image padding 32|
|Output data format||GeoJSON vector file in geographic projection containing all changes as polygons|
|Performance||Algorithm qualified on a large set of data encompassing different types of landscapes worldwide by Airbus R&D and operational team, with a high level of recall performance. High recall but low precision, so results might contain a significant amount of over detections. Works better with images of the same season, no snow and restrained angles on dense cities and high buildings.|
Terms & Conditions
View the End User License Agreement conditions.