DeltaSat: Automatic Change Detection on Commercial Satellite Imagery

With many existing commercial satellites currently orbiting the earth, and many more to be launched in the near future, a massive amount of image data is available.

This presents a low-cost opportunity to quickly detect important changes throughout the world by processing the data efficiently and effectively. To address this opportunity, Stottler Henke built the DeltaSat prototype for automatic change detection on satellite imagery.

DeltaSat incorporates ortho-rectification, geo-referencing, homographic projection, and linear normalization, as well as state-of-the-art image processing techniques such as adaptive thresholding and contour detection. DeltaSat also implemented prototype change classification, using artificial intelligence techniques to automate human analysts’ cognitive processes when interpreting satellite imagery.