Academy & Industry Research Collaboration Center (AIRCC)

Volume 12, Number 06, March 2022

Runway Extraction and Improved Mapping from Space Imagery


David A. Noever, PeopleTec, Inc., USA


Change detection methods applied to monitoring key infrastructure like airport runways represent an important capability for disaster relief and urban planning. The present work identifies two generative adversarial networks (GAN) architectures that translate reversibly between plausible runway maps and satellite imagery. The training capability was illustrated using paired images (satellite-map) from the same point of view and using the Pix2Pix architecture or conditional GANs. In the absence of available pairs, the CycleGAN architecture likewise showed that its four network heads (discriminator-generator pairs) also provided effective style transfer from raw image pixels to outline or feature maps. To emphasize the runway and tarmac boundaries, the experiments show that the traditional grey-tan map palette is not a required training input but can be augmented by higher contrast mapping palettes (redblack) for sharper runway boundaries. The research highlights a potentially novel use case (called “sketch2satellite”) where a human sketches the current runway boundaries and automates the machine output of plausible satellite images. Finally, faulty runway maps were identified where the published satellite and mapped runways disagree, but an automated update renders the correct map using GANs.


Generative Adversarial Networks, Satellite-to-Map, Pix2Pix, CycleGAN Architecture.