The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLIII-B1-2020
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLIII-B1-2020, 465–469, 2020
https://doi.org/10.5194/isprs-archives-XLIII-B1-2020-465-2020
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLIII-B1-2020, 465–469, 2020
https://doi.org/10.5194/isprs-archives-XLIII-B1-2020-465-2020

  06 Aug 2020

06 Aug 2020

GAN-BASED SYNTHESIS OF DEEP LEARNING TRAINING DATA FOR UAV MONITORING

D. Langenkämper, R. van Kevelaer, T. Möller, and T. W. Nattkemper D. Langenkämper et al.
  • Biodata Mining Group, Technical Faculty, Bielefeld University, 33615 Bielefeld, Germany

Keywords: deep learning, computer vision, inspection, Generative Adversarial Networks, low-shot learning

Abstract. Wind energy is a critical part of overcoming the use of fossil or nuclear energy usage. The price pressure on the renewable industry sector demands to cut the costs for costly regular inspections carried out by industrial climbers. Drone-based video-inspection reduces costs as well as increases the safety of inspection personal. To further increase the throughput, automatic or semi-automatic solutions to analyze these videos are needed. However, modern machine learning architectures need a lot of data to work reliably. This is by design a problem, as structural damage is rather rare in industrial infrastructure. Our proposed approach uses Generative Adversarial Networks to generate synthetic unmanned aerial vehicle imagery. This allows us to create a large enough training dataset (> 103) from a dataset, which is at least an order of magnitude smaller (approx. 102). We show that we can increase the classification accuracy of up to 6 percentage points.