Fully Convolutional Siamese Auto-encoder for Change Detection in UAV Aerial Images

  • Daniel Mesquita UFMG
  • Ronaldo Santos UFMG
  • Douglas Macharet UFMG
  • Mario Campos UFMG
  • Erickson Nascimento UFMG


Different applications in remote sensing, such as crop monitoring and visual surveillance, demand the automatic detection of changes from sets of images acquired over time. Most traditional approaches use satellite imagery, which besides the known issues such as cloud cover and image acquisition frequency for non-geostationary satellites, are very costly. In this context, with the recent technological advances, Unmanned Aerial Vehicles (UAVs) have become ubiquitous in numerous applications. In this paper, we present a Fully Convolutional Siamese Auto-encoder method for change detection in aerial images, in particular for those obtained with UAVs. We show that, by using an auto-encoder, we can further reduce the number of labeled samples required to achieve competitive results. We evaluated the performance of our approach on two different datasets, and results showed that our methodology outperforms state of the art, while demanding less training data.
Palavras-chave: Aerial images, change detection, fully convolutional neural networks (FCNNs), fully convolutional Siamese networks, unmanned aerial vehicles (UAVs).
Como Citar

Selecione um Formato
MESQUITA, Daniel; SANTOS, Ronaldo; MACHARET, Douglas; CAMPOS, Mario; NASCIMENTO, Erickson. Fully Convolutional Siamese Auto-encoder for Change Detection in UAV Aerial Images. In: CONFERENCE ON GRAPHICS, PATTERNS AND IMAGES (SIBGRAPI), 32. , 2019, Rio de Janeiro. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2019 . DOI: https://doi.org/10.5753/sibgrapi.2019.9826.