A general deep learning model for bird detection in high-resolution airborne imagery.

Journal: Ecological applications : a publication of the Ecological Society of America
PMID: 35708073

Abstract

Advances in artificial intelligence for computer vision hold great promise for increasing the scales at which ecological systems can be studied. The distribution and behavior of individuals is central to ecology, and computer vision using deep neural networks can learn to detect individual objects in imagery. However, developing supervised models for ecological monitoring is challenging because it requires large amounts of human-labeled training data, requires advanced technical expertise and computational infrastructure, and is prone to overfitting. This limits application across space and time. One solution is developing generalized models that can be applied across species and ecosystems. Using over 250,000 annotations from 13 projects from around the world, we develop a general bird detection model that achieves over 65% recall and 50% precision on novel aerial data without any local training despite differences in species, habitat, and imaging methodology. Fine-tuning this model with only 1000 local annotations increases these values to an average of 84% recall and 69% precision by building on the general features learned from other data sources. Retraining from the general model improves local predictions even when moderately large annotation sets are available and makes model training faster and more stable. Our results demonstrate that general models for detecting broad classes of organisms using airborne imagery are achievable. These models can reduce the effort, expertise, and computational resources necessary for automating the detection of individual organisms across large scales, helping to transform the scale of data collection in ecology and the questions that can be addressed.

Authors

  • Ben G Weinstein
    Department of Wildlife Ecology and Conservation, University of Florida, Gainesville, United States.
  • Lindsey Garner
    Department of Wildlife Ecology and Conservation, University of Florida, Gainesville, Florida, USA.
  • Vienna R Saccomanno
    California Oceans Program, The Nature Conservancy, Sacramento, California, USA.
  • Ashley Steinkraus
    Department of Wildlife Ecology and Conservation, University of Florida, Gainesville, Florida, USA.
  • Andrew Ortega
    Geomatics Program, University of Florida, Gainesville, Florida, USA.
  • Kristen Brush
    Montana State University, Bozeman, Montana, USA.
  • Glenda Yenni
    Department of Wildlife Ecology and Conservation, University of Florida, Gainesville, Florida, USA.
  • Ann E McKellar
    Environment and Climate Change Canada, Saskatoon, Saskatchewan, Canada.
  • Rowan Converse
    Center for the Advancement of Spatial Informatics Research and Education, University of New Mexico, Albuquerque, New Mexico, USA.
  • Christopher D Lippitt
    Department of Geography and Environmental Studies, University of New Mexico, Albuquerque, NM 87131, USA.
  • Alex Wegmann
    California Oceans Program, The Nature Conservancy, Sacramento, California, USA.
  • Nick D Holmes
    California Oceans Program, The Nature Conservancy, Sacramento, California, USA.
  • Alice J Edney
    Department of Zoology, University of Oxford, Oxford, UK.
  • Tom Hart
    Department of Zoology, University of Oxford, 11a Mansfield Road, Oxford, OX1 3SZ, UK. tom.hart@zoo.ox.ac.uk.
  • Mark J Jessopp
    School of Biological, Earth and Environmental Sciences, University College Cork, Cork, Ireland.
  • Rohan H Clarke
    School of Biological Sciences, Monash University, Melbourne, Victoria, Australia.
  • Dominik Marchowski
    Ornithological Station, Museum and Institute of Zoology, Polish Academy of Sciences, Gdańsk, Poland.
  • Henry Senyondo
    Department of Wildlife Ecology and Conservation, University of Florida, Gainesville, Florida, USA.
  • Ryan Dotson
    Quantaero, Nevada, Reno, USA.
  • Ethan P White
    Department of Wildlife Ecology and Conservation, University of Florida, Gainesville, United States.
  • Peter Frederick
    Department of Wildlife Ecology and Conservation, University of Florida, Gainesville, Florida, USA.
  • S K Morgan Ernest
    Department of Wildlife Ecology and Conservation, University of Florida, Gainesville, Florida, USA.

Keywords

External Resources

Popular Topics
Recent Journals