[MARMAM] New paper using aerial imagery to train AI to detect whales in satellite imagery

Alex Borowicz Alex.Borowicz at stonybrook.edu
Wed Oct 9 12:43:44 PDT 2019

Dear Colleagues,
We're excited to announce our new open-access paper, in which we train a
deep-learning algorithm to detect whales in high-resolution satellite
imagery using images captured from aerial surveys. We hope that this can be
a method that will help fill in some of our survey data gaps and help
better target at-sea work such as tissue sampling and tag deployment.

Aerial-trained deep learning networks for surveying cetaceans from
satellite imagery

   Borowicz A, Le H, Humphries G, Nehls G, Höschle C, Kosarev V, et al.
   (2019) Aerial-trained deep learning networks for surveying cetaceans from
   satellite imagery. PLoS ONE 14(10): e0212532

   Most cetacean species are wide-ranging and highly mobile, creating
   significant challenges for researchers by limiting the scope of data that
   can be collected and leaving large areas un-surveyed. Aerial surveys have
   proven an effective way to locate and study cetacean movements but are
   costly and limited in spatial extent. Here we present a semi-automated
   pipeline for whale detection from very high-resolution (sub-meter)
   satellite imagery that makes use of a convolutional neural network (CNN).
   We trained ResNet, and DenseNet CNNs using down-scaled aerial imagery and
   tested each model on 31 cm-resolution imagery obtained from the WorldView-3
   sensor. Satellite imagery was tiled and the trained algorithms were used to
   classify whether or not a tile was likely to contain a whale. Our best
   model correctly classified 100% of tiles with whales, and 94% of tiles
   containing only water. All model architectures performed well, with
   learning rate controlling performance more than architecture. While the
   resolution of commercially-available satellite imagery continues to make
   whale identification a challenging problem, our approach provides the means
   to efficiently eliminate areas without whales and, in doing so, greatly
   accelerates ocean surveys for large cetaceans.

Alex Borowicz
PhD Candidate
Ecology & Evolution
Stony Brook University


"That's true enough," said Candide, "but we must go and work in the garden."
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.uvic.ca/pipermail/marmam/attachments/20191009/652bba60/attachment.html>

More information about the MARMAM mailing list