pone.0263377.g003.tif (1.51 MB)
Fig 3 -
figure
posted on 2022-02-02, 18:33 authored by Stephanie M. Bilodeau, Austin W. H. Schwartz, Binfeng Xu, V. Paúl Pauca, Miles R. SilmanEmpty (left) and fish-filled (right) images captured at the same camera site within 24 hours. The empty or “NoFish” photo was assigned only a 0.09 probability of containing a fish by our ResNet-50 model, whereas the model predicted a 0.88 probability of the second photo containing at least one fish, hence its “Fish” designation. Probabilities like these were used to sort images into “Fish” and “NoFish” categories to streamline further analysis. Reprinted from [39] under a CC BY license with permission.
History
Usage metrics
Categories
Keywords
use machine learningterrestrial camera trappingshallow tropical reefscapesremote marine environmentsmarine systems duemarine animal behaviorfunction continuously underwaterfirst successful methodfeeding behavior relateddocumented benthic patternameras ), deployablespatially distributed cameraslater versions deployedcameras ’ potentialcameras proved ableaddressing fish movementdiv >< pcase study demonstratese c diver surveys revealedlarge spatial scalescase studytemporal scalesdeployed depthsbased surveysaddressing questionssuccessfully appliedsorting imagesrepeatable countsoverall accuracylarge volumelapse modefive monthsdevelop methodologydaylight hoursaquatic organismsaquatic environment15 minutes000 images
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC