Data from: Drones and convolutional neural networks facilitate automated and accurate cetacean species identification and photogrammetry
The flourishing application of drones within marine science provides more opportunity to conduct photogrammetric studies on large and varied populations of many different species. While these new platforms are increasing the size and availability of imagery datasets, established photogrammetry metho...
Main Authors: | , , , , , |
---|---|
Format: | Article in Journal/Newspaper |
Language: | unknown |
Published: |
2019
|
Subjects: | |
Online Access: | http://hdl.handle.net/10255/dryad.220866 https://doi.org/10.5061/dryad.7482v2n |
id |
ftdryad:oai:v1.datadryad.org:10255/dryad.220866 |
---|---|
record_format |
openpolar |
spelling |
ftdryad:oai:v1.datadryad.org:10255/dryad.220866 2023-05-15T14:04:38+02:00 Data from: Drones and convolutional neural networks facilitate automated and accurate cetacean species identification and photogrammetry Gray, Patrick C. Bierlich, Kevin C. Mantell, Sydney A. Friedlaender, Ari S. Goldbogen, Jeremy A. Johnston, David W. United States California Durham Western Antarctic Peninsula Antarctica 2019-07-24T14:55:23Z http://hdl.handle.net/10255/dryad.220866 https://doi.org/10.5061/dryad.7482v2n unknown doi:10.5061/dryad.7482v2n/1 doi:10.5061/dryad.7482v2n/2 doi:10.5061/dryad.7482v2n/3 doi:10.5061/dryad.7482v2n/4 doi:10.1111/2041-210x.13246 doi:10.5061/dryad.7482v2n Gray PC, Bierlich KC, Mantell SA, Friedlaender AS, Goldbogen JA, Johnston DW (2019) Drones and convolutional neural networks facilitate automated and accurate cetacean species identification and photogrammetry. Methods in Ecology and Evolution. http://hdl.handle.net/10255/dryad.220866 photogrammetry cetaceans convolutional neural network unoccupied aerial systems drones species identification deep learning population assessments Article 2019 ftdryad https://doi.org/10.5061/dryad.7482v2n https://doi.org/10.5061/dryad.7482v2n/1 https://doi.org/10.5061/dryad.7482v2n/2 https://doi.org/10.5061/dryad.7482v2n/3 https://doi.org/10.5061/dryad.7482v2n/4 https://doi.org/10.1111/2041-210x.13246 2020-01-01T16:31:00Z The flourishing application of drones within marine science provides more opportunity to conduct photogrammetric studies on large and varied populations of many different species. While these new platforms are increasing the size and availability of imagery datasets, established photogrammetry methods require considerable manual input, allowing individual bias in techniques to influence measurements, increasing error and magnifying the time required to apply these techniques. Here, we introduce the next generation of photogrammetry methods utilizing a convolutional neural network to demonstrate the potential of a deep learning‐based photogrammetry system for automatic species identification and measurement. We then present the same data analysed using conventional techniques to validate our automatic methods. Our results compare favorably across both techniques, correctly predicting whale species with 98% accuracy (57/58) for humpback whales, minke whales, and blue whales. Ninety percent of automated length measurements were within 5% of manual measurements, providing sufficient resolution to inform morphometric studies and establish size classes of whales automatically. The results of this study indicate that deep learning techniques applied to survey programs that collect large archives of imagery may help researchers and managers move quickly past analytical bottlenecks and provide more time for abundance estimation, distributional research, and ecological assessments. Article in Journal/Newspaper Antarc* Antarctic Antarctic Peninsula Antarctica Dryad Digital Repository (Duke University) Antarctic Antarctic Peninsula |
institution |
Open Polar |
collection |
Dryad Digital Repository (Duke University) |
op_collection_id |
ftdryad |
language |
unknown |
topic |
photogrammetry cetaceans convolutional neural network unoccupied aerial systems drones species identification deep learning population assessments |
spellingShingle |
photogrammetry cetaceans convolutional neural network unoccupied aerial systems drones species identification deep learning population assessments Gray, Patrick C. Bierlich, Kevin C. Mantell, Sydney A. Friedlaender, Ari S. Goldbogen, Jeremy A. Johnston, David W. Data from: Drones and convolutional neural networks facilitate automated and accurate cetacean species identification and photogrammetry |
topic_facet |
photogrammetry cetaceans convolutional neural network unoccupied aerial systems drones species identification deep learning population assessments |
description |
The flourishing application of drones within marine science provides more opportunity to conduct photogrammetric studies on large and varied populations of many different species. While these new platforms are increasing the size and availability of imagery datasets, established photogrammetry methods require considerable manual input, allowing individual bias in techniques to influence measurements, increasing error and magnifying the time required to apply these techniques. Here, we introduce the next generation of photogrammetry methods utilizing a convolutional neural network to demonstrate the potential of a deep learning‐based photogrammetry system for automatic species identification and measurement. We then present the same data analysed using conventional techniques to validate our automatic methods. Our results compare favorably across both techniques, correctly predicting whale species with 98% accuracy (57/58) for humpback whales, minke whales, and blue whales. Ninety percent of automated length measurements were within 5% of manual measurements, providing sufficient resolution to inform morphometric studies and establish size classes of whales automatically. The results of this study indicate that deep learning techniques applied to survey programs that collect large archives of imagery may help researchers and managers move quickly past analytical bottlenecks and provide more time for abundance estimation, distributional research, and ecological assessments. |
format |
Article in Journal/Newspaper |
author |
Gray, Patrick C. Bierlich, Kevin C. Mantell, Sydney A. Friedlaender, Ari S. Goldbogen, Jeremy A. Johnston, David W. |
author_facet |
Gray, Patrick C. Bierlich, Kevin C. Mantell, Sydney A. Friedlaender, Ari S. Goldbogen, Jeremy A. Johnston, David W. |
author_sort |
Gray, Patrick C. |
title |
Data from: Drones and convolutional neural networks facilitate automated and accurate cetacean species identification and photogrammetry |
title_short |
Data from: Drones and convolutional neural networks facilitate automated and accurate cetacean species identification and photogrammetry |
title_full |
Data from: Drones and convolutional neural networks facilitate automated and accurate cetacean species identification and photogrammetry |
title_fullStr |
Data from: Drones and convolutional neural networks facilitate automated and accurate cetacean species identification and photogrammetry |
title_full_unstemmed |
Data from: Drones and convolutional neural networks facilitate automated and accurate cetacean species identification and photogrammetry |
title_sort |
data from: drones and convolutional neural networks facilitate automated and accurate cetacean species identification and photogrammetry |
publishDate |
2019 |
url |
http://hdl.handle.net/10255/dryad.220866 https://doi.org/10.5061/dryad.7482v2n |
op_coverage |
United States California Durham Western Antarctic Peninsula Antarctica |
geographic |
Antarctic Antarctic Peninsula |
geographic_facet |
Antarctic Antarctic Peninsula |
genre |
Antarc* Antarctic Antarctic Peninsula Antarctica |
genre_facet |
Antarc* Antarctic Antarctic Peninsula Antarctica |
op_relation |
doi:10.5061/dryad.7482v2n/1 doi:10.5061/dryad.7482v2n/2 doi:10.5061/dryad.7482v2n/3 doi:10.5061/dryad.7482v2n/4 doi:10.1111/2041-210x.13246 doi:10.5061/dryad.7482v2n Gray PC, Bierlich KC, Mantell SA, Friedlaender AS, Goldbogen JA, Johnston DW (2019) Drones and convolutional neural networks facilitate automated and accurate cetacean species identification and photogrammetry. Methods in Ecology and Evolution. http://hdl.handle.net/10255/dryad.220866 |
op_doi |
https://doi.org/10.5061/dryad.7482v2n https://doi.org/10.5061/dryad.7482v2n/1 https://doi.org/10.5061/dryad.7482v2n/2 https://doi.org/10.5061/dryad.7482v2n/3 https://doi.org/10.5061/dryad.7482v2n/4 https://doi.org/10.1111/2041-210x.13246 |
_version_ |
1766275843095527424 |