The PNNL-TUNAMELT Dataset for Automated Detection around Marine Energy Devices
This labeled dataset released in the publication "PNNL-TUNAMELT: Towards automating the detection of interactions with marine energy devices" consists of 107451 labeled, acoustic camera video frames capturing marine life interactions around an underwater tidal turbine. The video frames are labeled with bounding boxes around each object in frame that a fish biologist has identified as being marine life. This dataset was created to promote automated target detection research and capabilities around underwater turbines in order to enable their safe deployment and operation. Disclaimer: no collisions were observed with the marine energy device while labeling and analyzing this dataset.
The data for this dataset was originally collected in 2010 around Ocean Renewable Power Company's (ORPC's) tidal turbine deployment in Cobscook Bay, Maine, USA, the results of which were published in [Viehman, H. A., and G. B. Zydlewski. 2015. Fish interactions with a commercial-scale tidal energy device in the natural environment. Estuaries Coasts 38: 241-252.]
The code, software tools, and a baseline automated target detection approach which were developed for this effort are located at https://github.com/tsnowak/pnnl-tunamelt (linked). To get started with using this dataset, please refer to this repository.
For further information, please refer to the linked github repository and associated publication, or feel free to reach out to the authors.
Citation Formats
TY - DATA
AB - This labeled dataset released in the publication "PNNL-TUNAMELT: Towards automating the detection of interactions with marine energy devices" consists of 107451 labeled, acoustic camera video frames capturing marine life interactions around an underwater tidal turbine. The video frames are labeled with bounding boxes around each object in frame that a fish biologist has identified as being marine life. This dataset was created to promote automated target detection research and capabilities around underwater turbines in order to enable their safe deployment and operation. Disclaimer: no collisions were observed with the marine energy device while labeling and analyzing this dataset.
The data for this dataset was originally collected in 2010 around Ocean Renewable Power Company's (ORPC's) tidal turbine deployment in Cobscook Bay, Maine, USA, the results of which were published in [Viehman, H. A., and G. B. Zydlewski. 2015. Fish interactions with a commercial-scale tidal energy device in the natural environment. Estuaries Coasts 38: 241-252.]
The code, software tools, and a baseline automated target detection approach which were developed for this effort are located at https://github.com/tsnowak/pnnl-tunamelt (linked). To get started with using this dataset, please refer to this repository.
For further information, please refer to the linked github repository and associated publication, or feel free to reach out to the authors.
AU - Nowak, Theodore
A2 - Staines, Garrett
A3 - Abdullai, Blerim
DB - Marine and Hydrokinetic Data Repository
DP - Open EI | National Renewable Energy Laboratory
DO -
KW - MHK
KW - Computer Vision
KW - Dataset
KW - Acoustic Camera
KW - Imaging Sonar
KW - Environmental Monitoring
KW - MHK Monitoring
KW - Collision Risk
KW - PNNL-TUNAMELT
KW - automated detection
KW - marine energy devices
KW - acoustic camera video frames
KW - tidal turbine
KW - code
KW - GitHub
KW - software tools
LA - English
DA - 2025/06/01
PY - 2025
PB - Pacific Northwest National Laboratory
T1 - The PNNL-TUNAMELT Dataset for Automated Detection around Marine Energy Devices
UR - https://mhkdr.openei.org/submissions/633
ER -
Nowak, Theodore, et al. The PNNL-TUNAMELT Dataset for Automated Detection around Marine Energy Devices. Pacific Northwest National Laboratory, 1 June, 2025, Marine and Hydrokinetic Data Repository. https://mhkdr.openei.org/submissions/633.
Nowak, T., Staines, G., & Abdullai, B. (2025). The PNNL-TUNAMELT Dataset for Automated Detection around Marine Energy Devices. [Data set]. Marine and Hydrokinetic Data Repository. Pacific Northwest National Laboratory. https://mhkdr.openei.org/submissions/633
Nowak, Theodore, Garrett Staines, and Blerim Abdullai. The PNNL-TUNAMELT Dataset for Automated Detection around Marine Energy Devices. Pacific Northwest National Laboratory, June, 1, 2025. Distributed by Marine and Hydrokinetic Data Repository. https://mhkdr.openei.org/submissions/633
@misc{MHKDR_Dataset_633,
title = {The PNNL-TUNAMELT Dataset for Automated Detection around Marine Energy Devices},
author = {Nowak, Theodore and Staines, Garrett and Abdullai, Blerim},
abstractNote = {This labeled dataset released in the publication "PNNL-TUNAMELT: Towards automating the detection of interactions with marine energy devices" consists of 107451 labeled, acoustic camera video frames capturing marine life interactions around an underwater tidal turbine. The video frames are labeled with bounding boxes around each object in frame that a fish biologist has identified as being marine life. This dataset was created to promote automated target detection research and capabilities around underwater turbines in order to enable their safe deployment and operation. Disclaimer: no collisions were observed with the marine energy device while labeling and analyzing this dataset.
The data for this dataset was originally collected in 2010 around Ocean Renewable Power Company's (ORPC's) tidal turbine deployment in Cobscook Bay, Maine, USA, the results of which were published in [Viehman, H. A., and G. B. Zydlewski. 2015. Fish interactions with a commercial-scale tidal energy device in the natural environment. Estuaries Coasts 38: 241-252.]
The code, software tools, and a baseline automated target detection approach which were developed for this effort are located at https://github.com/tsnowak/pnnl-tunamelt (linked). To get started with using this dataset, please refer to this repository.
For further information, please refer to the linked github repository and associated publication, or feel free to reach out to the authors.},
url = {https://mhkdr.openei.org/submissions/633},
year = {2025},
howpublished = {Marine and Hydrokinetic Data Repository, Pacific Northwest National Laboratory, https://mhkdr.openei.org/submissions/633},
note = {Accessed: 2025-07-05}
}
Details
Data from Jun 1, 2025
Last updated Jun 30, 2025
Submitted Jun 19, 2025
Organization
Pacific Northwest National Laboratory
Contact
Theodore Nowak
509.375.2531
Authors
Keywords
MHK, Computer Vision, Dataset, Acoustic Camera, Imaging Sonar, Environmental Monitoring, MHK Monitoring, Collision Risk, PNNL-TUNAMELT, automated detection, marine energy devices, acoustic camera video frames, tidal turbine, code, GitHub, software toolsDOE Project Details
Project Name Triton Initiative
Project Lead Samantha Eaves
Project Number FY25 AOP 232611
