gov.noaa.nmfs.inport:79020 eng UTF8 gov.noaa.nmfs.inport:69412 dataset Imagery Alaska Fisheries Science Center 7600 Sand Point Way N.E., Building 4 Seattle WA 98115 USA https://www.afsc.noaa.gov WWW:LINK-1.0-http--link Website Website for this organization information 0700-1700 Pacific Time resourceProvider 2026-03-31T22:02:58 ISO 19115-2 Geographic Information - Metadata Part 2 Extensions for imagery and gridded data ISO 19115-2:2009(E) Salmon tracking 2023-03-08 creation 2026-01-20 revision publication NOAA/NMFS/EDM 79020 https://www.fisheries.noaa.gov/inport/item/79020 WWW:LINK-1.0-http--link Full Metadata Record View the complete metadata record on InPort for more information about this dataset. information Pacific States Marine Fisheries Commission Lurbur, Moses <mlurbur@gmail.com> author Yochum, Noelle (206) 526-4432 7600 Sand Point Way NE, Bldg. 4 Seattle WA 98115 United States noelle.yochum@noaa.gov author Wilson, Katherine 2065264474 7600 Sand Point Way NE, Bldg. 4 Seattle FL 98115 United States katherine.wilson@noaa.gov author imageDigital This dataset contains the annotated data and object detection models used to evaluate multi-object tracking (MOT) algorithms for tracking Pacific salmon (Oncorhynchus spp.) in videos collected in a commercial walleye pollock (Gadus chalcogrammus) trawl in Alaska. The annotated data contains track annotations for all Pacific salmon and walleye pollock present and frame level annotations that describe the trawl and background conditions. The annotated data is composed of 11,572 salmon and 73,394 pollock annotations from 16,989 video frames that came from 184 video clips. Five sizes of YOLO12 detection models (n, s, m, l, and x) were fine-tuned and evaluated for pollock and salmon detection and four tracking algorithms (BoT-SORT, ByteTrack, Intersection over Union, and Centroid) were used to assess MOT of bycaught Pacific salmon in trawls. All five YOLO12 models are provided as part of this dataset. Bycatch reduction devices (BRDs) are used in the Alaska walleye pollock (Gadus chalcogrammus) fishery to reduce Pacific salmon (Oncorhynchus spp.) bycatch. Evaluation of BRD effectiveness often requires people to process collected or live-feed video, which can be a tedious, expensive, and time-consuming task. Deep learning can be used to automate the detection, classification, and tracking of fish in videos to support BRD and other fisheries bycatch work. This dataset was developed to evaluate the performance of widely-available, open-source pre-trained object detection models for the task of pollock and salmon detection and multi-object tracking models for salmon tracking. F/V Pacific Explorer, David Bryan, Deborah Sharpe , Katherine Hellen-Schneider, Matthew Callahan, Connor Fox Wilson, Katherine 2065264474 7600 Sand Point Way NE, Bldg. 4 Seattle FL 98115 United States katherine.wilson@noaa.gov pointOfContact asNeeded bycatch reduction computer vision multi-object tracking object detection trawl fisheries theme DOC/NOAA/NMFS/AFSC > Alaska Fisheries Science Center, National Marine Fisheries Service, NOAA, U.S. Department of Commerce dataCentre Global Change Master Directory (GCMD) Data Center Keywords 2017-04-24 publication 8.5 Earth Science Data and Information System, Earth Science Projects Division, Goddard Space Flight Center (GSFC), National Aeronautics and Space Administration (NASA) https://forum.earthdata.nasa.gov/app.php/tag/GCMD%2BKeywords GCMD Keyword Forum Page Global Change Master Directory (GCMD). 2025. GCMD Keywords, Version 22. Greenbelt, MD: Earth Science Data and Information System, Earth Science Projects Division, Goddard Space Flight Center (GSFC), National Aeronautics and Space Administration (NASA). URL (GCMD Keyword Forum Page): https://forum.earthdata.nasa.gov/app.php/tag/GCMD+Keywords information custodian AFSC/RACE/MACE: 2019-2020 Salmon and Pollock Object Detection project InPort otherRestrictions Cite As: Alaska Fisheries Science Center, [Date of Access]: Salmon tracking [Data Date Range], https://www.fisheries.noaa.gov/inport/item/79020. NOAA provides no warranty, nor accepts any liability occurring from any incomplete, incorrect, or misleading data, or from any incorrect, incomplete, or misleading use of the data. It is the responsibility of the user to determine whether or not the data is suitable for the intended purpose. license These data were produced by NOAA and are not subject to copyright protection in the United States. NOAA waives any potential copyright and related rights in these data worldwide through the Creative Commons Zero 1.0 Universal Public Domain Dedication (CC0-1.0). SPDX License: Creative Commons Zero v1.0 Universal (CC0-1.0) otherRestrictions Access Constraints: If using these data, please cite https://doi.org/10.1093/icesjms/fsaf168 and/or https://doi.org/[placeholder] unclassified NOAA Data Management Plan (DMP) NOAA/NMFS/EDM 79020 https://www.fisheries.noaa.gov/inportserve/waf/noaa/nmfs/afsc/dmp/pdf/79020.pdf WWW:LINK-1.0-http--link NOAA Data Management Plan (DMP) NOAA Data Management Plan for this record on InPort. information crossReference eng; US oceans All 7 tows from Trip 1 were within this extent. -166.03722222222 -165.59805555556 54.32472222222 54.69833333333 2020-08-07 2020-08-10 All 4 tows from trip 2 were within these bounds. -165.72888888889 -165.43055555556 54.51527777778 54.78472222222 2020-08-13 2020-08-16 All 26 tows were within this extent. -166 -164.5 54.33333333333 55.33333333333 2019-06-01 2019-06-30 PNG - Portable Network Graphics Format Uncompressed https://console.cloud.google.com/storage/browser/nmfs_odp_afsc/RACE/MACE/salmon_pollock_object_detection/ WWW:LINK-1.0-http--link 2019-2020 Salmon and Pollock Object Detection dataset This folder within the Alaska Fisheries Science Center's Google cloud storage contains the data and object detection models used for the Lurbur et al. 2026 study on multi-object tracking of salmon in trawl videos from the Alaska commercial pollock fishery. The 'annotated_data' folder contains a folder for each of the 184 video clips used in the study that provides video frames as png files and the annotations as a custom COCO json file. The 'models' folder contains the object detection models that were used in the study. The 'documentation' folder contains text files that provide details about the data and the cloud storage file structure. The 'tow_data' folder contains videos and data relevant to the Wilson et al. 2025 study and is not relevant to the Lurbur et al. 2026 study. download dataset Representativeness Video from within the trawl of the commercial pollock fishery in Alaska and the associated annotations of salmon and pollock. Accuracy See Wilson et al. 2025 and Lurbur et al. 2026 publications for details. Analytical Accuracy See Wilson et al. 2025 and Lurbur et al. 2026 publications for details. Horizontal Positional Accuracy See Wilson et al. 2025 and Lurbur et al. 2026 publications for details. Vertical Positional Accuracy See Wilson et al. 2025 and Lurbur et al. 2026 publications for details. Quantitation Limits See Wilson et al. 2025 and Lurbur et al. 2026 publications for details. Bias See Wilson et. 2025 publications for details. Comparability See Wilson et al. 2025 and Lurbur et al. 2026 publications for details. Completeness Measure See Wilson et al. 2025 and Lurbur et al. 2026 publications for details. Precision See Wilson et al. 2025 and Lurbur et al. 2026 publications for details. Analytical Precision See Wilson et al. 2025 and Lurbur et al. 2026 publications for details. Field Precision See Wilson et al. 2025 and Lurbur et al. 2026 publications for details. Sensitivity See Wilson et al. 2025 and Lurbur et al. 2026 publications for details. Detection Limit See Wilson et al. 2025 and Lurbur et al. 2026 publications for details. Completeness Report See Wilson et al. 2025 and Lurbur et al. 2026 publications for details. Conceptual Consistency See Wilson et al. 2025 and Lurbur et al. 2026 publications for details. The video portion of this project was collected onboard the FV Pacific Explorer by Noelle Yochum and Katherine Hellen-Schneider in 2019 and by Connor Fox in 2020 with Sexton housings equipped with Mobius cameras. Processing of video (identifying salmon and entering data) was conducted by David Bryan, Michelle Dyroy, Noelle Yochum, and Katherine Wilson. Video annotation (bounding box tracks of salmon and pollock) was done by Moses Lurbur, Deborah Sharpe, Katherine Hellen-Schneider, and Matthew Callahan. Annotation QA/QC was conducted by Katherine Wilson. Sexton Video Camera System Video Recorder The Sexton video camera system consists of an underwater housing rated to 1000m, 2 LED lights, a Mobius2 Action Camera, and a power supply (either 2 NiMH 19.2v or 2 Li-Ion 14.4v batteries). The Mobius2 camera can record 1800 HD video at 30 frames per second and is suitable for low light scenarios.