Relevance: primary · Type: action
Confidence100%
MIT Sea Grant collaborated with the Woodwell Climate Research Center and other collaborators to demonstrate a deep learning-based system for fish monitoring.
Relevance: supporting · Type: background
Confidence100%
Each spring, river herring migrate from Massachusetts coastal waters to freshwater rivers and streams to spawn.
Relevance: supporting · Type: background
Confidence100%
River herring populations have declined over the past several decades.
Relevance: supporting · Type: background
Confidence100%
River herring migration is monitored across the region primarily through traditional visual counting and volunteer-based programs.
Relevance: supporting · Type: background
Confidence100%
The annual river herring run begins in March.
Relevance: primary · Type: action
Confidence100%
Researchers from the Woodwell Climate Research Center, MIT Sea Grant, MIT CSAIL, MIT Lincoln Laboratory, and Intuit explored a monitoring method using underwater video and computer vision to supplement citizen science efforts.
Relevance: supporting · Type: background
Confidence100%
Zhongqi Chen is a researcher at the Woodwell Climate Research Center.
Relevance: supporting · Type: background
Confidence100%
Linda Deegan is a researcher at the Woodwell Climate Research Center.
Relevance: supporting · Type: background
Confidence100%
Robert Vincent is a researcher at MIT Sea Grant.
Relevance: supporting · Type: background
Confidence100%
Kevin Bennett is a researcher at MIT Sea Grant.
Relevance: supporting · Type: background
Confidence100%
Sara Beery is a researcher at MIT CSAIL.
Relevance: supporting · Type: background
Confidence100%
Timm Haucke is a researcher at MIT CSAIL.
Relevance: supporting · Type: background
Confidence100%
Austin Powell is a researcher at Intuit.
Relevance: supporting · Type: background
Confidence100%
Lydia Zuehsow is a researcher at MIT Lincoln Laboratory.
Relevance: primary · Type: event
Confidence100%
A paper describing this work was published in the journal Remote Sensing in Ecology and Conservation in February 2026.
Relevance: supporting · Type: background
Confidence100%
The paper is open-access.
Relevance: primary · Type: background
Confidence100%
The paper, titled “From snapshots to continuous estimates: Augmenting citizen science with computer vision for fish monitoring,” outlines how advancements in computer vision and deep learning, including object detection, tracking, and species classification, can automate fish counting with improved efficiency and data quality.
Relevance: supporting · Type: background
Confidence100%
Traditional fish monitoring methods are constrained by time, environmental conditions, and labor intensity.
Relevance: supporting · Type: background
Confidence100%
Volunteer visual counts are limited to brief daytime sampling windows.
Relevance: supporting · Type: background
Confidence100%
Volunteer visual counts miss nighttime fish movement and short migration pulses when hundreds of fish can pass by within a few minutes.
Relevance: supporting · Type: background
Confidence100%
Passive acoustic monitoring and imaging sonar technologies have advanced continuous fish monitoring under certain conditions.
Relevance: supporting · Type: background
Confidence100%
Manual review of underwater video is a low-cost option for continuous fish monitoring.
Relevance: supporting · Type: background
Confidence100%
Manual review of underwater video is labor-intensive and time-consuming.
Relevance: primary · Type: action
Confidence100%
The study presents a deep learning-based system for automated fish monitoring.
Relevance: primary · Type: action
Confidence100%
The team built a pipeline from in-field underwater cameras to video labeling and model training to achieve automated computer vision-powered fish counting.
Relevance: primary · Type: action
Confidence100%
Videos were collected from the Coonamessett River in Falmouth, the Ipswich River in Ipswich, and the Santuit River in Mashpee.
Relevance: primary · Type: action
Confidence100%
For training dataset preparation, the team selected video clips with variations in lighting, water clarity, fish species and density, time of day, and season.
Relevance: primary · Type: action
Confidence100%
The team used an open-source web platform to manually label video frames with bounding boxes to track fish movement.
Relevance: primary · Type: action
Confidence100%
The team labeled 1,435 video clips and annotated 59,850 frames.
Relevance: primary · Type: action
Confidence100%
The researchers compared computer vision counts with human video reviews, stream-side visual counts, and passive integrated transponder tagging data.
Relevance: primary · Type: event
Confidence100%
Models trained on diverse multi-site and multi-year data produced season-long, high-resolution counts consistent with traditional estimates.
Relevance: primary · Type: event
Confidence100%
The system provided insights into migration behavior, timing, and movement patterns linked to environmental factors.
Relevance: primary · Type: event
Confidence100%
Using video from the 2024 Coonamesset River migration, the system counted 42,510 river herring.
Relevance: primary · Type: event
Confidence100%
The system revealed that upstream migration peaked at dawn.
Relevance: primary · Type: event
Confidence100%
The system revealed that downstream migration was largely nocturnal.
Relevance: primary · Type: event
Confidence100%
The system revealed that fish utilized darker, quieter periods to avoid predators.
Relevance: supporting · Type: action
Confidence100%
The researchers intend to advance computer vision in fisheries management and provide a framework and best practices for integrating the technology into conservation efforts for a wide range of aquatic species.
Robert Vincent, researcher at MIT Sea Grant
Relevance: primary · Type: quote
Confidence100%
"MIT Sea Grant has been funding work on this topic for some time now, and this excellent work by Zhongqi Chen and colleagues will advance fisheries monitoring capabilities and improve fish population assessments for fisheries managers and conservation groups. It will also provide education and training for students, the public, and citizen science groups in support of the ecologically and culturally important river herring populations along our coasts."
Relevance: supporting · Type: background
Confidence100%
Traditional monitoring will continue until fisheries management agencies fully implement automated counting systems to maintain consistency in long-term datasets.
Relevance: supporting · Type: background
Confidence100%
The researchers view computer vision and citizen science as complementary approaches.
Relevance: supporting · Type: background
Confidence100%
Volunteers will be necessary for camera maintenance and for contributing directly to the computer vision workflow, including video annotation and model verification.
Relevance: supporting · Type: background
Confidence100%
The researchers envision that integrating citizen observations and computer vision-generated data will create a more comprehensive approach to environmental monitoring.
Relevance: supporting · Type: background
Confidence100%
This work was funded by MIT Sea Grant.
Relevance: supporting · Type: background
Confidence100%
Additional support was provided by the Northeast Climate Adaptation Science Center.
Relevance: supporting · Type: background
Confidence100%
Additional support was provided by an MIT Abdul Latif Jameel Water and Food Systems seed grant.
Relevance: supporting · Type: background
Confidence100%
Additional support was provided by the AI and Biodiversity Change Global Center, supported by the National Science Foundation and the Natural Sciences and Engineering Research Council of Canada.
Relevance: supporting · Type: background
Confidence100%
Additional support was provided by the MIT Undergraduate Research Opportunities Program.
forum Comments (0)
No comments yet. Be the first to comment.