Papers & Presentations

PowerPoint: Machine Learning for Big Fishery Visual Data

PowerPoint: Utilizing Wireless Technologies to Monitor New England Groundfish

Once a profitable fishery supporting coastal communities, the groundfish fishery in New England is in crisis. One factor is the mixed stock nature of the fishery where stocks with low catch limits constrain the harvest of more abundant stocks. In addition, climate-mediated changes in this region are unprecedented, and the impacts on marine fisheries resources, such as changes in distribution and productivity, are increasing. There is a need for more comprehensive and accurate data collection to understand the current and future state of the Gulf of Maine fishery resources. Adaptive electronic tools can help address these data gaps. The Gulf of Maine Research Institute (GMRI) has begun developing an extension of traditional electronic monitoring (EM), called the Fisheries… Read More »

PowerPoint: Integrated Operational Approach to EM & ER Onboard Fishing Vessels-Now and Tomorrow

Fisheries management is plagued by lack of verifiable data for compliance and stock assessments. Low vessel observer coverage, and delays in data transfer, transcription, integration, and analysis are the norm. These problems hinder timely analysis and updates of fishing regulations relevant to existing stocks. Fishing capacity is increasing around the world; incidental catches of protected species, bycatch discard, and IUU are critical issues. Technological advances and cost reduction bring much opportunity applicable to the fishing industry. The authors developed an integrated EMR solution and several technologies will be discussed.

Paper logbooks are limited to alphanumerical content and suffer from accuracy concerns. Electronic reporting (ER) replaces paper logbooks and allows for enhanced data entry and timely delivery. Electronic monitoring (EM) entails… Read More »

PowerPoint: Open Source Software Platform for Electronic Monitoring

Sea State Inc. has developed a suite of software applications to support commercial fisheries electronic monitoring (EM) data acquisition, review, summarization and archival. This National Fish and Wildlife Foundation funded project, in partnership with Saltwater Inc. and Chordata LLC, provides a platform that is adaptable to marine fisheries nationwide. The source code is available under the Gnu Public License, ensuring that all future enhancements will be open source as well.

The shipboard component of the software works on a wide range of hardware: from simple data-loggers without cameras up to systems with eight or more cameras. It is designed for flexibility – system behavior is controlled by configuration rules based on data from GPS, hydraulic pressure, reel rotation and video… Read More »

PowerPoint: Query Learning for Fish Identification

In fishery community, image-classification-based supervised approaches have been well studied for species identification. However, the classifiers learned from one dataset usually cannot directly be applied for a new dataset since the data captured from different years or regions often have large difference, like different colors, camera distortions and distributions of species. It is unreasonable and huge expensive to label the new data to train new classifiers. Hence, we propose a novel method with the combination of query learning and semi-supervised learning to address this challenge with only a small amount of data need to be labeled. First, an uncertainty measure is designed based on the distance between labeled and unlabeled samples in the transformed feature representations through support… Read More »

PowerPoint: Intelligent Monitoring Systems for Fishery Applications

The Observer program at Alaska Fisheries Science Center have been conducting research and development of new innovative electronic monitoring (EM) technologies to help address challenges for collecting scientific data to support catch estimation. Work focuses on development of new camera-based systems, methods, and tools while leveraging the latest development in computer vision to improve system functionality and offering potential solutions.

Chute and rail based camera systems are under developed to support and improve EM data collection from both trawl and longline fisheries. The camera chute system is technologically advanced and currently assesses image quality, catch count, length measurement and species identification. This functionality is also being actively developed for the rail camera systems. Further development of automated event –… Read More »

PowerPoint: Fishface – Exploring the use of image recognition software in fisheries management

Commercial fisheries increasingly are incorporating electronic monitoring (EM) into existing fishery-dependent data collection programs. Across the United States, seven fisheries in the Northeast (Atlantic herring and mackerel, and groundfish), West Coast (Groundfish, Pacific whiting), Alaska (fixed gear groundfish and Pacific halibut), and Highly Migratory Species (Atlantic Bluefin tuna) are using EM for catch accounting and/or compliance with catch retention requirements. Machine vision learning applications, based on image-training datasets, global positioning systems (GPS), and sensors, could substantially reduce data collection and processing costs for existing and future EM programs. Remotely collected data (video images, GPS, sensors) could be used for collecting data such as gear, time and area of effort, length and weight measurements, and species composition of catch and catch… Read More »

PowerPoint: Accelerating the Development of Automated Fish Identification for EM Systems- An Example from New England Groundfish

Automated fish measurement and speciation has the potential to revolutionize the fishery monitoring process, driving down costs and reducing the burden on human observers and video reviewers. Many efforts are underway to apply machine learning algorithms to different problems related to fisheries science. These efforts could benefit greatly from the large open data science community, exemplified by sites like Kaggle and DrivenData. These sites provide a platform to host an open data set and offer a prize for contestants to solve challenging machine learning problems. Using a grant from the National Fish & Wildlife Foundation, our goal was to create a high-quality data set to push forward the state of the art in automatic processing of Electronic Monitoring… Read More »

PowerPoint: Tracking and Measuring of Catch Events in Stereo Video for Longline Fisheries

Automated video analyses in monitoring fishery activities have drawn increasing attention due to its scalability and capability. Stereo videos, compared to monocular videos, can capture the depth of information in addition to color and texture; thus, it can be more robust in monitoring and capable of measuring the length of fish. In this work, we present a reliable tracking and measurement approach to stereo videos for catch events in longline fisheries. First, we combine background subtraction method with image object detector to detect fish in a stereo frame. Using the location and disparity information in the stereo frame, we can thus back-project the fish and track it in the 3D space. With the inferred 3D information, we separate… Read More »

Implementing EM for the Alaska Pot Cod Fleet

The North Pacific Fisheries Association (NPFA), a fishermen’s organization, and Saltwater Inc., an observer and EM service provider, received funding in 2016 from the National Fish and Wildlife Foundation (NFWF) to implement electronic monitoring (EM) in Alaska’s pot cod fishery.

The project highlights skipper engagement, integration of observers into the EM program, cross training of skilled EM personnel, a streamlined feedback loop between vessels and data, and local data review. A progress report, Implementing EM For Alaska’s Pot Cod Fleet, details key aspects of the program and is available at the link below.

Implementing EM for Alaska’s Pot Cod Fleet

For more information about this implementation, contact Abigail Turner – –  or Nancy Munro – .