Estimating reef fish size distributions with a mini remotely operated vehicle-integrated stereo camera system

Garner, Steven B.; Olsen, Aaron M.; Caillouet, Ryan; Campbell, Matthew D.; Patterson, William F.

doi:10.1371/journal.pone.0247985

Cited by 10 publications

(5 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Mobile underwater video cameras, either diver operated or on underwater drones, remotely operated vehicles (ROVs) or autonomous underwater vehicles, are beginning to replace traditional underwater visual census (UVC) techniques for habitats such as coral reefs, rocky reefs, deep waters, and offshore gas platforms (Andaloro et al, 2013;Goetze et al, 2015;Sward et al, 2019). Cameras have clear advantages over UVC: 1) cameras produce permanent records during a survey; 2) remote cameras on drones or ROVs can be used in deep waters inaccessible to divers, and in waters with dangerous animals; 3) remote cameras reduce costs associated with diving (Andaloro et al, 2013;Sward et al, 2019;Garner et al, 2021). Cameras can also avoid the known biases of UVC for different types of fish (Bernard et al, 2013;Sheaves et al, 2020).…”

Section: Introductionmentioning

confidence: 99%

Fish surveys on the move: Adapting automated fish detection and classification frameworks for videos on a remotely operated vehicle in shallow marine waters

et al. 2022

View full text Add to dashboard Cite

Mobile underwater cameras, diver-operated or on underwater vehicles, have become popular for monitoring fisheries. Replacing divers with cameras has clear advantages, such as creating permanent records and accessing waters unavailable to divers. The use of cameras, however, typically produces large quantities of video that are time-consuming to process manually. Automated analysis of underwater videos from stationary cameras using deep learning techniques has advanced considerably in recent years, but the use of mobile cameras potentially raises new challenges for existing methods. We tested how well three automation procedures for stationary underwater cameras, taking an object-centric rather than background-centric approach, performed on surveys of fish using a mobile camera. We analyzed underwear drone videos from reef and seagrass habitat to detect and count two marine fisheries species, luderick (Girella tricuspidata) and yellowfin bream (Acanthopagrus australis). Three convolutional neural network (CNN) frameworks were compared: Detectron Faster R-CNN, Detectron2 Faster R-CNN (using a Regional Proposal Network, RPN), and YOLOv5 (a single-stage detector, SSD). Models performed well overall. Per frame, overall F1 scores ranged 81.4 - 87.3%, precision 88.2 – 96.0%, and recall 73.2 - 88.2%. For quantifying MaxN per video, overall F1 ranged 85.9 – 91.4%, precision 81.9 – 95.3%, and recall 87.1 – 91.1%. For luderick, F1 was > 80% for all frameworks per frame and 89% or higher for MaxN. For yellowfin bream, F1 scores were lower (35.0 - 73.8% for frames, 43.4 - 73.0% for MaxN). Detectron2 performed poorly, and YOLOv5 and Detectron performed similarly with advantages depending on metrics and species. For these two frameworks, performance was as good as in videos from stationary cameras. Our findings show that object detection technology is very useful for extracting fish data from mobile underwater cameras for the system tested here. There is a need now to test performance over a wider range of environments to produce generalizable models. The key steps required area to test and enhance performance: 1. for suites of species in the same habitats with different water clarity, 2. in other coastal environments, 3. trialing cameras moving at different speeds, and 4. using different frame-rates.

show abstract

Section: Introductionmentioning

confidence: 99%

Fish surveys on the move: Adapting automated fish detection and classification frameworks for videos on a remotely operated vehicle in shallow marine waters

et al. 2022

View full text Add to dashboard Cite

show abstract

“…A 7.5-cm red laser scaler integrated into the ROV system enabled fish length to be estimated if both lasers struck an individual at an angle of incidence <20°from perpendicular. Bias correction for laser-scaled fish followed the method of Garner et al (2021).…”

Section: Data Collectionmentioning

confidence: 99%

Three trap designs evaluated for a deepwater lionfish fishery

et al. 2023

Self Cite

View full text Add to dashboard Cite

A deepwater (>40 m) fishery for invasive lionfish (Pterois volitans/miles) offers a potential means to control invasive lionfish densities and mitigate their impacts on reefs too deep for SCUBA removals. Trapping could provide a scalable solution—if an effective fishing gear with minimal environmental impacts could be permitted and adopted by fishers. We tested the efficacy of wooden slat lobster traps, wire sea bass traps, and experimental non-containment Gittings traps. One hundred deployments of each trap type were made at 120 mesophotic (38–78 m deep) natural reef sites in the northeastern Gulf of Mexico (29.6–30.1°N, 86.1–87.6°W). Reef sites were surveyed with remotely operated vehicles (ROV) before and after trap deployments, and remote time-lapse video cameras were affixed above 86 traps to sample in situ recruitment to the traps. The video data showed that lionfish were attracted to the vicinity of the three trap types at similar rates, but that lionfish rarely entered the lobster or sea bass traps. The high bycatch rates of sea bass traps suggested their use is likely unsuitable for targeting lionfish. Lobster traps had lower rates of bycatch, but their relatively high ratio of bycatch-to-lionfish catches suggests that modifications will be needed to make them more efficient. The Gittings traps had the highest lionfish catch rates and lowest bycatches of native fishes, but operational issues were also identified. They failed to open on 20% of deployments and one entangled a green sea turtle (Chelonia mydas). Even with the best-performing trap design, the average catch rate of lionfish was less than one lionfish per trap. A potential explanation could be the low biomass of lionfish observed during the ROV surveys, which averaged 0.3 kg lionfish per site. The time-lapse video data suggested that lionfish recruitment to Gittings traps could increase with higher densities of lionfish on the nearby reefs, if traps were retrieved after approximately two days of deployment, and if traps were retrieved during dawn or dusk. Further research, development, and testing is needed for lionfish traps, and critical bio- and techno-economic assessments appear warranted to evaluate the feasibility of a deepwater lionfish fishery.

show abstract

“…Stereo camera systems also have been widely used in fish length estimations [ 39 , 40 , 41 , 42 ], using disparity information to provide 3D information about an object [ 43 , 44 , 45 ]. In aquaculture, many are now putting their interest and efforts into integrating stereo cameras for fish length and biomass estimations [ 40 , 41 , 46 ].…”

Section: Related Workmentioning

confidence: 99%

A Two-Mode Underwater Smart Sensor Object for Precision Aquaculture Based on AIoT Technology

Chang

Ubina

Cheng

et al. 2022

Sensors

View full text Add to dashboard Cite

Monitoring the status of culture fish is an essential task for precision aquaculture using a smart underwater imaging device as a non-intrusive way of sensing to monitor freely swimming fish even in turbid or low-ambient-light waters. This paper developed a two-mode underwater surveillance camera system consisting of a sonar imaging device and a stereo camera. The sonar imaging device has two cloud-based Artificial Intelligence (AI) functions that estimate the quantity and the distribution of the length and weight of fish in a crowded fish school. Because sonar images can be noisy and fish instances of an overcrowded fish school are often overlapped, machine learning technologies, such as Mask R-CNN, Gaussian mixture models, convolutional neural networks, and semantic segmentation networks were employed to address the difficulty in the analysis of fish in sonar images. Furthermore, the sonar and stereo RGB images were aligned in the 3D space, offering an additional AI function for fish annotation based on RGB images. The proposed two-mode surveillance camera was tested to collect data from aquaculture tanks and off-shore net cages using a cloud-based AIoT system. The accuracy of the proposed AI functions based on human-annotated fish metric data sets were tested to verify the feasibility and suitability of the smart camera for the estimation of remote underwater fish metrics.

show abstract

Estimating reef fish size distributions with a mini remotely operated vehicle-integrated stereo camera system

Cited by 10 publications

References 48 publications

Fish surveys on the move: Adapting automated fish detection and classification frameworks for videos on a remotely operated vehicle in shallow marine waters

Fish surveys on the move: Adapting automated fish detection and classification frameworks for videos on a remotely operated vehicle in shallow marine waters

Three trap designs evaluated for a deepwater lionfish fishery

A Two-Mode Underwater Smart Sensor Object for Precision Aquaculture Based on AIoT Technology

Contact Info

Product

Resources

About