Use of computer vision onboard fishing vessels to quantify catches: The iObserver

Vilas, Carlos; Antelo, Luís T.; Martín-Rodríguez, Fernando; Morales, X.; Pérez‐Martín, Ricardo I.; Alonso, Antonio A.; Valeiras, J.; Abad, E.; Quinzán, M.; Barral-Martinez, Marta

doi:10.1016/j.marpol.2019.103714

Cited by 10 publications

(5 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These include discarding of non-commercial species, vessel size, distribution and operational range of the fleet, and the fact that the analysis of catch is based on time-consuming procedures such as visual sorting, taxonomic classification, counting, measuring, weighing, and/or tissue sampling. For all these reasons, the collection of fishery-dependent data is often limited to subsets of the fleet, compromising the accuracy and representativeness of the results achieved (Vilas et al 2019). On the other hand, fishery-independent data, which are mainly collected by scientific surveys explicitly designed to capture resource distribution and status, depend on huge operational ship-time costs (Dennis et al 2015).…”

Section: Introductionmentioning

confidence: 99%

All is fish that comes to the net: metabarcoding for rapid fisheries catch assessment

Russo

Maiello

Talarico

et al. 2021

Ecological Applications

View full text Add to dashboard Cite

Monitoring marine resource exploitation is a key activity in fisheries science and biodiversity conservation. Since research surveys are time consuming and costly, fishery‐dependent data (i.e., derived directly from fishing vessels) are increasingly credited with a key role in expanding the reach of ocean monitoring. Fishing vessels may be seen as widely ranging data‐collecting platforms, which could act as a fleet of sentinels for monitoring marine life, in particular exploited stocks. Here, we investigate the possibility of assessing catch composition of single hauls carried out by trawlers by applying DNA metabarcoding to the dense water draining from fishing nets just after the end of hauling operations (hereafter “slush”). We assess the performance of this approach in portraying β‐diversity and examining the quantitative relationship between species abundances in the catch and DNA amount in the slush (read counts generated by amplicon sequencing). We demonstrate that the assemblages identified using DNA in the slush satisfactorily mirror those returned by visual inspection of net content (about 71% of species and 86% of families of fish) and detect a strong relationship between read counts and species abundances in the catch. We therefore argue that this approach could be upscaled to serve as a powerful source of information on the structure of demersal assemblages and the impact of fisheries.

show abstract

Section: Introductionmentioning

confidence: 99%

All is fish that comes to the net: metabarcoding for rapid fisheries catch assessment

Russo

Maiello

Talarico

et al. 2021

Ecological Applications

View full text Add to dashboard Cite

show abstract

“…To maximise knowledge acquisition, marine research also relies on fisheries-dependent information, which is still largely based on traditional approaches, such as logbook data, visual inspection and sorting of species. These are usually performed by fisheries observers or the fishers themselves and, given that they require time, are consequently limited to subsets of the fleet, compromising the accuracy and representativeness of the results (Vilas et al, 2019). Promisingly, technological innovations are offering solutions to update and modernize fishery data collection (Bradley et al, 2019;Plet-Hansen et al, 2019).…”

mentioning

confidence: 99%

Little samplers, big fleet: eDNA metabarcoding from commercial trawlers enhances ocean monitoring

et al. 2022

View full text Add to dashboard Cite

“… Álvarez-Ellacuría et al (2020) propose the use of a deep convolutional network (Mask R-CNN) for unsupervised length estimation from images of European hake boxes collected at the fish market. Vilas et al (2020) address the problem of fish catch quantification on vessels using computer vision, and French et al (2019) the automated monitoring of fishing discards. However, none of the reviewed works above focuses on the analysis of images with varied fish species on auction trays at the fish market.…”

Section: Previous Workmentioning

confidence: 99%

Simultaneous, vision-based fish instance segmentation, species classification and size regression

Climent-Perez,

Galán-Cuenca,

Garcia-d’Urso

et al. 2024

PeerJ Computer Science

View full text Add to dashboard Cite

Overexploitation of fisheries is a worldwide problem, which is leading to a large loss of diversity, and affects human communities indirectly through the loss of traditional jobs, cultural heritage, etc. To address this issue, governments have started accumulating data on fishing activities, to determine biomass extraction rates, and fisheries status. However, these data are often estimated from small samplings, which can lead to partially inaccurate assessments. Fishing can also benefit of the digitization process that many industries are undergoing. Wholesale fish markets, where vessels disembark, can be the point of contact to retrieve valuable information on biomass extraction rates, and can do so automatically. Fine-grained knowledge about the fish species, quantities, sizes, etc. that are caught can be therefore very valuable to all stakeholders, and particularly decision-makers regarding fisheries conservation, sustainable, and long-term exploitation. In this regard, this article presents a full workflow for fish instance segmentation, species classification, and size estimation from uncalibrated images of fish trays at the fish market, in order to automate information extraction that can be helpful in such scenarios. Our results on fish instance segmentation and species classification show an overall mean average precision (mAP) at 50% intersection-over-union (IoU) of 70.42%, while fish size estimation shows a mean average error (MAE) of only 1.27 cm.

show abstract

Use of computer vision onboard fishing vessels to quantify catches: The iObserver

Cited by 10 publications

References 15 publications

All is fish that comes to the net: metabarcoding for rapid fisheries catch assessment

All is fish that comes to the net: metabarcoding for rapid fisheries catch assessment

Little samplers, big fleet: eDNA metabarcoding from commercial trawlers enhances ocean monitoring

Simultaneous, vision-based fish instance segmentation, species classification and size regression

Contact Info

Product

Resources

About