A Deep Learning Approach to Assist Sustainability of Demersal Trawling Operations

Bycatch in demersal trawl fisheries challenges their sustainability despite the implementation of the various gear technical regulations. A step towards extended control over the catch process can be established through a real-time catch monitoring tool that will allow fishers to react to unwanted c...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Maria Sokolova, Adrià Mompó Alepuz, Fletcher Thompson, Patrizio Mariani, Roberto Galeazzi, Ludvig Ahm Krag
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	deep learning innovation in fisheries digitized fishery automated catch description Environmental effects of industries and plants TD194-195 Renewable energy sources TJ807-830 Environmental sciences GE1-350
Acceso en línea:	https://doaj.org/article/2d631acd4bf742b29962e6d04684b5f4
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:2d631acd4bf742b29962e6d04684b5f4
record_format	dspace
spelling	oai:doaj.org-article:2d631acd4bf742b29962e6d04684b5f42021-11-25T19:00:27ZA Deep Learning Approach to Assist Sustainability of Demersal Trawling Operations10.3390/su1322123622071-1050https://doaj.org/article/2d631acd4bf742b29962e6d04684b5f42021-11-01T00:00:00Zhttps://www.mdpi.com/2071-1050/13/22/12362https://doaj.org/toc/2071-1050Bycatch in demersal trawl fisheries challenges their sustainability despite the implementation of the various gear technical regulations. A step towards extended control over the catch process can be established through a real-time catch monitoring tool that will allow fishers to react to unwanted catch compositions. In this study, for the first time in the commercial demersal trawl fishery sector, we introduce an automated catch description that leverages state-of-the-art region based convolutional neural network (Mask R-CNN) architecture and builds upon an in-trawl novel image acquisition system. The system is optimized for applications in <i>Nephrops</i> fishery and enables the classification and count of catch items during fishing operation. The detector robustness was improved with augmentation techniques applied during training on a custom high-resolution dataset obtained during extensive demersal trawling. The resulting algorithms were tested on video footage representing both the normal towing process and haul-back conditions. The algorithm obtained an F-score of 0.79. The resulting automated catch description was compared with the manual catch count showing low absolute error during towing. Current practices in demersal trawl fisheries are carried out without any indications of catch composition nor whether the catch enters the fishing gear. Hence, the proposed solution provides a substantial technical contribution to making this type of fishery more targeted, paving the way to further optimization of fishing activities aiming at increasing target catch while reducing unwanted bycatch.Maria SokolovaAdrià Mompó AlepuzFletcher ThompsonPatrizio MarianiRoberto GaleazziLudvig Ahm KragMDPI AGarticledeep learninginnovation in fisheriesdigitized fisheryautomated catch descriptionEnvironmental effects of industries and plantsTD194-195Renewable energy sourcesTJ807-830Environmental sciencesGE1-350ENSustainability, Vol 13, Iss 12362, p 12362 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	deep learning innovation in fisheries digitized fishery automated catch description Environmental effects of industries and plants TD194-195 Renewable energy sources TJ807-830 Environmental sciences GE1-350
spellingShingle	deep learning innovation in fisheries digitized fishery automated catch description Environmental effects of industries and plants TD194-195 Renewable energy sources TJ807-830 Environmental sciences GE1-350 Maria Sokolova Adrià Mompó Alepuz Fletcher Thompson Patrizio Mariani Roberto Galeazzi Ludvig Ahm Krag A Deep Learning Approach to Assist Sustainability of Demersal Trawling Operations
description	Bycatch in demersal trawl fisheries challenges their sustainability despite the implementation of the various gear technical regulations. A step towards extended control over the catch process can be established through a real-time catch monitoring tool that will allow fishers to react to unwanted catch compositions. In this study, for the first time in the commercial demersal trawl fishery sector, we introduce an automated catch description that leverages state-of-the-art region based convolutional neural network (Mask R-CNN) architecture and builds upon an in-trawl novel image acquisition system. The system is optimized for applications in <i>Nephrops</i> fishery and enables the classification and count of catch items during fishing operation. The detector robustness was improved with augmentation techniques applied during training on a custom high-resolution dataset obtained during extensive demersal trawling. The resulting algorithms were tested on video footage representing both the normal towing process and haul-back conditions. The algorithm obtained an F-score of 0.79. The resulting automated catch description was compared with the manual catch count showing low absolute error during towing. Current practices in demersal trawl fisheries are carried out without any indications of catch composition nor whether the catch enters the fishing gear. Hence, the proposed solution provides a substantial technical contribution to making this type of fishery more targeted, paving the way to further optimization of fishing activities aiming at increasing target catch while reducing unwanted bycatch.
format	article
author	Maria Sokolova Adrià Mompó Alepuz Fletcher Thompson Patrizio Mariani Roberto Galeazzi Ludvig Ahm Krag
author_facet	Maria Sokolova Adrià Mompó Alepuz Fletcher Thompson Patrizio Mariani Roberto Galeazzi Ludvig Ahm Krag
author_sort	Maria Sokolova
title	A Deep Learning Approach to Assist Sustainability of Demersal Trawling Operations
title_short	A Deep Learning Approach to Assist Sustainability of Demersal Trawling Operations
title_full	A Deep Learning Approach to Assist Sustainability of Demersal Trawling Operations
title_fullStr	A Deep Learning Approach to Assist Sustainability of Demersal Trawling Operations
title_full_unstemmed	A Deep Learning Approach to Assist Sustainability of Demersal Trawling Operations
title_sort	deep learning approach to assist sustainability of demersal trawling operations
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/2d631acd4bf742b29962e6d04684b5f4
work_keys_str_mv	AT mariasokolova adeeplearningapproachtoassistsustainabilityofdemersaltrawlingoperations AT adriamompoalepuz adeeplearningapproachtoassistsustainabilityofdemersaltrawlingoperations AT fletcherthompson adeeplearningapproachtoassistsustainabilityofdemersaltrawlingoperations AT patriziomariani adeeplearningapproachtoassistsustainabilityofdemersaltrawlingoperations AT robertogaleazzi adeeplearningapproachtoassistsustainabilityofdemersaltrawlingoperations AT ludvigahmkrag adeeplearningapproachtoassistsustainabilityofdemersaltrawlingoperations AT mariasokolova deeplearningapproachtoassistsustainabilityofdemersaltrawlingoperations AT adriamompoalepuz deeplearningapproachtoassistsustainabilityofdemersaltrawlingoperations AT fletcherthompson deeplearningapproachtoassistsustainabilityofdemersaltrawlingoperations AT patriziomariani deeplearningapproachtoassistsustainabilityofdemersaltrawlingoperations AT robertogaleazzi deeplearningapproachtoassistsustainabilityofdemersaltrawlingoperations AT ludvigahmkrag deeplearningapproachtoassistsustainabilityofdemersaltrawlingoperations
_version_	1718410392960499712

A Deep Learning Approach to Assist Sustainability of Demersal Trawling Operations

Ejemplares similares