Semantic segmentation of major macroalgae in coastal environments using high-resolution ground imagery and deep learning

Macroalgae are a fundamental component of coastal ecosystems and play a key role in shaping community structure and functioning. Macroalgae are currently threatened by diverse stressors, particularly climate change and invasive species, but they do not all respond in the same way to the stressors. E...

Descripción completa

Detalles Bibliográficos
Autores: Balado, Jesús, Rodríguez Pérez, José Ramón, Arias, Pedro, Olabarria, Celia, Martinez Sánchez, Joaquín
Tipo de recurso: artículo
Estado:Versión enviada para evaluación y publicación
Fecha de publicación:2021
País:España
Institución:Universidad de León
Repositorio:BULERIA. Repositorio Institucional de la Universidad de León
OAI Identifier:oai:buleria.unileon.es:10612/17918
Acceso en línea:https://hdl.handle.net/10612/17918
https://doi.org/10.1080/01431161.2020.1842543
Access Level:acceso abierto
Palabra clave:Ecología. Medio ambiente
Macroalgae
Intertidal rocky shore
Convolutional neural networks
Image processing
Semantic segmentation
id ES_9b00c1ca9ca3756fb261b3182d4c8eb2
oai_identifier_str oai:buleria.unileon.es:10612/17918
network_acronym_str ES
network_name_str España
repository_id_str
spelling Semantic segmentation of major macroalgae in coastal environments using high-resolution ground imagery and deep learningBalado, JesúsRodríguez Pérez, José RamónArias, PedroOlabarria, CeliaMartinez Sánchez, JoaquínEcología. Medio ambienteMacroalgaeIntertidal rocky shoreConvolutional neural networksImage processingSemantic segmentationMacroalgae are a fundamental component of coastal ecosystems and play a key role in shaping community structure and functioning. Macroalgae are currently threatened by diverse stressors, particularly climate change and invasive species, but they do not all respond in the same way to the stressors. Effective methods of collecting qualitative and quantitative information are essential to enable better, more efficient management of macroalgae. Acquisition of high-resolution images, in which macroalgae can be distinguished on the basis of their texture and colour, and the automated processing of these images are thus essential. Although ground images are useful, labelling is tedious. This study focuses on the semantic segmentation of five macroalgal species in high-resolution ground images taken in 0.5 x 0.5 m quadrats placed along an intertidal rocky shore at low tide. The target species, Bifurcaria bifurcata, Cystoseira tamariscifolia, Sargassum muticum, Sacchoriza polyschides and Codium spp., which predominate on intertidal shores, belong to different morpho-functional groups. The study explains how to convert vector-labelled data to raster-labelled data for adaptation to convolutional neural network (CNN) input. Three CNNs (MobileNetV2, Resnet18, Xception) were compared, and ResNet18 yielded the highest accuracy (91.9%). The macroalgae were correctly segmented, and the main confusion occurred at the borders between different macroalgal species, a problem derived from labelling errors. In addition, the interior and exterior of the quadrats were correctly delimited by the CNNs. The results were obtained from only one hundred labelled images and can be performed on personal computers, without the need to resort to external servers. The proposed method helps automation of the labelling process.SIFundación Biodiversidad, the Ministerio para la Transición Ecológica y 383 el Reto Demográfico through the Pleamar program, co-funded by the European Maritime and Fisheries Fund (EMFF), call 2018; and Xunta de Galicia for human resources and competitive reference groupsFundación Biodiversidad, the Ministerio para la Transición Ecológica y 383 el Reto Demográfico through the Pleamar program, co-funded by the European Maritime and Fisheries Fund (EMFF), call 2018; and Xunta de Galicia for human resources and competitive reference groupsMinisterio de Ciencia, Innovación y Universidades -Gobierno de EspañaTaylor & FrancisIngeniería Cartografica, Geodesica y FotogrametriaEscuela de Ingeniería Agraria y Forestal2021info:eu-repo/semantics/articleinfo:eu-repo/semantics/submittedVersionhttps://hdl.handle.net/10612/17918https://doi.org/10.1080/01431161.2020.1842543reponame:BULERIA. Repositorio Institucional de la Universidad de Leóninstname:Universidad de LeónInglésED481B-2019-061ED431C 2016-038RTI2018-095893-B-C21http://creativecommons.org/licenses/by-nc-nd/4.0/info:eu-repo/semantics/openAccessoai:buleria.unileon.es:10612/179182026-06-24T12:43:27Z
dc.title.none.fl_str_mv Semantic segmentation of major macroalgae in coastal environments using high-resolution ground imagery and deep learning
title Semantic segmentation of major macroalgae in coastal environments using high-resolution ground imagery and deep learning
spellingShingle Semantic segmentation of major macroalgae in coastal environments using high-resolution ground imagery and deep learning
Balado, Jesús
Ecología. Medio ambiente
Macroalgae
Intertidal rocky shore
Convolutional neural networks
Image processing
Semantic segmentation
title_short Semantic segmentation of major macroalgae in coastal environments using high-resolution ground imagery and deep learning
title_full Semantic segmentation of major macroalgae in coastal environments using high-resolution ground imagery and deep learning
title_fullStr Semantic segmentation of major macroalgae in coastal environments using high-resolution ground imagery and deep learning
title_full_unstemmed Semantic segmentation of major macroalgae in coastal environments using high-resolution ground imagery and deep learning
title_sort Semantic segmentation of major macroalgae in coastal environments using high-resolution ground imagery and deep learning
dc.creator.none.fl_str_mv Balado, Jesús
Rodríguez Pérez, José Ramón
Arias, Pedro
Olabarria, Celia
Martinez Sánchez, Joaquín
author Balado, Jesús
author_facet Balado, Jesús
Rodríguez Pérez, José Ramón
Arias, Pedro
Olabarria, Celia
Martinez Sánchez, Joaquín
author_role author
author2 Rodríguez Pérez, José Ramón
Arias, Pedro
Olabarria, Celia
Martinez Sánchez, Joaquín
author2_role author
author
author
author
dc.contributor.none.fl_str_mv Ingeniería Cartografica, Geodesica y Fotogrametria
Escuela de Ingeniería Agraria y Forestal
dc.subject.none.fl_str_mv Ecología. Medio ambiente
Macroalgae
Intertidal rocky shore
Convolutional neural networks
Image processing
Semantic segmentation
topic Ecología. Medio ambiente
Macroalgae
Intertidal rocky shore
Convolutional neural networks
Image processing
Semantic segmentation
description Macroalgae are a fundamental component of coastal ecosystems and play a key role in shaping community structure and functioning. Macroalgae are currently threatened by diverse stressors, particularly climate change and invasive species, but they do not all respond in the same way to the stressors. Effective methods of collecting qualitative and quantitative information are essential to enable better, more efficient management of macroalgae. Acquisition of high-resolution images, in which macroalgae can be distinguished on the basis of their texture and colour, and the automated processing of these images are thus essential. Although ground images are useful, labelling is tedious. This study focuses on the semantic segmentation of five macroalgal species in high-resolution ground images taken in 0.5 x 0.5 m quadrats placed along an intertidal rocky shore at low tide. The target species, Bifurcaria bifurcata, Cystoseira tamariscifolia, Sargassum muticum, Sacchoriza polyschides and Codium spp., which predominate on intertidal shores, belong to different morpho-functional groups. The study explains how to convert vector-labelled data to raster-labelled data for adaptation to convolutional neural network (CNN) input. Three CNNs (MobileNetV2, Resnet18, Xception) were compared, and ResNet18 yielded the highest accuracy (91.9%). The macroalgae were correctly segmented, and the main confusion occurred at the borders between different macroalgal species, a problem derived from labelling errors. In addition, the interior and exterior of the quadrats were correctly delimited by the CNNs. The results were obtained from only one hundred labelled images and can be performed on personal computers, without the need to resort to external servers. The proposed method helps automation of the labelling process.
publishDate 2021
dc.date.none.fl_str_mv 2021
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/submittedVersion
format article
status_str submittedVersion
dc.identifier.none.fl_str_mv https://hdl.handle.net/10612/17918
https://doi.org/10.1080/01431161.2020.1842543
url https://hdl.handle.net/10612/17918
https://doi.org/10.1080/01431161.2020.1842543
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv ED481B-2019-061
ED431C 2016-038
RTI2018-095893-B-C21
dc.rights.none.fl_str_mv http://creativecommons.org/licenses/by-nc-nd/4.0/
info:eu-repo/semantics/openAccess
rights_invalid_str_mv http://creativecommons.org/licenses/by-nc-nd/4.0/
eu_rights_str_mv openAccess
dc.publisher.none.fl_str_mv Taylor & Francis
publisher.none.fl_str_mv Taylor & Francis
dc.source.none.fl_str_mv reponame:BULERIA. Repositorio Institucional de la Universidad de León
instname:Universidad de León
instname_str Universidad de León
reponame_str BULERIA. Repositorio Institucional de la Universidad de León
collection BULERIA. Repositorio Institucional de la Universidad de León
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869414468241326080
score 15,300719

Similares en LA Referencia