Towards an automated protocol for wildlife density estimation using camera-traps

Camera-traps are valuable tools for estimating wildlife population density, and recently developed models enable density estimation without the need for individual recognition. Still, processing and analysis of camera-trap data are extremely time-consuming. While algorithms for automated species cla...

Descripción completa

Detalles Bibliográficos
Autores: Zampetti, Andrea, Mirante, Davide, Palencia, Pablo, Santini, Luca
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2024
País:España
Institución:Consejo Superior de Investigaciones Científicas (CSIC)
Repositorio:DIGITAL.CSIC. Repositorio Institucional del CSIC
OAI Identifier:oai:digital.csic.es:10261/380809
Acceso en línea:http://hdl.handle.net/10261/380809
https://api.elsevier.com/content/abstract/scopus_id/85208789059
Access Level:acceso abierto
Palabra clave:Automatization
Camera-traps
Density estimation
Distance sampling
Machine learning
MegaDetector
Random Encounter Model
Species classification
id ES_03e32fa86a04fdf8d03f26a7ba19760f
oai_identifier_str oai:digital.csic.es:10261/380809
network_acronym_str ES
network_name_str España
repository_id_str
dc.title.none.fl_str_mv Towards an automated protocol for wildlife density estimation using camera-traps
title Towards an automated protocol for wildlife density estimation using camera-traps
spellingShingle Towards an automated protocol for wildlife density estimation using camera-traps
Zampetti, Andrea
Automatization
Camera-traps
Density estimation
Distance sampling
Machine learning
MegaDetector
Random Encounter Model
Species classification
title_short Towards an automated protocol for wildlife density estimation using camera-traps
title_full Towards an automated protocol for wildlife density estimation using camera-traps
title_fullStr Towards an automated protocol for wildlife density estimation using camera-traps
title_full_unstemmed Towards an automated protocol for wildlife density estimation using camera-traps
title_sort Towards an automated protocol for wildlife density estimation using camera-traps
dc.creator.none.fl_str_mv Zampetti, Andrea
Mirante, Davide
Palencia, Pablo
Santini, Luca
author Zampetti, Andrea
author_facet Zampetti, Andrea
Mirante, Davide
Palencia, Pablo
Santini, Luca
author_role author
author2 Mirante, Davide
Palencia, Pablo
Santini, Luca
author2_role author
author
author
dc.contributor.none.fl_str_mv Ministero dell'Istruzione, dell'Università e della Ricerca
Universidad de Oviedo
Ministerio de Ciencia e Innovación (España)
Agencia Estatal de Investigación (España)
Zampetti, Andrea [0009-0001-0495-5416]
Mirante, Davide [0009-0003-2258-0907]
Palencia, Pablo [0000-0002-2928-4241]
Santini, Luca [0000-0002-5418-3688]
Consejo Superior de Investigaciones Científicas [https://ror.org/02gfc7t72]
dc.subject.none.fl_str_mv Automatization
Camera-traps
Density estimation
Distance sampling
Machine learning
MegaDetector
Random Encounter Model
Species classification
topic Automatization
Camera-traps
Density estimation
Distance sampling
Machine learning
MegaDetector
Random Encounter Model
Species classification
description Camera-traps are valuable tools for estimating wildlife population density, and recently developed models enable density estimation without the need for individual recognition. Still, processing and analysis of camera-trap data are extremely time-consuming. While algorithms for automated species classification are becoming more common, they have only served as supporting tools, limiting their true potential in being implemented in ecological analyses without human supervision. Here, we assessed the capability of two camera-trap based models to provide robust density estimates when image classification is carried out by machine learning algorithms. We simulated density estimation with Camera-Trap Distance Sampling (CT-DS) and Random Encounter Model (REM) under different scenarios of automated image classification. We then applied the two models to obtain density estimates of three focal species (roe deer Capreolus capreolus, red fox Vulpes vulpes and Eurasian badger Meles meles) in a reserve in central Italy. Species detection and classification was carried out both by the user and machine learning algorithms (respectively, MegaDetector and Wildlife Insights), and all outputs were used to estimate density and ultimately compared. Simulation results suggested that the CT-DS model could provide robust density estimates even at poor algorithm performances (down to 50% of correctly classified images), while the REM model is more unpredictable and depends on multiple factors. Density estimates obtained from the MegaDetector output were highly consistent for both models with the manually labelled images. While Wildlife Insights' performance differed greatly between species (recall: badger = 0.15; roe deer = 0.56; fox = 0.75), CT-DS estimates did not vary significantly; on the contrary, REM systematically overestimated density, with little overlap in standard errors. We conclude that CT-DS and REM models can be robust to the loss of images when machine learning algorithms are used to identify animals, with the CT-DS being an ideal candidate for applications in a fully unsupervised framework. We propose guidelines to evaluate when and how to integrate machine learning in the analysis of camera-trap data for density estimation, further strengthening the applicability of camera-traps as a cost-effective method for density estimation in (spatially and temporally) extensive multi-species monitoring programmes.
publishDate 2024
dc.date.none.fl_str_mv 2024
2025
2025
dc.type.none.fl_str_mv info:eu-repo/semantics/article
http://purl.org/coar/resource_type/c_6501
Publisher's version
info:eu-repo/semantics/publishedVersion
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv http://hdl.handle.net/10261/380809
https://api.elsevier.com/content/abstract/scopus_id/85208789059
url http://hdl.handle.net/10261/380809
https://api.elsevier.com/content/abstract/scopus_id/85208789059
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv Zampetti, A.; (2024); Data and code availability: Towards an automated protocol for wildlife density estimation using camera-traps; Zenodo; https://doi.org/10.5281/zenodo.13142672
https://doi.org/10.1111/2041-210X.14450
Sí
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv John Wiley & Sons
publisher.none.fl_str_mv John Wiley & Sons
dc.source.none.fl_str_mv reponame:DIGITAL.CSIC. Repositorio Institucional del CSIC
instname:Consejo Superior de Investigaciones Científicas (CSIC)
instname_str Consejo Superior de Investigaciones Científicas (CSIC)
reponame_str DIGITAL.CSIC. Repositorio Institucional del CSIC
collection DIGITAL.CSIC. Repositorio Institucional del CSIC
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869402757016846336
spelling Towards an automated protocol for wildlife density estimation using camera-trapsZampetti, AndreaMirante, DavidePalencia, PabloSantini, LucaAutomatizationCamera-trapsDensity estimationDistance samplingMachine learningMegaDetectorRandom Encounter ModelSpecies classificationCamera-traps are valuable tools for estimating wildlife population density, and recently developed models enable density estimation without the need for individual recognition. Still, processing and analysis of camera-trap data are extremely time-consuming. While algorithms for automated species classification are becoming more common, they have only served as supporting tools, limiting their true potential in being implemented in ecological analyses without human supervision. Here, we assessed the capability of two camera-trap based models to provide robust density estimates when image classification is carried out by machine learning algorithms. We simulated density estimation with Camera-Trap Distance Sampling (CT-DS) and Random Encounter Model (REM) under different scenarios of automated image classification. We then applied the two models to obtain density estimates of three focal species (roe deer Capreolus capreolus, red fox Vulpes vulpes and Eurasian badger Meles meles) in a reserve in central Italy. Species detection and classification was carried out both by the user and machine learning algorithms (respectively, MegaDetector and Wildlife Insights), and all outputs were used to estimate density and ultimately compared. Simulation results suggested that the CT-DS model could provide robust density estimates even at poor algorithm performances (down to 50% of correctly classified images), while the REM model is more unpredictable and depends on multiple factors. Density estimates obtained from the MegaDetector output were highly consistent for both models with the manually labelled images. While Wildlife Insights' performance differed greatly between species (recall: badger = 0.15; roe deer = 0.56; fox = 0.75), CT-DS estimates did not vary significantly; on the contrary, REM systematically overestimated density, with little overlap in standard errors. We conclude that CT-DS and REM models can be robust to the loss of images when machine learning algorithms are used to identify animals, with the CT-DS being an ideal candidate for applications in a fully unsupervised framework. We propose guidelines to evaluate when and how to integrate machine learning in the analysis of camera-trap data for density estimation, further strengthening the applicability of camera-traps as a cost-effective method for density estimation in (spatially and temporally) extensive multi-species monitoring programmes.D. M. was funded by the Italian Ministry of Education, University and Research, Programma Operativo Nazionale (CUP B85F21005360001). P. P. received support from the University of Oviedo through a Juan de la Cierva contract JDC2022-048567-I supported by ‘Ministerio de Ciencia e Innovación’, ‘Agencia Estatal de Investigacion’ and ‘NextGeneration EU’ (MCIN/AEI/10.13039/501100011033). All the authors would like to thank Gianfranco Pisa, Cristina Santocchi and all the staff of Tenuta Sant'Egidio for their precious field support and collaboration for data collection.Peer reviewedJohn Wiley & SonsMinistero dell'Istruzione, dell'Università e della RicercaUniversidad de OviedoMinisterio de Ciencia e Innovación (España)Agencia Estatal de Investigación (España)Zampetti, Andrea [0009-0001-0495-5416]Mirante, Davide [0009-0003-2258-0907]Palencia, Pablo [0000-0002-2928-4241]Santini, Luca [0000-0002-5418-3688]Consejo Superior de Investigaciones Científicas [https://ror.org/02gfc7t72]202520252024info:eu-repo/semantics/articlehttp://purl.org/coar/resource_type/c_6501Publisher's versioninfo:eu-repo/semantics/publishedVersionapplication/pdfhttp://hdl.handle.net/10261/380809https://api.elsevier.com/content/abstract/scopus_id/85208789059reponame:DIGITAL.CSIC. Repositorio Institucional del CSICinstname:Consejo Superior de Investigaciones Científicas (CSIC)InglésZampetti, A.; (2024); Data and code availability: Towards an automated protocol for wildlife density estimation using camera-traps; Zenodo; https://doi.org/10.5281/zenodo.13142672https://doi.org/10.1111/2041-210X.14450Síinfo:eu-repo/semantics/openAccessoai:digital.csic.es:10261/3808092026-05-22T06:33:51Z
score 15,811543

Similares en LA Referencia