Inteligência artificial na análise de dados de um plantio comercial de Eucalyptus saligna Smith por meio de imagens multiespectrais
Autor(a) principal: | |
---|---|
Data de Publicação: | 2024 |
Tipo de documento: | Tese |
Idioma: | por |
Título da fonte: | Manancial - Repositório Digital da UFSM |
Texto Completo: | http://repositorio.ufsm.br/handle/1/31734 |
Resumo: | The advancements in Remotely Piloted Aircraft (RPA), along with artificial intelligence, have been facilitating and enhancing field data collection in terms of temporal and spatial accuracy, with the possibility of creating customized datasets according to specific needs. The main goal of this thesis is to apply artificial intelligence techniques for identification and count estimation of Eucalyptus saligna Smith using high spatial resolution multispectral images onboard a Remotely Piloted Aircraft (RPA). The study area is located in the municipality of Eldorado do Sul - RS. Spectral reflectance curves of leaves were obtained using the Analytical Spectral Devices (ASD) FieldSpec® 3 spectroradiometer. For each of the three selected plant species, five measurements were taken, resulting in a total of 15 readings for each species. For this analysis, the wavelength ranges of the FieldSpec® 3 and those restricted to the four bands (Green, Red, RedEdge, and NIR) of the Parrot Sequoia® multispectral sensor were considered. Subsequently, the data were subjected to the Shapiro-Wilk test to check for normality at the 95% significance level. Given the absence of normal distribution, the non-parametric Kruskal-Wallis test was chosen. The results obtained from the wavelengths of the FieldSpec® 3 and the Parrot Sequoia® sensor demonstrated the ability to characterize and distinguish E. saligna from most spontaneous plants (SP), particularly in the RedEdge and NIR range. At another instance, images of the forest plantation at 180 days post-planting were also acquired via the RPA. Machine learning algorithms (Random Forest - RF) and deep learning (YOLOv8n) were used to identify, detect, and count E. saligna, respectively. The RF algorithm proved to be efficient in identifying E. saligna and the other thematic classes analyzed (SP, exposed soil, and shadow), achieving an overall accuracy of 93%. The YOLOv8n model showed promising results both in detection (recall of 0.93) and count estimation (0.915), demonstrating excellent performance. The results suggest that the use of RPA, multispectral images, and advanced technologies are highly promising for the forestry sector. |
id |
UFSM_ce272d97e53d9f7924dda2b012cd1897 |
---|---|
oai_identifier_str |
oai:repositorio.ufsm.br:1/31734 |
network_acronym_str |
UFSM |
network_name_str |
Manancial - Repositório Digital da UFSM |
repository_id_str |
|
spelling |
Inteligência artificial na análise de dados de um plantio comercial de Eucalyptus saligna Smith por meio de imagens multiespectraisArtificial intelligence in data analysis from a commercial planting of Eucalyptus saligna Smith using multispectral imagesAprendizado profundoAprendizado de máquinaAeronave remotamente pilotadaDeep learningMachine learningRemotely piloted aircraftCNPQ::CIENCIAS AGRARIAS::RECURSOS FLORESTAIS E ENGENHARIA FLORESTALThe advancements in Remotely Piloted Aircraft (RPA), along with artificial intelligence, have been facilitating and enhancing field data collection in terms of temporal and spatial accuracy, with the possibility of creating customized datasets according to specific needs. The main goal of this thesis is to apply artificial intelligence techniques for identification and count estimation of Eucalyptus saligna Smith using high spatial resolution multispectral images onboard a Remotely Piloted Aircraft (RPA). The study area is located in the municipality of Eldorado do Sul - RS. Spectral reflectance curves of leaves were obtained using the Analytical Spectral Devices (ASD) FieldSpec® 3 spectroradiometer. For each of the three selected plant species, five measurements were taken, resulting in a total of 15 readings for each species. For this analysis, the wavelength ranges of the FieldSpec® 3 and those restricted to the four bands (Green, Red, RedEdge, and NIR) of the Parrot Sequoia® multispectral sensor were considered. Subsequently, the data were subjected to the Shapiro-Wilk test to check for normality at the 95% significance level. Given the absence of normal distribution, the non-parametric Kruskal-Wallis test was chosen. The results obtained from the wavelengths of the FieldSpec® 3 and the Parrot Sequoia® sensor demonstrated the ability to characterize and distinguish E. saligna from most spontaneous plants (SP), particularly in the RedEdge and NIR range. At another instance, images of the forest plantation at 180 days post-planting were also acquired via the RPA. Machine learning algorithms (Random Forest - RF) and deep learning (YOLOv8n) were used to identify, detect, and count E. saligna, respectively. The RF algorithm proved to be efficient in identifying E. saligna and the other thematic classes analyzed (SP, exposed soil, and shadow), achieving an overall accuracy of 93%. The YOLOv8n model showed promising results both in detection (recall of 0.93) and count estimation (0.915), demonstrating excellent performance. The results suggest that the use of RPA, multispectral images, and advanced technologies are highly promising for the forestry sector.Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPESOs avanços das Aeronaves Remotamente Pilotadas (ARPs), juntamente com a inteligência artifical, vêm facilitando e melhorando a coleta dos dados de campo em termos de precisão temporal e espacial, com a possibilidade de criar conjuntos de dados personalizados de acordo com necessidades específicas. A presente tese tem como objetivo geral aplicar técnicas de inteligência artificial para identificação e estimativa de contagem de Eucalyptus saligna Smith, por meio da utilização de imagens multiespectrais de alta resolução espacial embarcado em Aeronave Remotamente Pilotada (ARP). A área de estudo está inserida no município de Eldorado do Sul - RS. Foram obtidas curvas de reflectância espectral das folhas por meio do espectrorradiômetro Analytical Spectral Devices (ASD) FieldSpec® 3. Para cada uma das três espécies de plantas selecionadas, foram realizadas cinco medições, resultando em um total de 15 leituras para cada espécie. Para essa análise, levou-se em consideração os intervalos de comprimento de onda do FieldSpec® 3 e os intervalos restritos às quatro bandas (Green, Red, RedEdge e NIR) do sensor multiespectral da Parrot Sequoia®. Posteriormente, os dados foram submetidos ao teste de Shapiro-Wilk para verificação da normalidade ao nível de significância de 95%. Dada a ausência da distribuição normal, optou-se pelo teste não-paramétrico de Kruskal-Wallis. Os resultados obtidos dos comprimentos de onda do FieldSpec® 3 e do sensor Parrot Sequoia® demonstraram capacidade de caracterizar e distinguir o E. saligna em relação à maioria das plantas espontâneas (PEs), principalmente na faixa da RedEdge e NIR. Em outro momento, também foram adquiridas imagens do plantio florestal aos 180 dias pós-plantio por meio da ARP. Algoritmos de aprendizado de máquina (Random Forest - RF) e aprendizado profundo (YOLOv8n) foram utilizados para identificar, detectar e contar o E. saligna, respetivamente. O algoritmo RF demonstrou ser eficiente na identificação do E. saligna e das demais classes temáticas analisadas (PEs, solo exposto e sombra), obtendo uma acurácia global de 93%. O modelo YOLOv8n demonstrou ser promissor tanto na detecção (recall de 0,93) quanto na estimativa de contagem (0,915) demonstrando um desempenho excelente. Os resultados obtidos sugerem que o uso de ARP, imagens multiespectrais e tecnologias avançadas são altamente promissores para o setor florestal.Universidade Federal de Santa MariaBrasilRecursos Florestais e Engenharia FlorestalUFSMPrograma de Pós-Graduação em Engenharia FlorestalCentro de Ciências RuraisPereira, Rudiney Soareshttp://lattes.cnpq.br/9479801378014588Benedetti, Ana Caroline PaimMarchesan, JulianaWeiler, Elenice BroettoEugenio, Fernando CoelhoFantinel, Roberta Aparecida2024-04-10T13:25:03Z2024-04-10T13:25:03Z2024-03-01info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/doctoralThesisapplication/pdfhttp://repositorio.ufsm.br/handle/1/31734porAttribution-NonCommercial-NoDerivatives 4.0 Internationalhttp://creativecommons.org/licenses/by-nc-nd/4.0/info:eu-repo/semantics/openAccessreponame:Manancial - Repositório Digital da UFSMinstname:Universidade Federal de Santa Maria (UFSM)instacron:UFSM2024-04-10T13:25:03Zoai:repositorio.ufsm.br:1/31734Biblioteca Digital de Teses e Dissertaçõeshttps://repositorio.ufsm.br/ONGhttps://repositorio.ufsm.br/oai/requestatendimento.sib@ufsm.br||tedebc@gmail.comopendoar:2024-04-10T13:25:03Manancial - Repositório Digital da UFSM - Universidade Federal de Santa Maria (UFSM)false |
dc.title.none.fl_str_mv |
Inteligência artificial na análise de dados de um plantio comercial de Eucalyptus saligna Smith por meio de imagens multiespectrais Artificial intelligence in data analysis from a commercial planting of Eucalyptus saligna Smith using multispectral images |
title |
Inteligência artificial na análise de dados de um plantio comercial de Eucalyptus saligna Smith por meio de imagens multiespectrais |
spellingShingle |
Inteligência artificial na análise de dados de um plantio comercial de Eucalyptus saligna Smith por meio de imagens multiespectrais Fantinel, Roberta Aparecida Aprendizado profundo Aprendizado de máquina Aeronave remotamente pilotada Deep learning Machine learning Remotely piloted aircraft CNPQ::CIENCIAS AGRARIAS::RECURSOS FLORESTAIS E ENGENHARIA FLORESTAL |
title_short |
Inteligência artificial na análise de dados de um plantio comercial de Eucalyptus saligna Smith por meio de imagens multiespectrais |
title_full |
Inteligência artificial na análise de dados de um plantio comercial de Eucalyptus saligna Smith por meio de imagens multiespectrais |
title_fullStr |
Inteligência artificial na análise de dados de um plantio comercial de Eucalyptus saligna Smith por meio de imagens multiespectrais |
title_full_unstemmed |
Inteligência artificial na análise de dados de um plantio comercial de Eucalyptus saligna Smith por meio de imagens multiespectrais |
title_sort |
Inteligência artificial na análise de dados de um plantio comercial de Eucalyptus saligna Smith por meio de imagens multiespectrais |
author |
Fantinel, Roberta Aparecida |
author_facet |
Fantinel, Roberta Aparecida |
author_role |
author |
dc.contributor.none.fl_str_mv |
Pereira, Rudiney Soares http://lattes.cnpq.br/9479801378014588 Benedetti, Ana Caroline Paim Marchesan, Juliana Weiler, Elenice Broetto Eugenio, Fernando Coelho |
dc.contributor.author.fl_str_mv |
Fantinel, Roberta Aparecida |
dc.subject.por.fl_str_mv |
Aprendizado profundo Aprendizado de máquina Aeronave remotamente pilotada Deep learning Machine learning Remotely piloted aircraft CNPQ::CIENCIAS AGRARIAS::RECURSOS FLORESTAIS E ENGENHARIA FLORESTAL |
topic |
Aprendizado profundo Aprendizado de máquina Aeronave remotamente pilotada Deep learning Machine learning Remotely piloted aircraft CNPQ::CIENCIAS AGRARIAS::RECURSOS FLORESTAIS E ENGENHARIA FLORESTAL |
description |
The advancements in Remotely Piloted Aircraft (RPA), along with artificial intelligence, have been facilitating and enhancing field data collection in terms of temporal and spatial accuracy, with the possibility of creating customized datasets according to specific needs. The main goal of this thesis is to apply artificial intelligence techniques for identification and count estimation of Eucalyptus saligna Smith using high spatial resolution multispectral images onboard a Remotely Piloted Aircraft (RPA). The study area is located in the municipality of Eldorado do Sul - RS. Spectral reflectance curves of leaves were obtained using the Analytical Spectral Devices (ASD) FieldSpec® 3 spectroradiometer. For each of the three selected plant species, five measurements were taken, resulting in a total of 15 readings for each species. For this analysis, the wavelength ranges of the FieldSpec® 3 and those restricted to the four bands (Green, Red, RedEdge, and NIR) of the Parrot Sequoia® multispectral sensor were considered. Subsequently, the data were subjected to the Shapiro-Wilk test to check for normality at the 95% significance level. Given the absence of normal distribution, the non-parametric Kruskal-Wallis test was chosen. The results obtained from the wavelengths of the FieldSpec® 3 and the Parrot Sequoia® sensor demonstrated the ability to characterize and distinguish E. saligna from most spontaneous plants (SP), particularly in the RedEdge and NIR range. At another instance, images of the forest plantation at 180 days post-planting were also acquired via the RPA. Machine learning algorithms (Random Forest - RF) and deep learning (YOLOv8n) were used to identify, detect, and count E. saligna, respectively. The RF algorithm proved to be efficient in identifying E. saligna and the other thematic classes analyzed (SP, exposed soil, and shadow), achieving an overall accuracy of 93%. The YOLOv8n model showed promising results both in detection (recall of 0.93) and count estimation (0.915), demonstrating excellent performance. The results suggest that the use of RPA, multispectral images, and advanced technologies are highly promising for the forestry sector. |
publishDate |
2024 |
dc.date.none.fl_str_mv |
2024-04-10T13:25:03Z 2024-04-10T13:25:03Z 2024-03-01 |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/doctoralThesis |
format |
doctoralThesis |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://repositorio.ufsm.br/handle/1/31734 |
url |
http://repositorio.ufsm.br/handle/1/31734 |
dc.language.iso.fl_str_mv |
por |
language |
por |
dc.rights.driver.fl_str_mv |
Attribution-NonCommercial-NoDerivatives 4.0 International http://creativecommons.org/licenses/by-nc-nd/4.0/ info:eu-repo/semantics/openAccess |
rights_invalid_str_mv |
Attribution-NonCommercial-NoDerivatives 4.0 International http://creativecommons.org/licenses/by-nc-nd/4.0/ |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.publisher.none.fl_str_mv |
Universidade Federal de Santa Maria Brasil Recursos Florestais e Engenharia Florestal UFSM Programa de Pós-Graduação em Engenharia Florestal Centro de Ciências Rurais |
publisher.none.fl_str_mv |
Universidade Federal de Santa Maria Brasil Recursos Florestais e Engenharia Florestal UFSM Programa de Pós-Graduação em Engenharia Florestal Centro de Ciências Rurais |
dc.source.none.fl_str_mv |
reponame:Manancial - Repositório Digital da UFSM instname:Universidade Federal de Santa Maria (UFSM) instacron:UFSM |
instname_str |
Universidade Federal de Santa Maria (UFSM) |
instacron_str |
UFSM |
institution |
UFSM |
reponame_str |
Manancial - Repositório Digital da UFSM |
collection |
Manancial - Repositório Digital da UFSM |
repository.name.fl_str_mv |
Manancial - Repositório Digital da UFSM - Universidade Federal de Santa Maria (UFSM) |
repository.mail.fl_str_mv |
atendimento.sib@ufsm.br||tedebc@gmail.com |
_version_ |
1805922038503702528 |