Identifying Nematode Damage on Soybean through Remote Sensing and Machine Learning Techniques
Autor(a) principal: | |
---|---|
Data de Publicação: | 2022 |
Outros Autores: | , , , , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UNESP |
Texto Completo: | http://dx.doi.org/10.3390/agronomy12102404 http://hdl.handle.net/11449/249296 |
Resumo: | Identifying nematode damage in large soybean areas is not always achievable in a practical way. Multispectral reflectance sensors have not been thoroughly evaluated to detect nematode damage in soybeans (Glycine max L.). The main research aims of this study were to: (i) determine the bivariate relationship between individual spectral bands and vegetation indices (VIs) relative to soybean conditions (symptomatic versus asymptomatic), and (ii) to select the best model for identifying plant conditions using three algorithms (logistic regression—LR, random forest—RF, conditional inference tree—CIT) and three options for data input using bands, vegetation indices (VIs), and bands plus VIs. The trial was conducted in Brazil on three on-farm soybean fields presenting different species of nematode infestation. Multispectral imagery was obtained using a drone-mounted MicaSense RedEdge® sensor. At each sampling, georeferenced point nematode infestation and spectral measurements of soybean plants were retrieved for the classification of symptomatic and asymptomatic areas, according to the threshold level adopted. Bivariate analysis of variance (ANOVA), LR, RF, and CIT were used to select the multispectral bands/VIs that discriminated among symptomatic and asymptomatic plants, assessing the best model via their respective parameters for accuracy, sensitivity, and specificity. The greatest classification accuracy (>0.70) was achieved when using the CIT algorithm with the spectral bands only, with green (560 ± 20 nm) and near-infrared (840 ± 40 nm) included as the main spectral input variables in the model. These results demonstrate the potential of combining remotely sensed data and machine learning to distinguish nematode-symptomatic and asymptomatic soybean plants. |
id |
UNSP_b79c4f55dd8e35eed1039da32c97a518 |
---|---|
oai_identifier_str |
oai:repositorio.unesp.br:11449/249296 |
network_acronym_str |
UNSP |
network_name_str |
Repositório Institucional da UNESP |
repository_id_str |
2946 |
spelling |
Identifying Nematode Damage on Soybean through Remote Sensing and Machine Learning Techniquesdigital agriculturedisease detectionmachine learningmultispectral mappingnematodesremote sensingIdentifying nematode damage in large soybean areas is not always achievable in a practical way. Multispectral reflectance sensors have not been thoroughly evaluated to detect nematode damage in soybeans (Glycine max L.). The main research aims of this study were to: (i) determine the bivariate relationship between individual spectral bands and vegetation indices (VIs) relative to soybean conditions (symptomatic versus asymptomatic), and (ii) to select the best model for identifying plant conditions using three algorithms (logistic regression—LR, random forest—RF, conditional inference tree—CIT) and three options for data input using bands, vegetation indices (VIs), and bands plus VIs. The trial was conducted in Brazil on three on-farm soybean fields presenting different species of nematode infestation. Multispectral imagery was obtained using a drone-mounted MicaSense RedEdge® sensor. At each sampling, georeferenced point nematode infestation and spectral measurements of soybean plants were retrieved for the classification of symptomatic and asymptomatic areas, according to the threshold level adopted. Bivariate analysis of variance (ANOVA), LR, RF, and CIT were used to select the multispectral bands/VIs that discriminated among symptomatic and asymptomatic plants, assessing the best model via their respective parameters for accuracy, sensitivity, and specificity. The greatest classification accuracy (>0.70) was achieved when using the CIT algorithm with the spectral bands only, with green (560 ± 20 nm) and near-infrared (840 ± 40 nm) included as the main spectral input variables in the model. These results demonstrate the potential of combining remotely sensed data and machine learning to distinguish nematode-symptomatic and asymptomatic soybean plants.Department of Engineering and Mathematical Sciences São Paulo State University ‘Júlio de Mesquita Filho’ (UNESP) School of Agricultural and Veterinarian Sciences, Via de Acesso Prof. Paulo Donato Castellane, SPDepartment of Agronomy Kansas State University, 1712 Claflin RoadDepartment of Crop and Soil Sciences University of Georgia Miller Plant SciencesDepartment of Crop Soil and Environmental Sciences Auburn University, 350 S College StDepartment of Agricultural Production Sciences São Paulo State University ‘Júlio de Mesquita Filho’ (UNESP) School of Agricultural and Veterinarian Sciences, Via de Acesso Prof. Paulo Donato Castellane, SPDepartment of Engineering and Mathematical Sciences São Paulo State University ‘Júlio de Mesquita Filho’ (UNESP) School of Agricultural and Veterinarian Sciences, Via de Acesso Prof. Paulo Donato Castellane, SPDepartment of Agricultural Production Sciences São Paulo State University ‘Júlio de Mesquita Filho’ (UNESP) School of Agricultural and Veterinarian Sciences, Via de Acesso Prof. Paulo Donato Castellane, SPUniversidade Estadual Paulista (UNESP)Kansas State UniversityMiller Plant SciencesAuburn UniversitySantos, Letícia Bernabé [UNESP]Bastos, Leonardo Mendesde Oliveira, Mailson Freire [UNESP]Soares, Pedro Luiz Martins [UNESP]Ciampitti, Ignacio Antonioda Silva, Rouverson Pereira [UNESP]2023-07-29T15:12:13Z2023-07-29T15:12:13Z2022-10-01info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articlehttp://dx.doi.org/10.3390/agronomy12102404Agronomy, v. 12, n. 10, 2022.2073-4395http://hdl.handle.net/11449/24929610.3390/agronomy121024042-s2.0-85140436297Scopusreponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPengAgronomyinfo:eu-repo/semantics/openAccess2024-06-06T15:18:03Zoai:repositorio.unesp.br:11449/249296Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462024-08-05T15:39:53.246395Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false |
dc.title.none.fl_str_mv |
Identifying Nematode Damage on Soybean through Remote Sensing and Machine Learning Techniques |
title |
Identifying Nematode Damage on Soybean through Remote Sensing and Machine Learning Techniques |
spellingShingle |
Identifying Nematode Damage on Soybean through Remote Sensing and Machine Learning Techniques Santos, Letícia Bernabé [UNESP] digital agriculture disease detection machine learning multispectral mapping nematodes remote sensing |
title_short |
Identifying Nematode Damage on Soybean through Remote Sensing and Machine Learning Techniques |
title_full |
Identifying Nematode Damage on Soybean through Remote Sensing and Machine Learning Techniques |
title_fullStr |
Identifying Nematode Damage on Soybean through Remote Sensing and Machine Learning Techniques |
title_full_unstemmed |
Identifying Nematode Damage on Soybean through Remote Sensing and Machine Learning Techniques |
title_sort |
Identifying Nematode Damage on Soybean through Remote Sensing and Machine Learning Techniques |
author |
Santos, Letícia Bernabé [UNESP] |
author_facet |
Santos, Letícia Bernabé [UNESP] Bastos, Leonardo Mendes de Oliveira, Mailson Freire [UNESP] Soares, Pedro Luiz Martins [UNESP] Ciampitti, Ignacio Antonio da Silva, Rouverson Pereira [UNESP] |
author_role |
author |
author2 |
Bastos, Leonardo Mendes de Oliveira, Mailson Freire [UNESP] Soares, Pedro Luiz Martins [UNESP] Ciampitti, Ignacio Antonio da Silva, Rouverson Pereira [UNESP] |
author2_role |
author author author author author |
dc.contributor.none.fl_str_mv |
Universidade Estadual Paulista (UNESP) Kansas State University Miller Plant Sciences Auburn University |
dc.contributor.author.fl_str_mv |
Santos, Letícia Bernabé [UNESP] Bastos, Leonardo Mendes de Oliveira, Mailson Freire [UNESP] Soares, Pedro Luiz Martins [UNESP] Ciampitti, Ignacio Antonio da Silva, Rouverson Pereira [UNESP] |
dc.subject.por.fl_str_mv |
digital agriculture disease detection machine learning multispectral mapping nematodes remote sensing |
topic |
digital agriculture disease detection machine learning multispectral mapping nematodes remote sensing |
description |
Identifying nematode damage in large soybean areas is not always achievable in a practical way. Multispectral reflectance sensors have not been thoroughly evaluated to detect nematode damage in soybeans (Glycine max L.). The main research aims of this study were to: (i) determine the bivariate relationship between individual spectral bands and vegetation indices (VIs) relative to soybean conditions (symptomatic versus asymptomatic), and (ii) to select the best model for identifying plant conditions using three algorithms (logistic regression—LR, random forest—RF, conditional inference tree—CIT) and three options for data input using bands, vegetation indices (VIs), and bands plus VIs. The trial was conducted in Brazil on three on-farm soybean fields presenting different species of nematode infestation. Multispectral imagery was obtained using a drone-mounted MicaSense RedEdge® sensor. At each sampling, georeferenced point nematode infestation and spectral measurements of soybean plants were retrieved for the classification of symptomatic and asymptomatic areas, according to the threshold level adopted. Bivariate analysis of variance (ANOVA), LR, RF, and CIT were used to select the multispectral bands/VIs that discriminated among symptomatic and asymptomatic plants, assessing the best model via their respective parameters for accuracy, sensitivity, and specificity. The greatest classification accuracy (>0.70) was achieved when using the CIT algorithm with the spectral bands only, with green (560 ± 20 nm) and near-infrared (840 ± 40 nm) included as the main spectral input variables in the model. These results demonstrate the potential of combining remotely sensed data and machine learning to distinguish nematode-symptomatic and asymptomatic soybean plants. |
publishDate |
2022 |
dc.date.none.fl_str_mv |
2022-10-01 2023-07-29T15:12:13Z 2023-07-29T15:12:13Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://dx.doi.org/10.3390/agronomy12102404 Agronomy, v. 12, n. 10, 2022. 2073-4395 http://hdl.handle.net/11449/249296 10.3390/agronomy12102404 2-s2.0-85140436297 |
url |
http://dx.doi.org/10.3390/agronomy12102404 http://hdl.handle.net/11449/249296 |
identifier_str_mv |
Agronomy, v. 12, n. 10, 2022. 2073-4395 10.3390/agronomy12102404 2-s2.0-85140436297 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
Agronomy |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.source.none.fl_str_mv |
Scopus reponame:Repositório Institucional da UNESP instname:Universidade Estadual Paulista (UNESP) instacron:UNESP |
instname_str |
Universidade Estadual Paulista (UNESP) |
instacron_str |
UNESP |
institution |
UNESP |
reponame_str |
Repositório Institucional da UNESP |
collection |
Repositório Institucional da UNESP |
repository.name.fl_str_mv |
Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP) |
repository.mail.fl_str_mv |
|
_version_ |
1808128547566387200 |