Woody Plant Encroachment in a Seasonal Tropical Savanna: Lessons about Classifiers and Accuracy from UAV Images
Woody plant encroachment in grassy ecosystems is a widely reported phenomenon associated with negative impacts on ecosystem functions. Most studies of this phenomenon have been carried out in arid and semi-arid grasslands. Therefore, studies in tropical regions, particularly savannas, which are comp...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-04-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/9/2342 |
| _version_ | 1797601776856203264 |
|---|---|
| author | Lucas Silva Costa Edson Eyji Sano Manuel Eduardo Ferreira Cássia Beatriz Rodrigues Munhoz João Vítor Silva Costa Leomar Rufino Alves Júnior Thiago Roure Bandeira de Mello Mercedes Maria da Cunha Bustamante |
| author_facet | Lucas Silva Costa Edson Eyji Sano Manuel Eduardo Ferreira Cássia Beatriz Rodrigues Munhoz João Vítor Silva Costa Leomar Rufino Alves Júnior Thiago Roure Bandeira de Mello Mercedes Maria da Cunha Bustamante |
| author_sort | Lucas Silva Costa |
| collection | DOAJ |
| description | Woody plant encroachment in grassy ecosystems is a widely reported phenomenon associated with negative impacts on ecosystem functions. Most studies of this phenomenon have been carried out in arid and semi-arid grasslands. Therefore, studies in tropical regions, particularly savannas, which are composed of grassland and woodland mosaics, are needed. Our objective was to evaluate the accuracy of woody encroachment classification in the Brazilian Cerrado, a tropical savanna. We acquired dry and wet season unmanned aerial vehicle (UAV) images using RGB and multispectral cameras that were processed by the support vector machine (SVM), decision tree (DT), and random forest (RF) classifiers. We also compared two validation methods: the orthomosaic and in situ methods. We targeted two native woody species: <i>Baccharis retusa</i> and <i>Trembleya parviflora</i>. Identification of these two species was statistically (<i>p</i> < 0.05) most accurate in the wet season RGB images classified by the RF algorithm, with an overall accuracy (OA) of 92.7%. Relating to validation assessments, the in situ method was more susceptible to underfitting scenarios, especially using an RF classifier. The OA was higher in grassland than in woodland formations. Our results show that woody encroachment classification in a tropical savanna is possible using UAV images and field surveys and is suggested to be conducted during the wet season. It is challenging to classify UAV images in highly diverse ecosystems such as the Cerrado; therefore, whenever possible, researchers should use multiple accuracy assessment methods. In the case of using in situ accuracy assessment, we suggest a minimum of 40 training samples per class and to use multiple classifiers (e.g., RF and DT). Our findings contribute to the generation of tools that optimize time and cost for the monitoring and management of woody encroachment in tropical savannas. |
| first_indexed | 2024-03-11T04:08:34Z |
| format | Article |
| id | doaj.art-f3293933ebcd4d41bb614efceded7cee |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-11T04:08:34Z |
| publishDate | 2023-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-f3293933ebcd4d41bb614efceded7cee2023-11-17T23:38:48ZengMDPI AGRemote Sensing2072-42922023-04-01159234210.3390/rs15092342Woody Plant Encroachment in a Seasonal Tropical Savanna: Lessons about Classifiers and Accuracy from UAV ImagesLucas Silva Costa0Edson Eyji Sano1Manuel Eduardo Ferreira2Cássia Beatriz Rodrigues Munhoz3João Vítor Silva Costa4Leomar Rufino Alves Júnior5Thiago Roure Bandeira de Mello6Mercedes Maria da Cunha Bustamante7Programa de Pós-Graduação em Ecologia, Instituto de Ciências Biológicas, Universidade de Brasília, Brasília 70910-900, BrazilEmpresa Brasileira de Pesquisa Agropecuária (Embrapa Cerrados), BR-020, Planaltina 73301-970, BrazilInstituto de Estudos Socioambientais, Universidade Federal do Goiás (UFG), Goiânia 74690-900, BrazilPrograma de Pós-Graduação em Ecologia, Instituto de Ciências Biológicas, Universidade de Brasília, Brasília 70910-900, BrazilInstituto de Estudos Socioambientais, Universidade Federal do Goiás (UFG), Goiânia 74690-900, BrazilInstituto de Estudos Socioambientais, Universidade Federal do Goiás (UFG), Goiânia 74690-900, BrazilPrograma de Pós-Graduação em Ecologia, Instituto de Ciências Biológicas, Universidade de Brasília, Brasília 70910-900, BrazilPrograma de Pós-Graduação em Ecologia, Instituto de Ciências Biológicas, Universidade de Brasília, Brasília 70910-900, BrazilWoody plant encroachment in grassy ecosystems is a widely reported phenomenon associated with negative impacts on ecosystem functions. Most studies of this phenomenon have been carried out in arid and semi-arid grasslands. Therefore, studies in tropical regions, particularly savannas, which are composed of grassland and woodland mosaics, are needed. Our objective was to evaluate the accuracy of woody encroachment classification in the Brazilian Cerrado, a tropical savanna. We acquired dry and wet season unmanned aerial vehicle (UAV) images using RGB and multispectral cameras that were processed by the support vector machine (SVM), decision tree (DT), and random forest (RF) classifiers. We also compared two validation methods: the orthomosaic and in situ methods. We targeted two native woody species: <i>Baccharis retusa</i> and <i>Trembleya parviflora</i>. Identification of these two species was statistically (<i>p</i> < 0.05) most accurate in the wet season RGB images classified by the RF algorithm, with an overall accuracy (OA) of 92.7%. Relating to validation assessments, the in situ method was more susceptible to underfitting scenarios, especially using an RF classifier. The OA was higher in grassland than in woodland formations. Our results show that woody encroachment classification in a tropical savanna is possible using UAV images and field surveys and is suggested to be conducted during the wet season. It is challenging to classify UAV images in highly diverse ecosystems such as the Cerrado; therefore, whenever possible, researchers should use multiple accuracy assessment methods. In the case of using in situ accuracy assessment, we suggest a minimum of 40 training samples per class and to use multiple classifiers (e.g., RF and DT). Our findings contribute to the generation of tools that optimize time and cost for the monitoring and management of woody encroachment in tropical savannas.https://www.mdpi.com/2072-4292/15/9/2342Cerradoobject-based image analysismesic biomeplant invasiondronemultispectral |
| spellingShingle | Lucas Silva Costa Edson Eyji Sano Manuel Eduardo Ferreira Cássia Beatriz Rodrigues Munhoz João Vítor Silva Costa Leomar Rufino Alves Júnior Thiago Roure Bandeira de Mello Mercedes Maria da Cunha Bustamante Woody Plant Encroachment in a Seasonal Tropical Savanna: Lessons about Classifiers and Accuracy from UAV Images Remote Sensing Cerrado object-based image analysis mesic biome plant invasion drone multispectral |
| title | Woody Plant Encroachment in a Seasonal Tropical Savanna: Lessons about Classifiers and Accuracy from UAV Images |
| title_full | Woody Plant Encroachment in a Seasonal Tropical Savanna: Lessons about Classifiers and Accuracy from UAV Images |
| title_fullStr | Woody Plant Encroachment in a Seasonal Tropical Savanna: Lessons about Classifiers and Accuracy from UAV Images |
| title_full_unstemmed | Woody Plant Encroachment in a Seasonal Tropical Savanna: Lessons about Classifiers and Accuracy from UAV Images |
| title_short | Woody Plant Encroachment in a Seasonal Tropical Savanna: Lessons about Classifiers and Accuracy from UAV Images |
| title_sort | woody plant encroachment in a seasonal tropical savanna lessons about classifiers and accuracy from uav images |
| topic | Cerrado object-based image analysis mesic biome plant invasion drone multispectral |
| url | https://www.mdpi.com/2072-4292/15/9/2342 |
| work_keys_str_mv | AT lucassilvacosta woodyplantencroachmentinaseasonaltropicalsavannalessonsaboutclassifiersandaccuracyfromuavimages AT edsoneyjisano woodyplantencroachmentinaseasonaltropicalsavannalessonsaboutclassifiersandaccuracyfromuavimages AT manueleduardoferreira woodyplantencroachmentinaseasonaltropicalsavannalessonsaboutclassifiersandaccuracyfromuavimages AT cassiabeatrizrodriguesmunhoz woodyplantencroachmentinaseasonaltropicalsavannalessonsaboutclassifiersandaccuracyfromuavimages AT joaovitorsilvacosta woodyplantencroachmentinaseasonaltropicalsavannalessonsaboutclassifiersandaccuracyfromuavimages AT leomarrufinoalvesjunior woodyplantencroachmentinaseasonaltropicalsavannalessonsaboutclassifiersandaccuracyfromuavimages AT thiagorourebandeirademello woodyplantencroachmentinaseasonaltropicalsavannalessonsaboutclassifiersandaccuracyfromuavimages AT mercedesmariadacunhabustamante woodyplantencroachmentinaseasonaltropicalsavannalessonsaboutclassifiersandaccuracyfromuavimages |