Almond cultivar identification using machine learning classifiers applied to UAV-based multispectral data

Guimaraes, Nathalie; Padua, Luis; Sousa, Joaquim J.; Bento, Albino; Couto, Pedro

Publication

Almond cultivar identification using machine learning classifiers applied to UAV-based multispectral data

2023Journal article

dc.contributor.author	Guimaraes, Nathalie
dc.contributor.author	Padua, Luis
dc.contributor.author	Sousa, Joaquim J.
dc.contributor.author	Bento, Albino
dc.contributor.author	Couto, Pedro
dc.date.accessioned	2023-01-03T15:10:36Z
dc.date.available	2023-01-03T15:10:36Z
dc.date.issued	2023
dc.description.abstract	In Portugal, almonds are a very important crop, due to their nutritional properties. In the northeastern part of the country, the almond sector has endured over time, with strong cultural traditions and key economic significance. In these areas, several cultivars are used. In effect, the presence of various almond cultivars implies differentiated management in irrigation, disease control, pruning system, and harvest planning. Therefore, cultivar classification is essential over large agricultural areas. Over the last decades, remote-sensing data have led to important breakthroughs in the classification of different cultivars for several crops. Nonetheless, for almonds, studies are incipient. Thus, this study aims to fill this knowledge gap and explore the classification of almond cultivars in an almond orchard. High-resolution multispectral data were acquired by an unmanned aerial vehicle (UAV). Vegetation indices (VIs) and tree structural parameters were, subsequently, estimated. To obtain an accurate cultivar identification, four machine learning classifiers, such as K-nearest neighbour (kNN), support vector machine (SVM), random forest (RF), and extreme gradient boosting (XGBoost), were applied and optimized through the fine-tuning process. The accuracy of machine learning classifiers was analysed. SVM and RF performed best with OAs of 76% and 74% using VIs and spectral bands (GREEN, GRVI, GN, REN, ClRE). Adding the canopy height model (CHM) improved performance, with RF and XGBoost having OAs of 88% and 84%. kNN performed worst with an OA of 73% using only VIs and spectral bands, 80% with VIs, spectral bands and CHM, and 93% with VIs, CHM, and tree crown area (TCA). The best performance was achieved by RF and XGBoost with OAs of 99% using VIs, CHM, and TCA. These results demonstrate the importance of the feature selection process. Moreover, this study reveals the feasibility of remote-sensing data and machine learning classifiers in the classification of almond cultivars.	pt_PT
dc.description.version	info:eu-repo/semantics/publishedVersion	pt_PT
dc.identifier.citation	Guimaraes, Nathalie; Padua, Luis; Sousa, Joaquim J.; Bento, Albino; Couto, Pedro (2023). Almond cultivar identification using machine learning classifiers applied to UAV-based multispectral data. International Journal of Remote Sensing. eISSN 1366-5901. 44:5, p. 1533-1555	pt_PT
dc.identifier.doi	10.1080/01431161.2023.2185913
dc.identifier.eissn	1366-5901
dc.identifier.issn	0143-1161
dc.identifier.uri	http://hdl.handle.net/10198/26242
dc.language.iso	eng	pt_PT
dc.publisher	Taylor & Francis
dc.rights.uri	http://creativecommons.org/licenses/by/4.0/	pt_PT
dc.subject	Unmanned aerial vehicles
dc.subject	Precision agriculture
dc.subject	Prunus dulcis
dc.subject	Tree cultivars classification
dc.subject	K-nearest neighbour
dc.subject	Support vector machine
dc.subject	Random forest
dc.subject	Extreme gradient boosting
dc.title	Almond cultivar identification using machine learning classifiers applied to UAV-based multispectral data	pt_PT
dc.type	journal article
dspace.entity.type	Publication
oaire.citation.title	International Journal of Remote Sensing	pt_PT
person.familyName	Bento
person.givenName	Albino
person.identifier.ciencia-id	D516-325A-9AD7
person.identifier.orcid	0000-0001-5215-785X
person.identifier.rid	N-9706-2016
person.identifier.scopus-author-id	35247694000
rcaap.rights	openAccess	pt_PT
rcaap.type	article	pt_PT
relation.isAuthorOfPublication	233115be-9d46-49d0-8b7d-2d64406d64a0
relation.isAuthorOfPublication.latestForDiscovery	233115be-9d46-49d0-8b7d-2d64406d64a0