About identification of features that affect the estimation of citrus harvest

Accurate models for early harvest estimation in citrus production generally involve expensive variables. The goal of this research work was to develop a model to provide early and accurate estimations of harvest using low-cost features. Given the original data may derive from tree measurements, met...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Bóbeda, Griselda R. R., Mazza, Silvia M., Rico , Noelia, Brenes Pérez, Cristian F., Gaiad, José E., Díaz Rodríguez, Susana Irene
Formato: Online
Lenguaje:eng
Publicado: Facultad de Ciencias Agrarias-UNCuyo 2023
Materias:
SVM
Acceso en línea:https://revistas.uncu.edu.ar/ojs3/index.php/RFCA/article/view/5452
id I11-R107article-5452
record_format ojs
spelling I11-R107article-54522023-09-14T18:41:42Z About identification of features that affect the estimation of citrus harvest About identification of features that affect the estimation of citrus harvest Bóbeda, Griselda R. R. Mazza, Silvia M. Rico , Noelia Brenes Pérez, Cristian F. Gaiad, José E. Díaz Rodríguez, Susana Irene MODIS SVM selección de variables aprendizaje automático naranja dulce tangor Murcott MODIS SVM selection of variables machine learning sweet orange Murcott tangor Accurate models for early harvest estimation in citrus production generally involve expensive variables. The goal of this research work was to develop a model to provide early and accurate estimations of harvest using low-cost features. Given the original data may derive from tree measurements, meteorological stations, or satellites, they have varied costs. The studied orchards included tangerines (Citrus reticulata x C. sinensis) and sweet oranges (C. sinensis) located in northeastern Argentina. Machine learning methods combined with different datasets were tested to obtain the most accurate harvest estimation. The final model is based on support vector machines with low-cost variables like species, age, irrigation, red and near-infrared reflectance in February and December, NDVI in December, rain during ripening, and humidity during fruit growth. Highlights: Red and near-infrared reflectance in February and December are helpful values to predict orange harvest. SVM is an efficient method to predict harvest. A ranking method to A ranking-based method has been developed to identify the variables that best predict orange production.   Accurate models for early harvest estimation in citrus production generally involve expensive variables. The goal of this research work was to develop a model to provide early and accurate estimations of harvest using low-cost features. Given the original data may derive from tree measurements, meteorological stations, or satellites, they have varied costs. The studied orchards included tangerines (Citrus reticulata x C. sinensis) and sweet oranges (C. sinensis) located in northeastern Argentina. Machine learning methods combined with different datasets were tested to obtain the most accurate harvest estimation. The final model is based on support vector machines with low-cost variables like species, age, irrigation, red and near-infrared reflectance in February and December, NDVI in December, rain during ripening, and humidity during fruit growth. Highlights: Red and near-infrared reflectance in February and December are helpful values to predict orange harvest. SVM is an efficient method to predict harvest. A ranking method to A ranking-based method has been developed to identify the variables that best predict orange production. Facultad de Ciencias Agrarias-UNCuyo 2023-06-27 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion application/pdf https://revistas.uncu.edu.ar/ojs3/index.php/RFCA/article/view/5452 10.48162/rev.39.096 Revista de la Facultad de Ciencias Agrarias UNCuyo; Vol. 55 No. 1 (2023): January-June; 65-74 Revista de la Facultad de Ciencias Agrarias UNCuyo; Vol. 55 Núm. 1 (2023): Enero-Junio; 65-74 1853-8665 0370-4661 eng https://revistas.uncu.edu.ar/ojs3/index.php/RFCA/article/view/5452/5568 Derechos de autor 2018 Revista de la Facultad de Ciencias Agrarias UNCuyo https://creativecommons.org/licenses/by-nc-sa/3.0/deed.es
institution Universidad Nacional de Cuyo
building Revistas en línea
filtrotop_str Revistas en línea
collection Revista de la Facultad de Ciencias Agrarias
journal_title_str Revista de la Facultad de Ciencias Agrarias
institution_str I-11
repository_str R-107
language eng
format Online
author Bóbeda, Griselda R. R.
Mazza, Silvia M.
Rico , Noelia
Brenes Pérez, Cristian F.
Gaiad, José E.
Díaz Rodríguez, Susana Irene
spellingShingle Bóbeda, Griselda R. R.
Mazza, Silvia M.
Rico , Noelia
Brenes Pérez, Cristian F.
Gaiad, José E.
Díaz Rodríguez, Susana Irene
About identification of features that affect the estimation of citrus harvest
MODIS
SVM
selección de variables
aprendizaje automático
naranja dulce
tangor Murcott
MODIS
SVM
selection of variables
machine learning
sweet orange
Murcott tangor
author_facet Bóbeda, Griselda R. R.
Mazza, Silvia M.
Rico , Noelia
Brenes Pérez, Cristian F.
Gaiad, José E.
Díaz Rodríguez, Susana Irene
author_sort Bóbeda, Griselda R. R.
title About identification of features that affect the estimation of citrus harvest
title_short About identification of features that affect the estimation of citrus harvest
title_full About identification of features that affect the estimation of citrus harvest
title_fullStr About identification of features that affect the estimation of citrus harvest
title_full_unstemmed About identification of features that affect the estimation of citrus harvest
title_sort about identification of features that affect the estimation of citrus harvest
description Accurate models for early harvest estimation in citrus production generally involve expensive variables. The goal of this research work was to develop a model to provide early and accurate estimations of harvest using low-cost features. Given the original data may derive from tree measurements, meteorological stations, or satellites, they have varied costs. The studied orchards included tangerines (Citrus reticulata x C. sinensis) and sweet oranges (C. sinensis) located in northeastern Argentina. Machine learning methods combined with different datasets were tested to obtain the most accurate harvest estimation. The final model is based on support vector machines with low-cost variables like species, age, irrigation, red and near-infrared reflectance in February and December, NDVI in December, rain during ripening, and humidity during fruit growth. Highlights: Red and near-infrared reflectance in February and December are helpful values to predict orange harvest. SVM is an efficient method to predict harvest. A ranking method to A ranking-based method has been developed to identify the variables that best predict orange production.  
publisher Facultad de Ciencias Agrarias-UNCuyo
publishDate 2023
url https://revistas.uncu.edu.ar/ojs3/index.php/RFCA/article/view/5452
topic MODIS
SVM
selección de variables
aprendizaje automático
naranja dulce
tangor Murcott
MODIS
SVM
selection of variables
machine learning
sweet orange
Murcott tangor
topic_facet MODIS
SVM
selección de variables
aprendizaje automático
naranja dulce
tangor Murcott
MODIS
SVM
selection of variables
machine learning
sweet orange
Murcott tangor
work_keys_str_mv AT bobedagriseldarr aboutidentificationoffeaturesthataffecttheestimationofcitrusharvest
AT mazzasilviam aboutidentificationoffeaturesthataffecttheestimationofcitrusharvest
AT riconoelia aboutidentificationoffeaturesthataffecttheestimationofcitrusharvest
AT brenesperezcristianf aboutidentificationoffeaturesthataffecttheestimationofcitrusharvest
AT gaiadjosee aboutidentificationoffeaturesthataffecttheestimationofcitrusharvest
AT diazrodriguezsusanairene aboutidentificationoffeaturesthataffecttheestimationofcitrusharvest
_version_ 1800220934433406976