Apple (Malus domestica) and pear (Pyrus communis) yield prediction after tree image analysis
Yield forecasting depends on accurate tree fruit counts and mean size estimation. This information is generally obtained manually, requiring many hours of work. Artificial vision emerges as an interesting alternative to obtaining more information in less time. This study aimed to test and train YOLO pre-trained models based on neural networks for the detection and count of pears and apples on trees after image analysis; while also estimating fruit size. Images of trees were taken during the day and at night in apple and pear trees while fruits were manually counted. Trained models were evaluated according to recall, precision and F1score. The correlation between detected and counted fruits was calculated while fruit size estimation was made after drawing straight lines on each fruit and using reference elements. The precision, recall and F1score achieved by the models were up to 0.86, 0.83 and 0.84, respectively. Correlation coefficients between fruit sizes measured manually and by images were 0.73 for apples and 0.80 for pears. The proposed methodologies showed promising results, allowing forecasters to make less time-consuming and accurate estimates compared to manual measurements.
Main Authors: | , , , , |
---|---|
Format: | info:ar-repo/semantics/artículo biblioteca |
Language: | eng |
Published: |
Facultad de Ciencias Agrarias, Universidad Nacional de Cuyo
2023-12
|
Subjects: | Malus Pumila, Pyrus Communis, Manzana, Pera, Apples, Pears, Fruit Detection, Artificial Vision, Yield Forecast, Detección de Frutos, Visión Artificial, Predicción de Cosecha, Malus Domestica, |
Online Access: | http://hdl.handle.net/20.500.12123/16325 https://revistas.uncu.edu.ar/ojs3/index.php/RFCA/article/view/6452 https://doi.org/10.48162/rev.39.104 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
oai:localhost:20.500.12123-16325 |
---|---|
record_format |
koha |
institution |
INTA AR |
collection |
DSpace |
country |
Argentina |
countrycode |
AR |
component |
Bibliográfico |
access |
En linea |
databasecode |
dig-inta-ar |
tag |
biblioteca |
region |
America del Sur |
libraryname |
Biblioteca Central del INTA Argentina |
language |
eng |
topic |
Malus Pumila Pyrus Communis Manzana Pera Apples Pears Fruit Detection Artificial Vision Yield Forecast Detección de Frutos Visión Artificial Predicción de Cosecha Malus Domestica Malus Pumila Pyrus Communis Manzana Pera Apples Pears Fruit Detection Artificial Vision Yield Forecast Detección de Frutos Visión Artificial Predicción de Cosecha Malus Domestica |
spellingShingle |
Malus Pumila Pyrus Communis Manzana Pera Apples Pears Fruit Detection Artificial Vision Yield Forecast Detección de Frutos Visión Artificial Predicción de Cosecha Malus Domestica Malus Pumila Pyrus Communis Manzana Pera Apples Pears Fruit Detection Artificial Vision Yield Forecast Detección de Frutos Visión Artificial Predicción de Cosecha Malus Domestica Del Brio, Dolores Tassile, Valentin Bramardi, Sergio Jorge Fernandez, Dario Eduardo Reeb, Pablo Daniel Apple (Malus domestica) and pear (Pyrus communis) yield prediction after tree image analysis |
description |
Yield forecasting depends on accurate tree fruit counts and mean size estimation. This information is generally obtained manually, requiring many hours of work. Artificial vision emerges as an interesting alternative to obtaining more information in less time. This study aimed to test and train YOLO pre-trained models based on neural networks for the detection and count of pears and apples on trees after image analysis; while also estimating fruit size. Images of trees were taken during the day and at night in apple and pear trees while fruits were manually counted. Trained models were evaluated according to recall, precision and F1score. The correlation between detected and counted fruits was calculated while fruit size estimation was made after drawing straight lines on each fruit and using reference elements. The precision, recall and F1score achieved by the models were up to 0.86, 0.83 and 0.84, respectively. Correlation coefficients between fruit sizes measured manually and by images were 0.73 for apples and 0.80 for pears. The proposed methodologies showed promising results, allowing forecasters to make less time-consuming and accurate estimates compared to manual measurements. |
format |
info:ar-repo/semantics/artículo |
topic_facet |
Malus Pumila Pyrus Communis Manzana Pera Apples Pears Fruit Detection Artificial Vision Yield Forecast Detección de Frutos Visión Artificial Predicción de Cosecha Malus Domestica |
author |
Del Brio, Dolores Tassile, Valentin Bramardi, Sergio Jorge Fernandez, Dario Eduardo Reeb, Pablo Daniel |
author_facet |
Del Brio, Dolores Tassile, Valentin Bramardi, Sergio Jorge Fernandez, Dario Eduardo Reeb, Pablo Daniel |
author_sort |
Del Brio, Dolores |
title |
Apple (Malus domestica) and pear (Pyrus communis) yield prediction after tree image analysis |
title_short |
Apple (Malus domestica) and pear (Pyrus communis) yield prediction after tree image analysis |
title_full |
Apple (Malus domestica) and pear (Pyrus communis) yield prediction after tree image analysis |
title_fullStr |
Apple (Malus domestica) and pear (Pyrus communis) yield prediction after tree image analysis |
title_full_unstemmed |
Apple (Malus domestica) and pear (Pyrus communis) yield prediction after tree image analysis |
title_sort |
apple (malus domestica) and pear (pyrus communis) yield prediction after tree image analysis |
publisher |
Facultad de Ciencias Agrarias, Universidad Nacional de Cuyo |
publishDate |
2023-12 |
url |
http://hdl.handle.net/20.500.12123/16325 https://revistas.uncu.edu.ar/ojs3/index.php/RFCA/article/view/6452 https://doi.org/10.48162/rev.39.104 |
work_keys_str_mv |
AT delbriodolores applemalusdomesticaandpearpyruscommunisyieldpredictionaftertreeimageanalysis AT tassilevalentin applemalusdomesticaandpearpyruscommunisyieldpredictionaftertreeimageanalysis AT bramardisergiojorge applemalusdomesticaandpearpyruscommunisyieldpredictionaftertreeimageanalysis AT fernandezdarioeduardo applemalusdomesticaandpearpyruscommunisyieldpredictionaftertreeimageanalysis AT reebpablodaniel applemalusdomesticaandpearpyruscommunisyieldpredictionaftertreeimageanalysis |
_version_ |
1798158382023049216 |
spelling |
oai:localhost:20.500.12123-163252023-12-22T12:36:13Z Apple (Malus domestica) and pear (Pyrus communis) yield prediction after tree image analysis Del Brio, Dolores Tassile, Valentin Bramardi, Sergio Jorge Fernandez, Dario Eduardo Reeb, Pablo Daniel Malus Pumila Pyrus Communis Manzana Pera Apples Pears Fruit Detection Artificial Vision Yield Forecast Detección de Frutos Visión Artificial Predicción de Cosecha Malus Domestica Yield forecasting depends on accurate tree fruit counts and mean size estimation. This information is generally obtained manually, requiring many hours of work. Artificial vision emerges as an interesting alternative to obtaining more information in less time. This study aimed to test and train YOLO pre-trained models based on neural networks for the detection and count of pears and apples on trees after image analysis; while also estimating fruit size. Images of trees were taken during the day and at night in apple and pear trees while fruits were manually counted. Trained models were evaluated according to recall, precision and F1score. The correlation between detected and counted fruits was calculated while fruit size estimation was made after drawing straight lines on each fruit and using reference elements. The precision, recall and F1score achieved by the models were up to 0.86, 0.83 and 0.84, respectively. Correlation coefficients between fruit sizes measured manually and by images were 0.73 for apples and 0.80 for pears. The proposed methodologies showed promising results, allowing forecasters to make less time-consuming and accurate estimates compared to manual measurements. Para pronosticar la producción es necesario contar el número de frutos de los árboles y estimar el tamaño medio. Esta información se obtiene manualmente y requiere mucha mano de obra experimentada. La visión artificial surge como alternativa para obtener más información en menos tiempo. Los objetivos del trabajo fueron entrenar modelos de visión artificial para detectar y contar el número de peras y manzanas en árboles a partir de imágenes; y medir diámetros de frutos en imágenes. Se usaron modelos pre-entrenados para detección de objetos basados en redes neuronales (YOLO). Se tomaron imágenes de árboles de día y de noche, y los frutos de cada planta fueron contados manualmente. Los modelos se evaluaron según sensibilidad, precisión y F1score; y se calculó la correlación entre frutos detectados y contados. La estimación de diámetros se realizó trazando líneas rectas sobre cada fruto y utilizando elementos de referencia. La precisión, sensibilidad y F1score alcanzados por los modelos fueron 0,86, 0,83 y 0,84, respectivamente. Las correla-ciones entre diámetros medidos manualmente y por imágenes fueron de 0,73 en manzanas y 0,80 en peras. Las metodologías propuestas permitieron realizar estimaciones a partir de imágenes con una precisión aceptable y en menor tiempo respecto de las mediciones manuales. EEA Alto Valle Fil: Del Brío, Dolores. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina. Instituto Nacional de Tecnología Agropecuaria (INTA). Estación Experimental Agropecuaria Alto Valle; Argentina Fil: Tassile, Valentín. Universidad Nacional del Comahue. Facultad de Ciencias y Tecnología de los Alimentos; Argentina Fil: Bramardi, Sergio Jorge. Universidad Nacional del Comahue. Departamento de Estadística; Argentina Fil: Fernandez, Darío Eduardo. Instituto Nacional de Tecnología Agropecuaria (INTA). Estación Experimental Agropecuaria Alto Valle; Argentina Fil: Reeb, Pablo Daniel. Universidad Nacional del Comahue. Departamento de Estadística; Argentina 2023-12-22T12:19:57Z 2023-12-22T12:19:57Z 2023-12 info:ar-repo/semantics/artículo info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion http://hdl.handle.net/20.500.12123/16325 https://revistas.uncu.edu.ar/ojs3/index.php/RFCA/article/view/6452 1853-8665 https://doi.org/10.48162/rev.39.104 eng info:eu-repo/semantics/openAccess http://creativecommons.org/licenses/by-nc-sa/4.0/ Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) application/pdf Facultad de Ciencias Agrarias, Universidad Nacional de Cuyo Revista de la Facultad de Ciencias Agrarias / Universidad Nacional de Cuyo 55 (2) : 1-11 (2023) |