Convolutional neural networks to assess bergamot essential oil content in the field from smartphone images

The essential oil (EO) extracted from bergamot peel (Citrus bergamia, Risso et Poiteau) is appreciated in perfumery and gastronomy. Notably, 90 % of the bergamot EO production is concentrated in the Province of Reggio Calabria (Southern Italy) under a protected designation of origin (PDO). The early...

Full description

Bibliographic Details
Main Authors: Anello, Matteo, Mateo, Fernando, Bernardi, Bruno, Giuffrè, Angelo Maria, Blasco, José, Gómez-Sanchis, Juan
Format: article
Language:Inglés
Published: Elsevier 2024
Subjects:
Online Access:https://hdl.handle.net/20.500.11939/8960
https://www.sciencedirect.com/science/article/pii/S092666902401210X?via%3Dihub
_version_ 1855032886038626304
author Anello, Matteo
Mateo, Fernando
Bernardi, Bruno
Giuffrè, Angelo Maria
Blasco, José
Gómez-Sanchis, Juan
author_browse Anello, Matteo
Bernardi, Bruno
Blasco, José
Giuffrè, Angelo Maria
Gómez-Sanchis, Juan
Mateo, Fernando
author_facet Anello, Matteo
Mateo, Fernando
Bernardi, Bruno
Giuffrè, Angelo Maria
Blasco, José
Gómez-Sanchis, Juan
author_sort Anello, Matteo
collection ReDivia
description The essential oil (EO) extracted from bergamot peel (Citrus bergamia, Risso et Poiteau) is appreciated in perfumery and gastronomy. Notably, 90 % of the bergamot EO production is concentrated in the Province of Reggio Calabria (Southern Italy) under a protected designation of origin (PDO). The early estimation of EO content in fruits is fundamental to help farmers in their decision at harvesting period. The application of advanced modelling techniques based on artificial intelligence and digital device technology can contribute to this goal. This study proposes a method to estimate the EO content of fruits in the field using classification and regression models based on a deep learning approach in two cultivars: cv. “Fantastico” and cv. “Femminello”. The first step was to capture images of the fruit in the Red, Green, and Blue colours (RGB) using a mid-range smartphone camera and a portable inspection chamber designed and developed for this study. The acquisition of the images was carried out in the field. The fruits were collected and transported to the laboratory, where the EO was extracted using steam hydrodistillation. Custom-built convolutional neural networks (CNN) and three transfer learning models (VGG-16, VGG-19, and Xception architectures) were trained and applied for classification (among different discrete levels of oil content) and regression (to predict the EO content). The classification results showed an accuracy of 0.795 and 0.797 on the test samples of the two cultivars separately, while the best regression model achieved a minimum mean squared error of 0.12 and 0.04 for each cultivar, respectively. The results showed the effectiveness of the approach tested and how modelling each variety independently can lead to better performance for the CNNs tested.
format article
id ReDivia8960
institution Instituto Valenciano de Investigaciones Agrarias (IVIA)
language Inglés
publishDate 2024
publishDateRange 2024
publishDateSort 2024
publisher Elsevier
publisherStr Elsevier
record_format dspace
spelling ReDivia89602025-04-25T14:49:39Z Convolutional neural networks to assess bergamot essential oil content in the field from smartphone images Anello, Matteo Mateo, Fernando Bernardi, Bruno Giuffrè, Angelo Maria Blasco, José Gómez-Sanchis, Juan Cv. Fantastico Cv. Femminello Smartphone camera N01 Agricultural engineering Essential oils Deep learning The essential oil (EO) extracted from bergamot peel (Citrus bergamia, Risso et Poiteau) is appreciated in perfumery and gastronomy. Notably, 90 % of the bergamot EO production is concentrated in the Province of Reggio Calabria (Southern Italy) under a protected designation of origin (PDO). The early estimation of EO content in fruits is fundamental to help farmers in their decision at harvesting period. The application of advanced modelling techniques based on artificial intelligence and digital device technology can contribute to this goal. This study proposes a method to estimate the EO content of fruits in the field using classification and regression models based on a deep learning approach in two cultivars: cv. “Fantastico” and cv. “Femminello”. The first step was to capture images of the fruit in the Red, Green, and Blue colours (RGB) using a mid-range smartphone camera and a portable inspection chamber designed and developed for this study. The acquisition of the images was carried out in the field. The fruits were collected and transported to the laboratory, where the EO was extracted using steam hydrodistillation. Custom-built convolutional neural networks (CNN) and three transfer learning models (VGG-16, VGG-19, and Xception architectures) were trained and applied for classification (among different discrete levels of oil content) and regression (to predict the EO content). The classification results showed an accuracy of 0.795 and 0.797 on the test samples of the two cultivars separately, while the best regression model achieved a minimum mean squared error of 0.12 and 0.04 for each cultivar, respectively. The results showed the effectiveness of the approach tested and how modelling each variety independently can lead to better performance for the CNNs tested. 2024-08-29T07:40:03Z 2024-08-29T07:40:03Z 2024 article publishedVersion Anello, M., Mateo, F., Bernardi, B., Giuffre, A. M., Blasco, J., & Gómez-Sanchis, J. (2024). Convolutional neural networks to assess bergamot essential oil content in the field from smartphone images. Industrial Crops and Products, 220, 119233. 0926-6690 https://hdl.handle.net/20.500.11939/8960 10.1016/j.indcrop.2024.119233 https://www.sciencedirect.com/science/article/pii/S092666902401210X?via%3Dihub en This research was funded by the Italian Ministry of University and Research (MUR) in the framework of a doctorate grant funded within the National Operational Program "Research and Innovation 2014-2020, Innovative Doctorates with Industrial Characterisation". PhD course in Agricultural, Food and Forestry Sciences (SAAF) at the University Mediterranea of Reggio Calabria (XXXVI cycle). This research is co-funded by the projects AEI PID2019-107347RRC33 and GVA-PROMETEO CIPROM/2021/014. info:eu-repo/grantAgreement/GVA/PROMETEO 2021/014 Attribution-NonCommercial-NoDerivatives 4.0 Internacional http://creativecommons.org/licenses/by-nc-nd/4.0/ openAccess Elsevier electronico
spellingShingle Cv. Fantastico
Cv. Femminello
Smartphone camera
N01 Agricultural engineering
Essential oils
Deep learning
Anello, Matteo
Mateo, Fernando
Bernardi, Bruno
Giuffrè, Angelo Maria
Blasco, José
Gómez-Sanchis, Juan
Convolutional neural networks to assess bergamot essential oil content in the field from smartphone images
title Convolutional neural networks to assess bergamot essential oil content in the field from smartphone images
title_full Convolutional neural networks to assess bergamot essential oil content in the field from smartphone images
title_fullStr Convolutional neural networks to assess bergamot essential oil content in the field from smartphone images
title_full_unstemmed Convolutional neural networks to assess bergamot essential oil content in the field from smartphone images
title_short Convolutional neural networks to assess bergamot essential oil content in the field from smartphone images
title_sort convolutional neural networks to assess bergamot essential oil content in the field from smartphone images
topic Cv. Fantastico
Cv. Femminello
Smartphone camera
N01 Agricultural engineering
Essential oils
Deep learning
url https://hdl.handle.net/20.500.11939/8960
https://www.sciencedirect.com/science/article/pii/S092666902401210X?via%3Dihub
work_keys_str_mv AT anellomatteo convolutionalneuralnetworkstoassessbergamotessentialoilcontentinthefieldfromsmartphoneimages
AT mateofernando convolutionalneuralnetworkstoassessbergamotessentialoilcontentinthefieldfromsmartphoneimages
AT bernardibruno convolutionalneuralnetworkstoassessbergamotessentialoilcontentinthefieldfromsmartphoneimages
AT giuffreangelomaria convolutionalneuralnetworkstoassessbergamotessentialoilcontentinthefieldfromsmartphoneimages
AT blascojose convolutionalneuralnetworkstoassessbergamotessentialoilcontentinthefieldfromsmartphoneimages
AT gomezsanchisjuan convolutionalneuralnetworkstoassessbergamotessentialoilcontentinthefieldfromsmartphoneimages