Precision viticulture: Automatic selection of the regions of interest from moving wagon hyperspectral images of grapes for efficient SSC prediction
Precision viticulture is increasingly being applied to automate and optimize grape production in the vineyard. This paper describes the development of a method for automatic selection of regions of interest from hyperspectral images obtained of a row of vines and intended for prediction of soluble s...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-03-01
|
| Series: | Smart Agricultural Technology |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S277237552400039X |
| _version_ | 1797246722480537600 |
|---|---|
| author | Alessandro Benelli Chiara Cevoli Angelo Fabbri Søren Balling Engelsen Klavs Martin Sørensen |
| author_facet | Alessandro Benelli Chiara Cevoli Angelo Fabbri Søren Balling Engelsen Klavs Martin Sørensen |
| author_sort | Alessandro Benelli |
| collection | DOAJ |
| description | Precision viticulture is increasingly being applied to automate and optimize grape production in the vineyard. This paper describes the development of a method for automatic selection of regions of interest from hyperspectral images obtained of a row of vines and intended for prediction of soluble solids content. For this purpose, a dataset consisting of hyperspectral images of a row of ‘Sangiovese’ wine grapes was adopted. Hyperspectral images were acquired directly in the field by means of a hyperspectral imaging Vis/NIR system (400–1000 nm) mounted on a ground-based vehicle. The analyses were carried out on 17 different days, under clear or partly cloudy conditions, in the period between post-veraison and harvest. The vineyard row of Sangiovese vines was divided into 11 sections and a hyperspectral image for each section for each day of analysis was acquired. The regions of interest of the hyperspectral images, comprising the areas representing the grapes, were selected using a PLS-DA-based method. The best PLS-DA model provided excellent results, with sensitivity and specificity values of 0.991 and 0.996, respectively. The mean spectra of the selected regions of interest (ROI) were finally used to predict the soluble solids content (SSC) of the grapes by PLS regression to a primary reference analysis. The results of SSC predictions using the automatic selection of ROIs (R2CV = 0.74 and RMSECV = 0.86 °Brix) were on par with similar regression based on carefully manual selection of ROIs (R2CV = 0.73 and RMSECV = 0.87 °Brix). |
| first_indexed | 2024-04-24T19:47:19Z |
| format | Article |
| id | doaj.art-5a91524773774f9b807de12dcc93aa0c |
| institution | Directory Open Access Journal |
| issn | 2772-3755 |
| language | English |
| last_indexed | 2024-04-24T19:47:19Z |
| publishDate | 2024-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Smart Agricultural Technology |
| spelling | doaj.art-5a91524773774f9b807de12dcc93aa0c2024-03-25T04:18:21ZengElsevierSmart Agricultural Technology2772-37552024-03-017100434Precision viticulture: Automatic selection of the regions of interest from moving wagon hyperspectral images of grapes for efficient SSC predictionAlessandro Benelli0Chiara Cevoli1Angelo Fabbri2Søren Balling Engelsen3Klavs Martin Sørensen4Department for Innovation in Biological, Agro-Food and Forest Systems, University of Tuscia, Via San Camillo de Lellis snc, Viterbo 01100, Italy; Department of Agricultural and Food Sciences, Alma Mater Studiorum, University of Bologna, Piazza Goidanich 60, Cesena 47521, Italy; Corresponding author at: Via San Camillo de Lellis snc, Viterbo 01100, Italy.Department of Agricultural and Food Sciences, Alma Mater Studiorum, University of Bologna, Piazza Goidanich 60, Cesena 47521, ItalyDepartment of Agricultural and Food Sciences, Alma Mater Studiorum, University of Bologna, Piazza Goidanich 60, Cesena 47521, ItalyFood Analytics & Biotechnology, Department of Food Science, University of Copenhagen, Rolighedsvej 26, Frederiksberg 1958, DenmarkFood Analytics & Biotechnology, Department of Food Science, University of Copenhagen, Rolighedsvej 26, Frederiksberg 1958, DenmarkPrecision viticulture is increasingly being applied to automate and optimize grape production in the vineyard. This paper describes the development of a method for automatic selection of regions of interest from hyperspectral images obtained of a row of vines and intended for prediction of soluble solids content. For this purpose, a dataset consisting of hyperspectral images of a row of ‘Sangiovese’ wine grapes was adopted. Hyperspectral images were acquired directly in the field by means of a hyperspectral imaging Vis/NIR system (400–1000 nm) mounted on a ground-based vehicle. The analyses were carried out on 17 different days, under clear or partly cloudy conditions, in the period between post-veraison and harvest. The vineyard row of Sangiovese vines was divided into 11 sections and a hyperspectral image for each section for each day of analysis was acquired. The regions of interest of the hyperspectral images, comprising the areas representing the grapes, were selected using a PLS-DA-based method. The best PLS-DA model provided excellent results, with sensitivity and specificity values of 0.991 and 0.996, respectively. The mean spectra of the selected regions of interest (ROI) were finally used to predict the soluble solids content (SSC) of the grapes by PLS regression to a primary reference analysis. The results of SSC predictions using the automatic selection of ROIs (R2CV = 0.74 and RMSECV = 0.86 °Brix) were on par with similar regression based on carefully manual selection of ROIs (R2CV = 0.73 and RMSECV = 0.87 °Brix).http://www.sciencedirect.com/science/article/pii/S277237552400039XHyperspectral imagingGrapeAutomaticClassificationPLS-DAPLS |
| spellingShingle | Alessandro Benelli Chiara Cevoli Angelo Fabbri Søren Balling Engelsen Klavs Martin Sørensen Precision viticulture: Automatic selection of the regions of interest from moving wagon hyperspectral images of grapes for efficient SSC prediction Smart Agricultural Technology Hyperspectral imaging Grape Automatic Classification PLS-DA PLS |
| title | Precision viticulture: Automatic selection of the regions of interest from moving wagon hyperspectral images of grapes for efficient SSC prediction |
| title_full | Precision viticulture: Automatic selection of the regions of interest from moving wagon hyperspectral images of grapes for efficient SSC prediction |
| title_fullStr | Precision viticulture: Automatic selection of the regions of interest from moving wagon hyperspectral images of grapes for efficient SSC prediction |
| title_full_unstemmed | Precision viticulture: Automatic selection of the regions of interest from moving wagon hyperspectral images of grapes for efficient SSC prediction |
| title_short | Precision viticulture: Automatic selection of the regions of interest from moving wagon hyperspectral images of grapes for efficient SSC prediction |
| title_sort | precision viticulture automatic selection of the regions of interest from moving wagon hyperspectral images of grapes for efficient ssc prediction |
| topic | Hyperspectral imaging Grape Automatic Classification PLS-DA PLS |
| url | http://www.sciencedirect.com/science/article/pii/S277237552400039X |
| work_keys_str_mv | AT alessandrobenelli precisionviticultureautomaticselectionoftheregionsofinterestfrommovingwagonhyperspectralimagesofgrapesforefficientsscprediction AT chiaracevoli precisionviticultureautomaticselectionoftheregionsofinterestfrommovingwagonhyperspectralimagesofgrapesforefficientsscprediction AT angelofabbri precisionviticultureautomaticselectionoftheregionsofinterestfrommovingwagonhyperspectralimagesofgrapesforefficientsscprediction AT sørenballingengelsen precisionviticultureautomaticselectionoftheregionsofinterestfrommovingwagonhyperspectralimagesofgrapesforefficientsscprediction AT klavsmartinsørensen precisionviticultureautomaticselectionoftheregionsofinterestfrommovingwagonhyperspectralimagesofgrapesforefficientsscprediction |