A Novel Model Fusion Approach for Greenhouse Crop Yield Prediction
In this work, we have proposed a novel methodology for greenhouse tomato yield prediction, which is based on a hybrid of an explanatory biophysical model—the Tomgro model, and a machine learning model called CNN-RNN. The Tomgro and CNN-RNN models are calibrated/trained for predicting tomato yields w...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-12-01
|
Series: | Horticulturae |
Subjects: | |
Online Access: | https://www.mdpi.com/2311-7524/9/1/5 |
_version_ | 1797441798022365184 |
---|---|
author | Liyun Gong Miao Yu Vassilis Cutsuridis Stefanos Kollias Simon Pearson |
author_facet | Liyun Gong Miao Yu Vassilis Cutsuridis Stefanos Kollias Simon Pearson |
author_sort | Liyun Gong |
collection | DOAJ |
description | In this work, we have proposed a novel methodology for greenhouse tomato yield prediction, which is based on a hybrid of an explanatory biophysical model—the Tomgro model, and a machine learning model called CNN-RNN. The Tomgro and CNN-RNN models are calibrated/trained for predicting tomato yields while different fusion approaches (linear, Bayesian, neural network, random forest and gradient boosting) are exploited for fusing the prediction result of individual models for obtaining the final prediction results. The experimental results have shown that the model fusion approach achieves more accurate prediction results than the explanatory biophysical model or the machine learning model. Moreover, out of different model fusion approaches, the neural network one produced the most accurate tomato prediction results, with means and standard deviations of root mean square error (RMSE), r2-coefficient, Nash-Sutcliffe efficiency (NSE) and percent bias (PBIAS) being 17.69 ± 3.47 g/m<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mrow></mrow><mn>2</mn></msup></semantics></math></inline-formula>, 0.9995 ± 0.0002, 0.9989 ± 0.0004 and 0.1791 ± 0.6837, respectively. |
first_indexed | 2024-03-09T12:30:16Z |
format | Article |
id | doaj.art-046594dc207e45ff8d85965d313d8b00 |
institution | Directory Open Access Journal |
issn | 2311-7524 |
language | English |
last_indexed | 2024-03-09T12:30:16Z |
publishDate | 2022-12-01 |
publisher | MDPI AG |
record_format | Article |
series | Horticulturae |
spelling | doaj.art-046594dc207e45ff8d85965d313d8b002023-11-30T22:29:29ZengMDPI AGHorticulturae2311-75242022-12-0191510.3390/horticulturae9010005A Novel Model Fusion Approach for Greenhouse Crop Yield PredictionLiyun Gong0Miao Yu1Vassilis Cutsuridis2Stefanos Kollias3Simon Pearson4School of Computer Science, University of Lincoln, Lincoln LN6 7TS, UKSchool of Computer Science, University of Lincoln, Lincoln LN6 7TS, UKSchool of Computer Science, University of Lincoln, Lincoln LN6 7TS, UKSchool of Computer Science, University of Lincoln, Lincoln LN6 7TS, UKLincoln Institute for Agri-Food Technology, University of Lincoln, Lincoln LN6 7TS, UKIn this work, we have proposed a novel methodology for greenhouse tomato yield prediction, which is based on a hybrid of an explanatory biophysical model—the Tomgro model, and a machine learning model called CNN-RNN. The Tomgro and CNN-RNN models are calibrated/trained for predicting tomato yields while different fusion approaches (linear, Bayesian, neural network, random forest and gradient boosting) are exploited for fusing the prediction result of individual models for obtaining the final prediction results. The experimental results have shown that the model fusion approach achieves more accurate prediction results than the explanatory biophysical model or the machine learning model. Moreover, out of different model fusion approaches, the neural network one produced the most accurate tomato prediction results, with means and standard deviations of root mean square error (RMSE), r2-coefficient, Nash-Sutcliffe efficiency (NSE) and percent bias (PBIAS) being 17.69 ± 3.47 g/m<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mrow></mrow><mn>2</mn></msup></semantics></math></inline-formula>, 0.9995 ± 0.0002, 0.9989 ± 0.0004 and 0.1791 ± 0.6837, respectively.https://www.mdpi.com/2311-7524/9/1/5biophysical modeldeep neural networkrecurrent neural networkconvolutional neural networkmodel fusioncrop yield prediction |
spellingShingle | Liyun Gong Miao Yu Vassilis Cutsuridis Stefanos Kollias Simon Pearson A Novel Model Fusion Approach for Greenhouse Crop Yield Prediction Horticulturae biophysical model deep neural network recurrent neural network convolutional neural network model fusion crop yield prediction |
title | A Novel Model Fusion Approach for Greenhouse Crop Yield Prediction |
title_full | A Novel Model Fusion Approach for Greenhouse Crop Yield Prediction |
title_fullStr | A Novel Model Fusion Approach for Greenhouse Crop Yield Prediction |
title_full_unstemmed | A Novel Model Fusion Approach for Greenhouse Crop Yield Prediction |
title_short | A Novel Model Fusion Approach for Greenhouse Crop Yield Prediction |
title_sort | novel model fusion approach for greenhouse crop yield prediction |
topic | biophysical model deep neural network recurrent neural network convolutional neural network model fusion crop yield prediction |
url | https://www.mdpi.com/2311-7524/9/1/5 |
work_keys_str_mv | AT liyungong anovelmodelfusionapproachforgreenhousecropyieldprediction AT miaoyu anovelmodelfusionapproachforgreenhousecropyieldprediction AT vassiliscutsuridis anovelmodelfusionapproachforgreenhousecropyieldprediction AT stefanoskollias anovelmodelfusionapproachforgreenhousecropyieldprediction AT simonpearson anovelmodelfusionapproachforgreenhousecropyieldprediction AT liyungong novelmodelfusionapproachforgreenhousecropyieldprediction AT miaoyu novelmodelfusionapproachforgreenhousecropyieldprediction AT vassiliscutsuridis novelmodelfusionapproachforgreenhousecropyieldprediction AT stefanoskollias novelmodelfusionapproachforgreenhousecropyieldprediction AT simonpearson novelmodelfusionapproachforgreenhousecropyieldprediction |