Journal of Researches in Mechanics of Agricultural Machinery

Journal of Researches in Mechanics of Agricultural Machinery

Applying the Radom Forest method to determine the amount of sugar, acidity and hardness of cherries with the benefit of spectroscopic technique (Vis-NIR)

Document Type : Research Paper

Authors
Gholamhossein Shahgholi 1
Sina Feizolahzadeh Ardabili 2
1 Department of Biosystems Engineering,, Faculty of agriculture and natural resources, University of Mohaghegh Ardabili, Ardabil, Iran
2 Department of Biosystems Engineering, Faculty of agriculture anf natural resurces, University of Mohaghegh Ardabili, Ardabil, Iran
10.22034/JRMAM.2024.14719.700
Abstract
This study investigates the application of random forest algorithm in relation to spectroscopic techniques, especially Vis-NIR (Vis-NIR) spectroscopy, to predict key quality characteristics of cherries. The aim of this article is to provide a reliable and efficient method for non-destructive evaluation of sugar content, acidity and hardness which are important factors determining cherry quality and consumer acceptance. In this study, 200 random samples of cherry products were prepared. Spectral data were obtained. The spectral data obtained from Vis-NIR spectroscopy were evaluated by the Relief method, and five effective wavelengths were selected separately for each of the values of the dependent variables of sugar, acidity, and hardness values. The selected effective wavelengths were then used as input features for a random forest model, which was trained on a dataset containing samples with known sugar content, acidity levels, and hardness measurements. The performance of the model was evaluated in terms of prediction accuracy and robustness through cross-validation and independent testing using root mean square error (RMSE) and correlation coefficient (CC). The results showed that the effectiveness of the random forest algorithm in accurately predicting the sugar content, acidity and hardness of cherries based on spectral information. The proposed approach provides a rapid, non-destructive and cost-effective solution for quality assessment in the cherry industry and enables producers and stakeholders to make informed decisions about harvesting, sorting and post-harvest processes.
Keywords
spectroscopy
cherry
non-destructive testing
feature selection
random forest
Subjects
Andersen, P. V., Afseth, N. K., Aaby, K., Gaarder, M. Ø., Remberg, S. F., & Wold, J. P. (2023). Prediction of chemical and sensory properties in strawberries using Raman spectroscopy. Post-harvest Biology and Technology, 201: 112370.
 
Cao, J., Li, X., Liu, Y., Leng, F., Li, X., Sun, C., & Chen, K. (2015). Bioassay‐Based Isolation and Identification of Phenolics from Sweet Cherry That Promote Active Glucose Consumption by HepG2 Cells. Journal of food science, 80(2): C234-C240.
 
Díaz-Mula, H. M., Castillo, S., Martínez-Romero, D., Valero, D., Zapata, P. J., Guillén, F., & Serrano, M. (2009). Sensory, nutritive and functional properties of sweet cherry as affected by cultivar and ripening stage. Food Science and Technology International, 15(6): 535-543.
 
Fazari, A., Pellicer-Valero, O. J., Gómez-Sanchıs, J., Bernardi, B., Cubero, S., Benalia, S., Zimbalatti, G., & Blasco, J. (2021). Application of deep convolutional neural networks for the detection of anthracnose in olives using VIS/NIR hyperspectral images. Computers and Electronics in Agriculture, 187: 106252.
 
Genuer, R., Poggi, J.-M., Genuer, R., & Poggi, J.-M. (2020). Random forests. Springer.
 
Guo, Y., Ni, Y., & Kokot, S. (2016). Evaluation of chemical components and properties of the jujube fruit using near infrared spectroscopy and chemometrics. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 153: 79-86.
 
Kalopesa, E., Gkrimpizis, T., Samarinas, N., Tsakiridis, N. L., & Zalidis, G. C. (2023). Rapid Determination of Wine Grape Maturity Level from pH, Titratable Acidity, and Sugar Content Using Non-Destructive In Situ Infrared Spectroscopy and Multi-Head Attention Convolutional Neural Networks. Sensors, 23(23): 9536.
 
Kim, S. Y., Hong, S. J., Kim, E., Lee, C. H., & Kim, G. (2023). Application of ensemble neural-network method to integrated sugar content prediction model for citrus fruit using Vis/NIR spectroscopy. Journal of Food Engineering, 338: 111254.
 
Kira, K., & Rendell, L. A. (1992). A practical approach to feature selection. In Machine learning proceedings 1992. Elsevier. pp. 249-256
 
Lu, R. (2001). Predicting firmness and sugar content of sweet cherries using near–infrared diffuse reflectance spectroscopy. Transactions of the ASAE, 44(5): 1265.
 
Mahmoodi-Eshkaftaki, M., Mahbod, M., & Ghenaatian, H. R. (2024). Non-destructive estimation of biomass characteristics: Combining hyperspectral imaging data with neural networks. Renewable energy, 224: 120137.
 
Parmar, A., Katariya, R., & Patel, V. (2019). A review on random forest: An ensemble classifier. International conference on intelligent data communication technologies and internet of things (ICICI) 2018.
 
Pourdarbani, R., Sabzi, S., & Arribas, J. I. (2021). Non-destructive estimation of three apple fruit properties at various ripening levels with optimal Vis-NIR spectral wavelength regression data. Heliyon, 7(9).
 
Pourdarbani, R., Sabzi, S., Kalantari, D., & Arribas, J. I. (2020). Non-destructive visible and short-wave near-infrared spectroscopic data estimation of various physicochemical properties of Fuji apple (Malus pumila) fruits at different maturation stages. Chemometrics and Intelligent Laboratory Systems, 206: 104147.
 
Rong, D., Wang, H., Ying, Y., Zhang, Z., & Zhang, Y. (2020). Peach variety detection using VIS-NIR spectroscopy and deep learning. Computers and Electronics in Agriculture, 175: 105553.
 
Ruiz-Altisent, M., Ruiz-Garcia, L., Moreda, G., Lu, R., Hernandez-Sanchez, N., Correa, E., Diezma, B., Nicolaï, B., & García-Ramos, J. (2010). Sensors for product characterization and quality of specialty crops—A review. Computers and Electronics in agriculture, 74(2): 176-194.
 
Shao, Y., Xuan, G., Hu, Z., Gao, Z., & Liu, L. (2019). Determination of the bruise degree for cherry using Vis-NIR reflection spectroscopy coupled with multivariate analysis. PloS One, 14(9): e0222633.
 
Timm, E., Guyer, D., Brown, G., & Schulte, N. (1995). Michigan sweet cherry color measurement and prototype color chip development. Applied Engineering in Agriculture, 11(3): 403-407.
 
Urbanowicz, R. J., Meeker, M., La Cava, W., Olson, R. S., & Moore, J. H. (2018). Relief-based feature selection: Introduction and review. Journal of biomedical informatics, 85: 189-203.
 
Walsh, J., Neupane, A., & Li, M. (2024). Evaluation of 1D convolutional neural network in estimation of mango dry matter content. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 124003.
 
Zhang, C., & Ma, Y. (2012). Ensemble machine learning: methods and applications. Springer.