Prediction of the Maturity of Greenhouse Grapes Based on Imaging Technology
Xinguang Wei,Linlin Wu,Dong Ge,Mingze Yao,Yikui Bai
DOI: https://doi.org/10.34133/2022/9753427
IF: 6.961
2022-03-30
Plant Phenomics
Abstract:To predict grape maturity in solar greenhouses, a plant phenotype-monitoring platform (Phenofix, France) was used to obtain RGB images of grapes from expansion to maturity. Horizontal and longitudinal diameters, compactness, soluble solid content (SSC), titratable acid content, and the SSC/acid of grapes were measured and evaluated. The color values (R, G, B, H, S, and I) of the grape skin were determined and subjected to a back-propagation neural network algorithm (BPNN) to predict grape maturity. The results showed that the physical and chemical properties (PCP) of the three varieties of grapes changed significantly during the berry expansion stage and the color-changing maturity stage. According to the normalized rate of change of the PCP indicators, the ripening process of the three varieties of grapes could be divided into two stages: an immature stage (maturity coefficient Mc < 0.7) and a mature stage (after which color changes occurred) (0.7 ≤ Mc < 1). When predicting grape maturity based on the R, G, B, H, I, and S color values, the R, G, and I as well as G, H, and I performed well for Drunk Incense, Muscat Hamburg, and Xiang Yue grape maturity prediction. The GPI ranked in the top three (up to 0.87) when the above indicators were used in combination with BPNN to predict the grape Mc by single-factor and combined-factor analysis. The results showed that the prediction accuracy (RG and HI) of the two-factor combination was better for Drunk Incense, Muscat Hamburg, and Xiang Yue grapes (with recognition accuracies of 79.3%, 78.2%, and 79.4%, respectively), and all of the predictive values were higher than those of the single-factor predictions. Using a confusion matrix to compare the accuracy of the Mc's predictive ability under the two-factor combination method, the prediction accuracies were in the following order: Xiang Yue (88%) > Muscat Hamburg (81.3%) > Drunk Incense (76%). The results of this study provide an effective way to predict the ripeness of grapes in the greenhouse.