Moisture content prediction of cigar leaves air-curing process based on stacking ensemble learning model

Frontiers in Plant Science, Journal Year: 2025, Volume and Issue: 16

Published: March 26, 2025

Introduction Accurately determining the moisture content of cigar leaves during air-curing process is crucial for quality preservation. Traditional measurement techniques are often subjective and destructive, limiting their practical application. Methods In this study, we propose a stacking ensemble learning model non-destructive prediction, leveraging image-based analysis naturally suspended leaves. front rear surface images were collected throughout process. Color texture features extracted from these images, filtering method was applied to remove redundant variables. To ensure optimal selection, entropy weight employed comprehensively evaluate candidate machine models, leading construction model. Furthermore, SHAP quantify contribution each input feature prediction results. Results The model, comprising MLP, RF, GBDT as base learners LR meta-learner, achieved superior accuracy ( R 2 test =0.989) outperforms than traditional models ranged 0.961 0.982). revealed that (45.5%) leaf (38.5%) most influential predictors, with airing period AP ), f * , G ASM identified key predictors. Conclusion This study provides feasible scalable solution real-time monitoring content, offering effective technical support similar agricultural food drying applications.

Language: Английский

The Effect of Anaerobic Fermentation on the Aroma Components of Cigar Tobacco Leaves

Hans Journal of Agricultural Sciences, Journal Year: 2024, Volume and Issue: 14(12), P. 1339 - 1347

Published: Jan. 1, 2024

Language: Английский

Quality prediction of air-cured cigar tobacco leaf using region-based neural networks combined with visible and near-infrared hyperspectral imaging

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Dec. 28, 2024

Language: Английский