What question did this study set out to answer?

This research aims to develop a non-destructive method for monitoring chemical composition during tobacco curing using multimodal data.

In-situ and non-destructive monitoring of chemical components in tobacco leaf during curing based on multimodal data and machine learning

View Full Paper

Ask AI

Bookmark

Authors

Multimodal data predicts chemical composition in tobacco leaves during curing, indicating a new monitoring approach.

This research aims to develop a non-destructive method for monitoring chemical composition during tobacco curing using multimodal data.
Developed an online monitoring framework combining visible-light imaging, temperature, and humidity sensors.
Extracted colorimetric features from RGB, Lab, and YUV color models as inputs for machine learning models.
Trained models on datasets from two tobacco-growing regions and tested with external data for performance evaluation.
Achieved predictive accuracy for carbohydrates and amino acids with validation R² values exceeding 0.90.
Identified relative humidity and V value as dominant predictors for most chemical components, with temperature affecting alanine variation.
Demonstrated feasibility of using visual and environmental data for process monitoring and quality optimization.

Shi et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark

View Full Paper

Ask AI

Bookmark

Authors

Multimodal data predicts chemical composition in tobacco leaves during curing, indicating a new monitoring approach.

This research aims to develop a non-destructive method for monitoring chemical composition during tobacco curing using multimodal data.
Developed an online monitoring framework combining visible-light imaging, temperature, and humidity sensors.
Extracted colorimetric features from RGB, Lab, and YUV color models as inputs for machine learning models.
Trained models on datasets from two tobacco-growing regions and tested with external data for performance evaluation.
Achieved predictive accuracy for carbohydrates and amino acids with validation R² values exceeding 0.90.
Identified relative humidity and V value as dominant predictors for most chemical components, with temperature affecting alanine variation.
Demonstrated feasibility of using visual and environmental data for process monitoring and quality optimization.

Shi et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark