Related Experiment Video
Updated: Sep 20, 2025
10:20
Evaluation of Photosynthetic Behaviors by Simultaneous Measurements of Leaf Reflectance and Chlorophyll Fluorescence Analyses
Published on: August 9, 2019
12.8K
Intelligent chlorophyll estimation by attention-integrated deep learning and dual-modal fusion in tencha drying using
Huilin Chang1, Jiazhen Cai1, Qin Ouyang1
1School of Food and Biological Engineering, Jiangsu University, Zhenjiang, China.
Journal of the Science of Food and Agriculture
|May 23, 2025
Summary
Snapshot multispectral technology and spectral-image fusion accurately predict chlorophyll content in tencha. This non-invasive method enhances quality control during the drying process for premium matcha production.
Area of Science:
- Agricultural Science
- Food Science
- Spectroscopy
Background:
- Chlorophyll content in tencha is critical for sensory evaluation of matcha.
- Drying process significantly impacts tencha's chlorophyll levels.
- Accurate monitoring is essential for quality control.
Purpose of the Study:
- To assess chlorophyll content in drying tencha using multispectral technology.
- To develop and compare predictive models for chlorophyll levels.
- To evaluate the efficacy of spectral-image fusion for non-invasive monitoring.
Main Methods:
- Snapshot multispectral imaging (660–924 nm) was employed.
- Reflectance data were fused with grayscale texture features.
- Convolutional neural network (CNN) and SE-Res18 models were developed and compared.
Main Results:
- Fusion approach significantly improved prediction accuracy compared to single data sources.
- The SE-Res18 model achieved high correlation coefficients (0.9814 training, 0.9337 testing).
- Low relative percent deviation (2.79) indicates model precision.
Conclusions:
- Snapshot multispectral technology combined with spectral-image fusion is viable for precise chlorophyll monitoring.
- This approach offers a rapid, non-invasive method for tencha quality control.
- The technology enhances quality assurance in tencha production.
