Enhancing fruit SSC detection accuracy via a light attenuation theory-based correction method to mitigate measurement
Penghui Liu1, Yuanhao Zheng1, Hao Tian2
1College of Biosystems Engineering and Food Science, Zhejiang University, 866 Yuhangtang Road, Hangzhou 310058, PR China; Key Laboratory of Intelligent Equipment and Robotics for Agriculture of Zhejiang Province, Hangzhou 310058, PR China; The National Key Laboratory of Agricultural Equipment Technology, Beijing 100083, PR China.
Food Research International (Ottawa, Ont.)
|November 30, 2024
Summary
Fruit quality detection accuracy improves by correcting for measurement orientation variations. A new method based on light attenuation theory enhances spectral data, boosting apple soluble solids content (SSC) prediction models.
Area of Science:
- Agricultural Science
- Spectroscopy
- Data Analysis
Background:
- Nondestructive fruit quality detection is crucial for the agro-product industry.
- Accuracy is challenged by factors like fruit orientation and environmental changes.
Purpose of the Study:
- To investigate the impact of measurement orientation on apple spectra and soluble solids content (SSC) detection.
- To develop a correction method for orientation-induced spectral variations.
Main Methods:
- Collected visible/near-infrared (Vis/NIR) spectra (550-950 nm) across four orientations.
- Developed local and global calibration models.
- Applied a light attenuation theory-based method for spectral correction.
Main Results:
- Measurement orientation significantly altered spectral intensity and reduced model predictive power.
- Global models showed less susceptibility to orientation variation than local models.
- Orientation correction substantially improved model performance (Rp2, RPD) and reduced errors (RMSEP).
Conclusions:
- Measurement orientation significantly impacts fruit quality detection accuracy.
- The proposed light attenuation-based correction method effectively mitigates orientation effects.
- This approach offers a cost-effective and universal solution for reliable online fruit quality assessment.
