Answers are grounded in this paper and cite their sources.
This study employs a deep-learning method (Y-Net) to estimate 10 tea flavor-related chemical compounds (TFCCs) using fresh tea shoot reflectance and transmittance. The unique aspect of Y-Net lies in its utilization of dual inputs, reflectance, and transmittance, which are seamlessly integrated within the Y-Net architecture. This architecture harnesses the power of a convolutional neural network-based residual network to fuse tea shoot spectra effectively. This strategic combination enhances the capacity of the model to discern intricate patterns in the optical characteristics of fresh tea shoots, providing a comprehensive framework for TFCC estimation. In this study, we sampled tea shoots in Alishan, Central Taiwan, and measured their reflectance and transmittance ( n = 2032) within the optical region (400–2500 nm) using a portable spectroradiometer. Corresponding TFCCs were quantified using the high-performance liquid chromatography analysis. To enhance Y-Net, we employed data augmentation techniques for model training. We compared the performances of Y-Net with seven other commonly utilized statistical, machine-/deep-learning models. Furthermore, we assessed the prediction accuracies of Y-Net, and Y-Net using spectra within the visible–near-infrared (Vis-NIR) regions (for higher energy throughput and low-cost instruments) and reflectance only (for airborne/spaceborne remote sensing applications). The results showed that overall Y-Net, i.e., mean root mean square error (RMSE) ± standard deviation (SD) = 0.65 ± 0.58 mg g –1 , outperformed the other statistical, machine and deep learning models (≥2.33 ± 2.14 mg g –1 ), demonstrating its superiority in predicting TFCCs. In addition, this original Y-Net also yielded significantly lower mean RMSE ± SD compared with Vis-NIR (3.15 ± 3.35 mg g –1 ) and reflectance-only (2.23 ± 2.25 mg g –1 ) Y-Nets using validation data. This study highlights the feasibility of using spectroscopy and Y-Net to assess minor biochemical components in fresh tea shoots and sheds light on the potential of the proposed approach for effective regional monitoring of tea shoot quality.
No takes yet. Share an insight, caveat, or question.
Denaro et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: