Innovative strategies for protein content determination in dried laver (Porphyra spp.): Evaluation of preprocessing methods and machine learning algorithms through short-wave infrared imaging
Eunghee Kim, Jong‐Jin Park, Gyuseok Lee, Jeong‐Seok Cho, Seulki Park, Dae-Yong Yun, Kee‐Jai Park, Jeong‐Ho Lim
Abstract
In this study, we explored the application of Short-Wave Infrared (SWIR) hyperspectral imaging combined with Competitive Adaptive Reweighted Sampling (CARS) and advanced regression models for the non-destructive assessment of protein content in dried laver. Utilizing a spectral range of 900–1700 nm, we aimed to refine the quality control process by selecting informative wavelengths through CARS and applying various preprocessing techniques (standard normal variate [SNV], Savitzky-Golay filtering [SG], Orthogonal Signal Correction [OSC], and StandardScaler [SS]) to enhance the model's accuracy. The SNV-OSC-StandardScaler- Support vector regression (SVR) model trained on CARS-selected wavelengths significantly outperformed the other configurations, achieving a prediction determination coefficient (Rp 2 ) of 0.9673, root mean square error of prediction of 0.4043, and residual predictive deviation of 5.533. These results highlight SWIR hyperspectral imaging's potential as a rapid and precise tool for assessing dried laver quality, aiding food industry quality control and dried laver market growth. • Utilized SWIR hyperspectral imaging for non-destructive protein content determination in dried laver. • Applied Competitive Adaptive Reweighted Sampling (CARS) for efficient wavelength selection. • Demonstrated superior performance of the SNV-OSC-StandardScaler-SVR model in protein prediction. • Achieved high predictive accuracy with Rp 2 of 0.9673, indicating effective model calibration and validation. • Developed visual chemical maps to depict protein distribution, enhancing practical quality assessment.