초분광 반사광 영상을 이용한 방울토마토 내부품질 인자 예측

Hyperspectral reflectance imaging technology was used to predict internal quality of cherry tomatoes with the spectral range of 400-1000 nm. Partial least square (PLS) regression method was used to predict firmness, sugar content, and acid content. The PLS models were developed with several preprocessing methods, such as normalization, standard normal variate (SNV), multiplicative scatter correction (MSC), and derivative of Savitzky Golay. The performance of the prediction models were investigated to find the best combination of the preprocessing and PLS models. The coefficients of determination (R_p²) and standard errors of prediction (SEP) for the prediction of firmness, sugar content, and acid content of cherry tomatoes from green to red ripening stages were 0.876 and 1.875 kgf with mean of normalization, 0.823 and 0.388°Bx with maximum of normalization, and 0.620 and 0.208% with maximum of normalization, respectively.