Non-destructive prediction of total soluble solids and titratable acidity in Kyoho grape using hyperspectral imaging and deep learning algorithm

Total soluble solids (TSS) and titratable acidity (TA) are essential quality properties for postharvest commercialisation of grapes. This study aimed to estimate the TSS and TA in grapes using hyperspectral imaging (HSI) technique in the range of 400-1001 nm. A deep learning-based stacked auto-encoders (SAE) algorithm was developed to extract deep spectral features from pixel-level spectra. Then, these features with a compensation factor (i.e. size of fruits) were fed into partial least squares (PLS) and least squares support vector machine (LSSVM) for predicting TSS and TA in grapes. Additionally, competitive adaptive reweighed sampling and successive projections algorithm as conventional wavelength selection approaches were also investigated for comparison. The optimal prediction accuracy was achieved by the SAE-LSSVM model with size compensation, where RP2=0.9237$$ {R}_P<^>2=0.9237 $$, RMSEP = 0.5041% and RPD=3.25$$ RPD=3.25 $$ for TSS; Rp2=0.9216$$ {R}_p<^>2=0.9216 $$, RMSEP = 0.1091 g L-1 and RPD=3.21$$ RPD=3.21 $$ for TA. The results suggested that SAE has great potential for extracting features from pixel-level hyperspectral image data; the well-performed deep learning model SAE-LSSVM with size compensation can be used for rapid and non-destructive predicting TSS and TA in grapes, which may provide a valuable reference for internal quality evaluation of postharvest fruits via HSI technique.

Automated summary

This study utilized hyperspectral imaging technique and deep learning-based stacked auto-encoders algorithm to estimate total soluble solids and titratable acidity in grapes. The results showed that the deep learning model performed well in predicting the quality properties of grapes, and the combination with size compensation further improved the prediction accuracy. This research provides a valuable reference for non-destructive evaluation of postharvest fruits using hyperspectral imaging technique.