Methodological analysis demonstrates improved SSC prediction in plums using machine learning and optical modeling techniques, suggesting enhanced fruit quality assessment.
Soluble Solid Content (SSC) is a key parameter for assessing fruit quality, reflecting sweetness and nutritional value. Traditional methods for SSC measurement, such as chemical assays, require destructive sampling, limiting their applicability to large-scale quality monitoring. This study presents a nondestructive SSC detection method combining multispectral imaging, hybrid machine learning (RF-PLS), and physics-based optical modeling. Optical data of plum samples were acquired using a multispectral camera (350–950 nm). Noise reduction was performed via Savitzky-Golay filtering and SNV transformation, followed by wavelength selection using SPA and RF-FI scoring. By integrating the Kubelka-Munk theory to decode light propagation in fruit flesh, we established a hybrid model that synergizes physical insights with machine learning. The RF-PLS framework achieved superior accuracy (R²=0.89, RMSE=2.17%) over standalone algorithms, validating the power of multimodal fusion in SSC prediction.
No takes yet. Share an insight, caveat, or question.
Mingsheng Cai (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: