A Portable Nondestructive Instrument Based on Laser Backscattering Imaging to Detect Firmness and Soluble Solids Content of Peaches
J. Yang, M. Xu, L-Q Pan and K. Tu
A small volume, digital display and nondestructive instrument based on laser backscattering imaging was designed and developed to determine the internal quality of peaches; namely the firmness and soluble solids content. According to the relationship between backscattering image parameters and internal quality parameters of the peaches, we found that pixel area, uniformity and entropy were correlated with the physical and chemical parameters of the peaches. The partial least squares discrimination analysis (PLS-DA) and support vector classification (SVC) models were built with the peach image feature parameters as inputs and the results showed that the grading accuracy of SVC models on firmness and soluble solids content were better than the PLS-DA models. For Hujingmilu and Zaobaihua peaches, the overall classification accuracies of soluble solids content were 94 and 92% and the prediction accuracy were 94 and 91%, respectively. Additionally, the overall classification accuracies of firmness were 95 and 93%, and the prediction accuracy were 94 and 93% respectively. This research demonstrates the feasibility of developing portable quality classification instruments based on laser backscattering imaging for fruits.
Keywords: Diode laser, peach, laser scattering image, soluble solids content, firmness, pixel area, uniformity, entropy, partial least squares discrimination analysis (PLS-DA), support vector classification (SVC)