Nondestructive detection of total phenolic content of loquat based on hyperspectral imaging
To evaluate the feasibility of using hyperspectral imaging(HSI)to predict the total phenol content(TPC)in loquat during postharvest cold storage,the present experiment was conducted to collect the spectral information of 115 loquat samples using HSI in the Vis-NIR wavelength band(363-1 026 nm),and then the TPC was determined by the forintol method.The anomalous samples were eliminated by the Monte Carlo algorithm and then the competitive adaptive reweighted sampling(CARS)algorithm was used to select the featured bands.Nonlinear iterative partial least square(NIPALS)and simple partial least square(SIMPLS)algorithms of TPC in loquat were established and made a validation respectively.As the result shows,47 featured bands,accounting for 7.62%of the total number of wavelengths,are selected by CARS.CARS-NIPALS and CARS-SIMPLS reach high prediction accuracy,with the 2cR of the best principal component 0.922 1 and 0.916 0,and the R2p of the best principal component 0.816 6 and 0.812 8 respectively.NIPALS and SIMPLS perform similarly.Therefore,it can be shown that HSI can effectively predict the TPC of loquat during cold storage.
food packaging and storagenonlinear iterative partial least square(NIPALS)simple partial least square(SIMPLS)loquattotal phenolichyperspectral imaging(HSI)
NETL
NSTL
万方数据
