Non-destructive determination of pericarp break force of Lingwu long jujube by hyperspectral imaging technology

Using hyperspectral imaging technology to measure pericarp break force of Lingwu jujube could provide a scientific method for the non- destructive measurement of external quality of Lingwu jujube. The hyperspectral image of 120 Lingwu long jujubes were acquired from 400 nm to 1000 nm,and pretreatment methods were used to process diffuse reflectance spectroscopy. The successive projections algorithm( SPA),competitive adaptive reweighed sampling( CARS) and principal uninformative variable elimination( UVE) were used to select characteristic wavelengths.A partial least squares regression( PLSR) and a least squares support vector machine( LS-SVM) model were established based on full spectra and selected characteristic wavelengths for predicting pericarp break force of Lingwu long jujube.The result showed that standardized normal variate( SNV) pretreatment method was best,and the correlation coefficient of cross calibration( Rcv) of the PLSR model built was 0.8207,root mean square error of cross validation( RMSEP) reached 9.9630. The SPA,CARS and UVE method were used to select characteristic wavelengths with 29,31 and 31 from 125 wavelengths,respectively.LS- SVM model based on full spectra was the best,and correlation coefficient of prediction( Rp) and root- mean- square error of prediction( RMSEP) of the model was 0.9555 and 3.8282,respectively. The results indicated that the non- destructive measurement for pericarp break force of Lingwu jujube based on hyperspectral imaging technology was feasible.