Standardization of near infrared hyperspectral imaging for wheat single kernel sorting according to deoxynivalenol level

The spatial recognition feature of near infrared hyperspectral imaging (HSI-NIR) makes it potentially suitable for Fusarium and deoxynivalenol (DON) management in single kernels to break with heterogeneity contamination wheat batches move towards individual kernel sorting provide more quick, environmental-friendly non-destructive analysis than wet-chemistry techniques. aim this study was standardize HSI-NIR damage DON presence, predict the level classify grains according EU maximum limit (1250 µg/kg). Visual inspection on infection symptoms HPLC determination were used as reference methods. scanned both crease-up crease-down position different image captures. spectra pretreated by Multiplicative Scatter Correction (MSC) Standard Normal Variate (SNV), 1st 2nd derivatives normalisation, they evaluated also removing spectral tails. best fitted predictive model SNV data (R2 0.88 RMSECV 4.8 mg/kg) which 7 characteristic wavelengths used. Linear Discriminant Analysis (LDA), Naïve Bayes K-nearest Neighbours models classified 100% accuracy derivative symptomatology 98.9 98.4% correctness spectra, respectively. starting point results are encouraging future investigations technique application overcome their heterogeneity.