Visualization and prediction of TVB-N content in chilled pork by hyperspectral imaging

TVB-N is one of the important indexes freshness chilled pork. To present more accurately and completely, a visualization technology chemical composition was applied optimized for pork detection. Two hyperspectral cameras covering visible near-infrared wavebands were used to detect content. GA, SPA, iPLSR, siPLSR employed select characteristic from spectra data extracted images, PLSR models established on full wavebands. According spectral data, distribution maps obtained, URV ICV evaluate performance maps. The results showed that, both achieved good accuracy, that cluster in range 600-750 nm, 1020-1120 1450-1570 nm. optimize maps, prediction reconstructed under guidance ICV. This study provided method predicting content imaging aspect online testing pork, which improves efficiency quality testing.