Near-infrared spectral imaging for quantitative analysis of active component in counterfeit imidacloprid using PLS regression

Near-infrared (NIR) imaging systems simultaneously record spectral and spatial information. Near-infrared imaging was applied to the identification of imidacloprid in both artificially mixed samples and commercial formulation in this study. The distributions of imidacloprid and additive in the heterogeneous counterfeit were obtained by the relationship imaging (RI) mode. Furthermore a series of samples which consisted of different contents of uniformly distributed imidacloprid were prepared and three data cubes were generated at each content level. Extracted spectra from those images were imported to establish the partial least squares model. The model's results were: R2 99.21%, RMSEC 0.0306, RMSECV 0.0183, RMSECV/mean value 0.0348 and RSEP 0.0784. The prediction relative error of commercial formulation is 0.0680, indicating the predicted value was correlated to the real content. Lastly the chemical value reconstruction image of imidacloprid formulation products was calculated by MATLAB program. NIR microscopy imaging manifests herein its potential in qualitatively identifying the active component in counterfeit pesticide and quantifying the active component in scanned image.