Detection of the Early Fungal Infection of Citrus by Fourier Transform Near-Infrared Spectra

Early fungal infection of citrus is one the common diseases found during storage period citrus, and fungus that infects will spread to entire batch as degree deepens, causing enormous economic losses. Therefore, early detection fundamental. The purpose this study explore qualitative identification infections in by using Fourier transform near-infrared (FT-NIR) combined with a variety chemometric methods. First, discrete wavelet (DWT) used filter noise spectral signal, then PLS-DA model, helps discriminate healthy from infected Citrus. Subsequently, four different feature variable selection methods were introduced, Then, linear discriminant analysis (LDA) support vector machine (SVM) two classifiers establish model for infection. modeling results show SVM effect better than LDA, DWT-CARS-SVM based on RBF kernel function has best result, accuracy rates training set test are 100% 97%. indicate FT-NIR spectroscopy, methods, able distinguish citrus.