An approach for identifying of Fusarium infected single maize grains based on diffuse reflectance in visible and near infrared region is proposed in the paper.Spectral characteristics were collected in the range 400-2500 viking g-5 nm in steps of 2 nm.Soft independent modeling of class analogy (SIMCA) is used for data processing.
Maize grains classification is based on SIMCA classifier 4 post backdrop stand and Probabilistic neural network (PNN).Recognition accuracy which is achieved for both classes of grains is respectively 99.89% for healthy, and 93.
7% for infected.