Identification of Nanguo Pear Maturity Based on Information Fusion

Maturity is not only an important factor affecting the internal quality of the Nanguo pear, but also an important theoretical basis for grading online fruit. Based on the hyperspectral imaging technology, in this paper, backpropagation neural network and support vector machine models are established to identify Nanguo pear maturity by information fusion of spectral features and image features. The results show that the identification results of the support vector machine based on information fusion of spectral features and image features are the best, and the recognition rate is above 95%. Among them, the recognition rates of immature and mature samples reach 100%.