Internally mildewed sunflower seeds, which cannot be recognized and discarded based on their appearance, pose a serious risk to human health. Thus, there is a need for a rapid non-destructive mildew grade discrimination method. Currently, few reports are available regarding this process. In this study, a method based on the combination of the near-infrared diffuse reflectance and near-infrared diffuse transmission (NIRr-NIRt) fusion spectra and a one-dimension convolutional neural network (1D-CNN) is proposed. The NIRr-NIRt fusion spectra can provide more complementary and comprehensive information, and therefore better discrimination accuracy, than a single spectrum. The first derivative (FD) preprocessing method could further improve the discrimination effect. By comparison against three conventional machine learning algorithms (artificial neural network (ANN), support vector machine (SVM), and K-nearest neighbor (KNN)), the 1D-CNN model based on the fusion spectra was found to perform the best. The mean prediction accuracy was 2.01%, 5.97%, and 10.55% higher than that of the ANN, SVM, and KNN models, respectively. These results indicate that the CNN model was able to precisely classify the mildew grades with a prediction accuracy of 97.60% and 94.04% for the training and test set, respectively. Thus, this study provides a non-destructive and rapid method for classifying the mildew grade of sunflower seeds with the potential to be applied in the quality control of sunflower seeds.

Internally mildewed sunflower seeds pose a significant risk to human health. To control internal mildew, it is imperative to study its source in the main production area of China, which has been little investigated. Here, high-throughput sequencing was used to characterize the fungal and fungus-seed communities. Alpha diversity and ANOSIM analyses showed mildew did not alter the fungal compositions significantly. STAMP analysis showed that the sunflower seeds were most vulnerable to internal mildew during the field-planting stage. Alternaria was the predominant mildew-causing pathogen of sunflower seeds for consumption, which may originate from seed transmission and colonize at the seed-development stage. Finally, only a few seeds developed internal mildew with a worrisome level of Alternaria contamination in the humid field climate. NMDS analysis showed that climatic factors also played important roles in shaping microbial change during storage, with a relative humidity (RH) of 67% being the critical threshold in normal-temperature warehouses. Internal mildew never occurred below the RH threshold for the microbial community structure, which hardly changed after an average storage duration. The results indicated that a combination of field management to combat Alternaria, pretreatment with 5 KGy γ-irradiation and drying at the time of storage will minimize or prevent internal mildew. This work also provides an empirical framework for studies of mildewing in other shelled seeds.