A Deep Learning Based Method for Identifying the Purity of Corn Seed Varieties with Hyperspectral Features
In order to improve the recognition accuracy of corn seeds and improve the yield and quality of corn,a method based on hyperspectral imaging and deep learning technology including data preprocessing,image segmentation,and im-proved convolutional neural network(CNN)based on hyperspectral corn seed purity identification method.The research results show that the improved CNN has significant effects in improving training performance,and its accuracy,precision,recall rate and F1 score are superior to traditional machine learning and other deep learning methods.
万方数据
维普
