As an important food crop, potato is often at-tacked by pests and diseases. Traditional pest identi-fication relies on the visual observation of agricul-tural workers for empirical distinction, which has a small detection range, high labor intensity, and low operating efficiency.This paper takes potato pests and diseases images under natural conditions as the research object, and uses image processing and pat-tern recognition technology to automatically classify pests and diseases. Firstly, in view of the problems oftraditional potato pest detection methods,a potato pest detection model based on Faster R-CNN is pro-posed; Secondly, the residual convolutional network is used to extract image features, Max-pooling is a down-sampling method, the feature pyramid net-work is introduced into the RPN network to generate object proposals, and the convolutional neural net-work structure is optimized; Finally, construct a po-tato pest data set, and implement model training and testing to detect potato pests. The test results based on the TensorFlow framework show that the opti-mized neural network algorithm has an average recognition accuracy of97.8 for typical potato pest images. The optimized convolutional neural network recognition model has stronger robustness and ap-plicability, and can provide a reference for the iden-tification and intelligent diagnosis of potato and other crop pests.
展开▼
