Research On Rice Disease Detection System Based On Artificial Intelligence

With the improvement of computing power of computers and chips,research in the field of artificial intelligence has re-emerged.Among them,Convolutional Neural Network(CNN)is widely used in image recognition in various fields,and gradually shifts to mobile and embedded end.Rice disease detection has always been the primary problem in the detection of plant diseases.How to accurately and efficiently detect rice diseases and increase rice yield is a hot issue in agriculture.Based on the theory of artificial intelligence,the rice disease detection system is studied.The system is mainly designed from three parts: image acquisition,temperature and humidity collection and disease detection model.ARM is used as the main control unit of the embedded platform to realize rice lesion image collection and environmental temperature and humidity detection.The collected image is used as the input of the detection model to complete the off-line detection of rice diseases,and the diagnosis is assisted by environmental factors such as temperature and relative humidity.The model design was carried out on the PC,and the TensorFlow deep learning framework was built.The design and training of the rice disease detection model was completed by convolutional neural network,and the optimized model was transplanted to the embedded end to realize the detection and recognition function of the terminal.Tests show that the rice disease detection system can complete the basic functions,the image acquisition rate is 25 frames / sec,the resolution of the collected lesion image is 640 × 480;the system measured humidity range is 0 ~ 99.9% RH,the temperature range is-40~80°C;the average accuracy of disease identification was 96%.