| Rice is one of the main food crops in China,and more than 60% of China’s population takes rice as the staple food.Rice disease is one of the main problems faced during rice growth.Therefore,the prevention and management of rice diseases have a significant impact on China’s food security and output.The basic premise of disease control is the correct diagnosis of the disease.Due to the relatively small income from agricultural production,the number of agricultural practitioners is small,and the level of education is generally not high.When faced with problems of rice planting and disease diagnosis,practitioners still rely on past experience,mainly through skill training and knowledge transfer to improve their coping ability,lacking efficient and scientific means.In recent years,with the improvement of computing power and the progress of related theoretical research,the deep learning method with neural network as the main method has been widely used in various fields.In this study,a rice disease identification network was constructed based on convolutional neural network,and based on the rice planting area in Changsha,combined with the information visualization system,the intelligence and visualization of rice planting and disease diagnosis and control were designed and realized from a macro perspective.The rice disease identification function designed and implemented in this research has great practical value in actual agricultural production.Users can quickly photograph crop disease samples through tools such as cameras and mobile phones,and then upload them to the disease identification system.The system can quickly diagnose diseases,help agricultural producers discover rice disease problems in a timely and accurate manner,and improve planting efficiency.The main work and results of this research are as follows:(1)Obtain data related to rice planting through reptiles,including rice varieties,rice diseases,and rice planting-related information in Changsha,and construct a rice disease picture data-set based on rice disease pictures obtained from multiplechannels;(2)A rice disease identification network was constructed based on the convolutional neural network,and the rice blast,bacterial blight and bacterial brown spot were identified.The results show that the accuracy of the recognition network constructed in this paper reaches 95.9%,and the training time is the least,which is better than the comparison models Alex Net and VGG16;(3)Combined with the Django framework,and using Python,html,css,js and other related technologies to design and implement a planting information visualization system based on the rice area of Changsha,the system functions include:login and registration function,home page display function,environmental factor analysis Function,social factor analysis function,rice variety query function,rice disease identification function. |