Rice Field Pest And Disease Identification System Based On Convolutional Neural Networ

Rice,as an important economic crop in the world,will inevitably encounter diseases and insect pests in the natural growth process.Its diseases and insect pests will directly affect the quality and yield of crops and bring huge economic losses to farmers.The traditional identification of artificial rice pests and diseases relies heavily on professionals to identify them on the rice field.This method consumes manpower and material resources,and is subjective,which is not conducive to disaster prevention and control.Therefore,the combination of computer technology and disease and pest identification can overcome the disadvantages brought by traditional artificial identification.Deep learning is a popular technology at present,and convolutional neural networks(CNN)is one of the important branches of deep learning.Its algorithm maturity is high,and there are many examples of application in agriculture.Therefore,this paper selects CNN as the basic network and proposes a rice field disease and pest identification system based on CNN,which has high practical value and good application prospects.The main research contents of this paper are as follows:(1)In view of the poor performance of traditional CNN in the identification of diseases and insect pests,this paper selects Res Net50 network model as the basic network model architecture of this paper,and introduces the attention mechanism to improve Res Net50,and uses the improved network model to compare with Res Net50 and other six comparison algorithms on the rice disease and insect pests data set,The results show that our proposed algorithm model significantly improves the accuracy of rice pest identification.(2)In view of the lack of professional agricultural knowledge of the general public and the lack of agricultural experts,this paper designs and develops a Res Net50 network model based on the attention mechanism,and designs and implements a system based on this model.Through this system,the general public can use Internet devices such as mobile phones to accurately prevent and identify rice diseases and insect pests in the paddy field.At the same time,for the idea of relevant researchers using the system,It is of great significance to extend it to other fields of crop pest identification.