The main goal of the smart agricultural system is to automatically monitor farmland through automatic irrigation and pest detection framework.Traditional agricultural methods have low crop yields and require a lot of manpower.Therefore,this paper proposes a scheme of smart agricultural monitoring system based on the Internet of Things.The main function of the system is automatic irrigation and plant disease detection,using machine learning algorithm to accurately predict the amount of water needed in farmland,and automatically identify pests according to the needs of farmland.The pest detection module uses proximity algorithm and support vector machine learning algorithm to accurately predict plant diseases.Extracting convenient features from plant leaves and then using these features for classification is helpful to detect whether plants are infected with insect pests.The system monitors,analyzes,evaluates and controls farmland to realize automatic irrigation of water and identification of plant diseases.The machine learning algorithm is numerically analyzed,and the accuracy of classification is up to 84%.
deep learningneural networkInternet of Thingsclassificationfeature extraction
国家科技期刊平台
NSTL
万方数据
维普
