Design And Implementation Of Intelligent Vision Monitoring System For Intelligent Agriculture

China is a big agricultural country,and with the continuous advancement of science technology,the development of facilities agriculture is becoming more and more rapid.Smart agriculture will surely become the trend of the development of modern agriculture in the future.As an important part of modern agriculture,the monitoring and control of agricultural environment is an effective means to grasp the information of agricultural environment and the growth states of crops.In the process of agricultural production,natural factors affect the growth of crops,such as soil temperature and humidity,ambient temperature and humidity,rainfall,wind speed,and CO₂ concentration.Extensive traditional agricultural management methods cannot achieve refined and intelligent management of agricultural environmental information.The application of modern industrial control technology,wireless communication technology,and embedded technology to agricultural production can improve the agricultural production management mode and the agricultural production efficiency.A set of intelligent visual monitoring system for intelligent agriculture is designed in this paper.The system uses various sensors to collect data of temperature,humidity,CO₂concentration and light intensity in different planting areas,and sends data to coordinator nodes through ZigBee wireless sensor network.The coordinator node controls the peripherals and communicates with the embedded main controller through the RS232serial port.The embedded main controller side access to plant video information through the camera,the H.264 encoding for video compression,using TCP protocol and PC server to connect,with clonal selection algorithm and K neighbor classifier blade recognition algorithm based on the results shown in the PC interface design based on Qt.According to the recognition result,different control strategies are invoked,the modular management is implemented based on C++programming,and the data storage and historical query functions are implemented by transplanting SQLite database.The system has many characteristics,such as intelligence,high recognition rate,low cost and easy management and so on.Through the test,the system's environmental information acquisition and video image acquisition can be effectively displayed on the host computer,and through the host computer to send control instructions to control the external implementation agencies,This system can also achieve database storage and history query function.The designed plant leaf recognition classifier has great identification ability and can effectively identify plant species in different planting areas.So,this system has certain practical significance and promotion significance in the field of smart agriculture.