Research And Design Of Intelligent Spraying Management System For Vegetable Greenhouses Based On Image Recognition

China is the world’s largest vegetable growing country,and the vegetable farming has gradually become one of the ways for farmers to get rid of poverty and get rich.In the process of vegetable planting,there are many kinds of pests and diseases,and the occurrence rules are complicated.Most farmers mainly rely on subjective judgments for pest control,and the high false positive rate is unable to timely and accurately control pests and diseases,which directly affects the quality of vegetables and causes huge economic losses.With the rapid development of agricultural science and technology,greenhouses are required to provide a more scientific growth environment for vegetable growth,and intelligent fertilization is carried out according to the nutrients required for different growth stages of different vegetables.Based on the above problems,this paper researches and develops a greenhouse intelligent spray management system based on image recognition.The system can help the vegetable farmers to automatically identify and control vegetable pests and diseases in a timely and accurate manner.It also has the intelligent configuration function of nutrient solution,and intelligently sprays the nutrient solution according to different growth stages of different vegetables.which solves the problems of fertilization,low efficiency in pests removal,labor intensity and waste of resources,then improves the quality of vegetables.In this paper,the research on a greenhouse intelligent spray management system based on image recognition is divided into three sections : Identification and prevention of pests and diseases、Intelligent spraying of nutrient solution and Environmental monitoring.Identification and prevention of pests and diseases: This section mainly realizes the functions of image acquisition,image pre-processing,feature extraction,classification and pest control of tomato leaves.For the tomato image,which is susceptible to natural light and angle,the image is pre-processed by color space transformation and median filtering algorithm.Then the histogram feature extraction algorithm is used to extract the color and texture features of tomato pests and diseases images and generate corresponding feature vectors.Finally,the K-nearest neighbor algorithm based on weighted distance is used to classify and identify the tomato pest and disease image,and the final diagnosis result is converted into a control command and sent to the main controller to realize the pest control function.Intelligent spraying of nutrient solution: This section includes intelligent configuration of nutrient solution and intelligent spray.Each tank is equipped with a flow sensor and an adjustable opening electric valve.The ratio control system is used to strictly control the batchingaccuracy and realize the intelligent configuration of the nutrient solution.The intelligent spraying part is intelligently sprayed according to the different nutrient solution required for different growth stages of different vegetables.Environmental monitoring part: This section mainly collects information on internal environmental parameters such as air temperature and humidity,light intensity and carbon dioxide concentration in greenhouses.And then transferring the information to the server for data analysis and storage through NB-IoT wireless communication module.Users can query the internal environmental parameters of the greenhouse through the WEB interface and display screen to realize real-time monitoring of the greenhouse environment.In addition,the user can control the electrical equipment inside the greenhouse and maintain the dynamic balance of the vegetable growth environment.In summary,the intelligent spray management system studied in this paper provides great convenience for vegetable farmers lacking knowledge of pest control,improves the recognition rate of vegetable pests and diseases,timely and accurately sprays prevention and control,and can be intelligently sprayed according to the nutrient solution required for different growth stages of different vegetables,which has important research significance for realizing intelligent spray management of greenhouses.