Search

CN-121999356-A - Intelligent identification method and equipment for plant diseases and insect pests of facility agriculture

CN121999356ACN 121999356 ACN121999356 ACN 121999356ACN-121999356-A

Abstract

The invention discloses an intelligent identification method and equipment for plant diseases and insect pests of facility agriculture, comprising the following steps of collecting data, wherein cameras and small weather stations are arranged in a plurality of greenhouses in the facility agriculture, the cameras are responsible for capturing pictures of crops, the weather stations are responsible for collecting environmental parameters, the data can be uploaded to a server through a network, the network comprises a network and a wireless network (WiFi, 4G, bluetooth and the like), the data is uploaded to the server by adopting http and mqtt protocols, the pictures are uploaded by adopting the http, the weather station data is uploaded by adopting the mqtt protocol, the cameras and the small weather stations are arranged in a plurality of greenhouses in the collected data, the cameras are responsible for capturing the pictures of the crops, the weather stations are responsible for capturing the environmental parameters, the data can be uploaded to the server through the network, the original equipment is used for collecting the data, the cost is not increased for the greenhouses, and a large amount of data is also obtained as training samples.

Inventors

  • YANG LINNAN
  • GAO LUTAO
  • ZHANG LILIAN
  • PENG LIN

Assignees

  • 云南农业大学

Dates

Publication Date
20260508
Application Date
20240322

Claims (6)

  1. 1. An intelligent identification method for plant diseases and insect pests of facility agriculture comprises the following steps: (1) The data acquisition is that cameras and small weather stations are arranged in a plurality of greenhouses in facility agriculture, the cameras are responsible for capturing pictures of crops, the weather stations are responsible for acquiring environmental parameters, an intelligent spore capturing instrument monitors the stock of disease spores and the diffusion dynamics of the disease spores, provides data for predicting and preventing epidemic and infection of diseases, an intelligent pest situation measuring and reporting lamp monitors and pre-warns the ecology of pests, traps and photographs the pests, and the data can be uploaded to a server through a network; Where the network includes a network and a wireless network (WiFi, 4G, bluetooth, etc.). (2) Uploading data to a server, wherein the data is uploaded to the server by adopting http and mqtt protocols; The data uploaded to the server (2) also comprises all positions of the greenhouse and other relevant information. (3) Manually marking training data, and inputting data by the model; The data annotation and training model (3) comprises pictures of crops, meteorological data, weather data, which are used to train the model. (4) And (3) model release, namely after data marking and training the model (3), training to obtain a plant disease and insect pest identification model by using a deep learning method, testing the trained model by using an AI (advanced technology) device, and releasing the model into the device formally after the plant disease and insect pest identification accuracy reaches 90%. (5) And (3) finishing output, namely after the model release (4) is finished, the AI equipment collects pictures, meteorological data and weather data of crops, and whether the crops have diseases and insect pests or not is identified through the model. (6) And in the model release (4), the equipment periodically reads the model information from the server, and when a new model is found, the latest model is downloaded from the server to the AI equipment to replace the old model, and the new model is used for identifying plant diseases and insect pests, so that the accuracy is effectively improved.
  2. 2. An apparatus comprising the intelligent identification method for plant diseases and insect pests of claim 1, wherein the data are marked and trained, the meteorological data in the model (3) are acquired as ground climate data month data of a data center of a meteorological office, discrete point data are obtained, and the point data are converted into surface data by adopting kriging interpolation.
  3. 3. The intelligent identification device for plant diseases and insect pests of claim 2, wherein the data acquisition (1) comprises surface temperature and soil moisture.
  4. 4. The intelligent identification device for plant diseases and insect pests of the facility agriculture according to claim 2, wherein the output (5) is finished, if the existing plant diseases and insect pests are identified, the information of the plant diseases and insect pests is displayed in the LED large screen, meanwhile, the information of the plant diseases and insect pests is uploaded to the server, the server can inform a greenhouse planter, and the information is transmitted to relevant specialists for identification and authentication. The expert will again confirm the pest information and add it to the server training.
  5. 5. The intelligent identification device for plant diseases and insect pests according to claim 2, wherein the data is uploaded to the server (2) and the pictures are uploaded by http, and the weather station data is uploaded by mqtt protocol.
  6. 6. The intelligent recognition device for the plant diseases and insect pests of the facility agriculture according to claim 2, wherein the meteorological data of the data labeling and training model (3) comprise parameters such as greenhouse temperature and humidity, soil parameters, greenhouse external wind speed and the like.

Description

Intelligent identification method and equipment for plant diseases and insect pests of facility agriculture Technical Field The invention relates to the technical field of plant protection, in particular to a method, a device, equipment and a storage medium for identifying plant diseases and insect pests. Background The existing agricultural facility planting greenhouse is provided with a camera, a small weather station (for collecting parameters such as temperature and humidity, wind speed, oxygen content of soil and the like of surrounding environment) and an LED large screen (for displaying environmental parameter information), and meanwhile, the LED large screen also acquires weather information from a remote server (weather data are acquired from a national weather information center) and displays the weather information. Based on the existing hardware and acquisition parameters, the current equipment is only used for security protection and equipment management, the system cannot identify the plant diseases and insect pests of crops planted in a greenhouse, needs to manually observe on site and identify or take pictures to transmit to an expert to identify the plant diseases and insect pests, consumes a large amount of manpower resources to identify the plant diseases and insect pests, cannot quickly identify the plant diseases and insect pests, has time delay, and can cause serious loss if not handled in time. Therefore, we propose a method and a device for intelligently identifying plant diseases and insect pests in facility agriculture to solve the problems. Disclosure of Invention The invention aims to provide a method and equipment for intelligently identifying plant diseases and insect pests of facility agriculture, which are used for solving the problems that the prior hardware and acquisition parameters are provided in the background technology, the prior equipment is only used for security protection and equipment management, a system cannot identify plant diseases and insect pests of crops planted in a greenhouse, the plant diseases and insect pests of crops need to be manually observed on site, identified or photographed and transmitted to an expert to identify the plant diseases and insect pests, a large amount of manpower resources are consumed to identify the plant diseases and insect pests, the plant diseases and insect pests cannot be quickly identified, and delay exists, and if the plant diseases and insect pests are not timely processed, significant loss can be caused. In order to achieve the above purpose, the present invention provides the following technical solutions: a method and equipment for intelligently identifying plant diseases and insect pests of facility agriculture comprise the following steps: The data acquisition is that cameras and small weather stations are arranged in a plurality of greenhouses in facility agriculture, the cameras are responsible for capturing pictures of crops, the weather stations are responsible for acquiring environmental parameters, an intelligent spore capturing instrument monitors the stock of disease spores and the diffusion dynamics of the disease spores, provides data for predicting and preventing epidemic and infection of diseases, an intelligent pest situation measuring and reporting lamp monitors and pre-warns the ecology of pests, traps and photographs the pests, and the data can be uploaded to a server through a network; Where the network includes a network and a wireless network (WiFi, 4G, bluetooth, etc.). Uploading data to a server, wherein the data is uploaded to the server by adopting http and mqtt protocols; the data uploaded to the server also includes all the positions of the greenhouse and other relevant information. Manually marking training data, and inputting data by the model; the data annotates and trains the model including pictures of the crop, meteorological data, weather data, which is used to train the model. And (3) model release, namely after data marking and training the model, training the model by using a deep learning method to obtain a plant disease and insect pest identification model, putting the trained model into an AI (advanced technology) equipment for testing, and releasing the model into the equipment formally after the plant disease and insect pest identification accuracy reaches 90%. And (3) finishing outputting, namely acquiring pictures, meteorological data and weather data of crops by the AI equipment after the model is released, and identifying whether the crops have diseases and insect pests or not by the model. In the model release, the equipment periodically reads the model information from the server, when a new model is found, the latest model is downloaded from the server to the AI equipment, the old model is replaced, and the new model is used for identifying plant diseases and insect pests, so that the accuracy is effectively improved. Furthermore, the data are marked and the meteorolog