Citrus shoot early warning platform and method based on time sequence prediction
A time-series prediction and new shoot technology, which is applied in the direction of forecasting, fertilization methods, sub-station devices, etc., can solve the problems of damage to mesophyll, different effects and inaccuracy of citrus new shoots, etc., to achieve precise pesticide application and unmanned management Effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0032] like Figure 1-4 A citrus sprout early warning platform and method based on time series prediction is shown, including an edge information perception system, a communication network, and a platform server; wherein, the edge information perception system includes several cameras, edge computing equipment, and soil fertility sensors; the communication network includes wireless Communication, local area network communication, edge information perception system transmits data to platform server through communication network. , new shoot forecast and early warning.
[0033] The agricultural information perception system is deployed according to the five-point sampling method or the division of farmland areas. The agricultural situation perception system includes several agricultural situation perception sub-nodes, such as figure 1 As shown in the figure, each agricultural situation perception sub-node is a single whole, and is only responsible for the agricultural situatio...
Embodiment 2
[0046] like Figure 1-4 As shown in the figure, the present invention provides a method and platform for early warning of citrus shoots based on time series prediction. Decode and identify the information, obtain the identification results of citrus shoots, and package and transmit the agricultural situation perception sub-node number, video or image information, identification results, and soil fertility information to the remote platform server.
[0047] The platform server accepts the incoming data from the agricultural situation perception sub-node at the port, disassembles it according to the data source and data type, stores it in the database, and compares the proportion of new shoots with the pre-set early warning threshold. The obtained soil fertility data, seasons and other information are used for early warning of citrus shoots.
[0048] When monitoring personnel log in to the platform, the server reads the data information stored in the database for front-end disp...
Embodiment 3
[0054] This example takes the early warning monitoring of citrus summer shoots as an example, such as figure 2 The agricultural situation perception sub-nodes shown are deployed in the farmland in the form of five-point sampling method according to the division of the farmland area, such as figure 1 The video image data signal collected by the camera is transmitted to the edge computing platform through the network cable. The edge computing device uses its own computing power to decode the video signal, and calls the deep learning model for target detection and classification, and obtains the budding stage, the growth stage, and the mature stage. According to the number of new shoots in the period, the classified data is encapsulated in JSON format. The soil fertility sensor in this area communicates with the edge computing device through the network cable, and the edge computing device encapsulates the data into the JSON data format. The edge computing device transmits data ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


