Fertilization and topdressing intelligent decision-making system and method based on rice open space sensing data

Abstract

The invention discloses a fertilization and topdressing intelligent decision-making system and method based on rice open space sensing data. The method comprises the following steps: acquiring agricultural condition data through a multi-source data acquisition module formed by an unmanned aerial vehicle multispectral camera, a ground sensor network and a meteorological station; generating a standardized data set by adopting a space-time fusion algorithm; the CNN-LSTM hybrid model is used for realizing rice growth prediction, disease, pest and weed identification and maturity evaluation; dynamically generating a base fertilizer / topdressing prescription map, a pesticide application scheme and an agricultural machine path plan in combination with an agricultural rule base; and the intelligent agricultural machine is linked through the Internet of Things interface to execute accurate operation. Millimeter-level sensing of the farmland state, variable control over fertilizer application and pesticide application and autonomous execution of agricultural machine operation are achieved, and compared with a traditional mode, the fertilizer utilization rate is increased by 35% or above, the pesticide application amount is reduced by 40%, and the comprehensive yield is increased by 15-20%.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent agriculture, and particularly relates to an intelligent decision-making system and method for fertilization and topdressing based on rice field and aerial perception data. Background Art

[0002] Traditional decision-making for agricultural fertilization and topdressing mostly relies on manual experience or static soil test data, and has the following problems:

[0003] Poor data timeliness: The traditional soil sampling cycle is long, and it is difficult to timely reflect the dynamic growth requirements of rice;

[0004] Lagged identification of diseases, pests and weeds: Relying on manual inspections, it is difficult to achieve early warning;

[0005] Rough agricultural machinery operations: Fertilization and pesticide application lack precise prescription guidance, resulting in waste of resources and environmental pollution.

[0006] Therefore, how to provide an intelligent decision-making system and method for fertilization and topdressing based on rice field and aerial perception data to solve the problems existing in the prior art is of great significance for its application. Summary of the Invention

[0007] In view of this, the purpose of this application is to provide an intelligent decision-making system and method for fertilization and topdressing based on rice field and aerial perception data to solve the problems of poor data timeliness, lagged identification of diseases, pests and weeds, and rough agricultural machinery operations existing in traditional agricultural fertilization and topdressing decision-making, which mostly rely on manual experience or static soil test data.

[0008] To achieve the above purpose, the present invention provides the following technical solutions:

[0009] An intelligent decision-making system for fertilization and topdressing based on rice field and aerial perception data, characterized by including the following modules:

[0010] Data acquisition module: Used to obtain multi-source agricultural situation data, including unmanned aerial vehicle remote sensing images, satellite remote sensing data, soil physical and chemical parameters, rice growth parameters and meteorological data collected by ground sensors;

[0011] Data processing and analysis module: Perform spatio-temporal fusion, feature extraction and outlier correction on the multi-source data to generate a standardized agricultural situation data set;

[0012] Agricultural situation dynamic prediction module: Based on deep learning and time series analysis algorithms, construct a rice growth state prediction model, a diseases, pests and weeds identification model and a maturity evaluation model, and dynamically output rice growth trend parameters, diseases, pests and weeds distribution maps and maturity prediction results;

[0013] Prescription map generation module: Combine the agronomic requirement rule base and generate soil base fertilizer prescription maps, topdressing prescription maps, pest, disease, weed control prescription maps, and lodging risk warning maps according to the prediction results;

[0014] Intelligent decision-making execution module: Link the prescription maps with agricultural machinery equipment to achieve variable fertilization, precise pesticide application, and agricultural machinery path planning.

[0015] As a preferred technical solution of the present invention, the data acquisition module includes:

[0016] A multi-spectral camera carried by a drone, used to obtain the spectral reflectance and vegetation index of the rice canopy;

[0017] A ground sensor network, which real-time monitors soil humidity, pH value, conductivity, and nitrogen, phosphorus, and potassium content;

[0018] A weather station device, which collects temperature, humidity, rainfall, and light intensity data.

[0019] As a preferred technical solution of the present invention, the agricultural situation dynamic prediction module adopts a hybrid model combining a convolutional neural network (CNN) and a long short-term memory network (LSTM), which is used to identify the risks of pests, diseases, and weeds in the rice growth stage and predict the maturity.

[0020] As a preferred technical solution of the present invention, the prescription map generation module dynamically generates prescription maps for fertilization amount, pesticide application amount, and agricultural machinery operation path according to the threshold rules in the agronomic requirement rule base.

[0021] As a preferred technical solution of the present invention, the intelligent decision-making execution module transmits the prescription map to the intelligent agricultural machinery equipment through the Internet of Things interface to achieve precise operation.

[0022] An intelligent decision-making method for fertilization and topdressing based on rice air-ground perception data, including the following steps:

[0023] A. Obtain air-ground perception data during the rice growth cycle through multi-source data acquisition technology;

[0024] B. Perform spatio-temporal alignment and fusion on the data, and extract the spectral characteristics of the rice canopy, the distribution characteristics of soil nutrients, and the environmental parameter characteristics;

[0025] C. Use a convolutional neural network (CNN) and a long short-term memory network (LSTM) to construct a time-series dynamic prediction model, identify the risks of pests, diseases, and weeds in the rice growth stage, and predict the maturity;

[0026] D. Combine the threshold rules in the agronomic knowledge base to dynamically generate prescription maps for fertilization amount, pesticide application amount, and agricultural machinery operation path;

[0027] E. Transmit the prescription map to the intelligent agricultural machinery equipment through the Internet of Things interface to achieve precise operation.

[0028] As a preferred technical solution of the present invention, the multi-source data acquisition technology includes unmanned aerial vehicle remote sensing images, satellite remote sensing data, soil physical and chemical parameters and meteorological data collected by ground sensors.

[0029] As a preferred technical solution of the present invention, the time-series dynamic prediction model adopts a hybrid model combining a convolutional neural network (CNN) and a long short-term memory network (LSTM) to identify the risks of diseases, pests and weeds in the rice growth stage and predict the maturity.

[0030] As a preferred technical solution of the present invention, the prescription map generation step dynamically generates prescription maps for fertilization amount, pesticide application amount and agricultural machinery operation path according to the threshold rules in the agronomic requirement rule base.

[0031] As a preferred technical solution of the present invention, the precise operation step transmits the prescription map to the intelligent agricultural machinery equipment through the Internet of Things interface to achieve variable fertilization, precise pesticide application and agricultural machinery path planning.

[0032] Compared with the prior art, the beneficial effects of the present invention are:

[0033] The present invention obtains agricultural situation data through a multi-source data acquisition module composed of an unmanned aerial vehicle multi-spectral camera, a ground sensor network and a weather station; generates a standardized data set by using a spatio-temporal fusion algorithm; uses a CNN-LSTM hybrid model to realize rice growth trend prediction, disease, pest and weed identification and maturity assessment; dynamically generates base fertilizer / top dressing prescription maps, pesticide application plans and agricultural machinery path planning in combination with an agronomic rule base; and links intelligent agricultural machinery through the Internet of Things interface to execute precise operation. The present invention realizes millimeter-level perception of the farmland state, variable control of fertilization and pesticide application, and autonomous execution of agricultural machinery operation, improves the fertilizer utilization rate by more than 35% compared with the traditional method, reduces the pesticide dosage by 40%, and increases the comprehensive yield by 15-20%.

[0034] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, so as to be implemented in accordance with the content of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the following takes the preferred embodiments of the present application and combines with the drawings to describe in detail as follows.

[0035] Those skilled in the art will understand the above and other purposes, advantages and features of the present application more clearly according to the following detailed description of the specific embodiments of the present application in conjunction with the drawings. Description of the Drawings

[0036] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on these drawings. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts do not necessarily draw according to the actual scale.

[0037] Figure 1 It is the system block diagram of the present invention;

[0038] Figure 2 It is the method flowchart of the present invention. Detailed implementation manners

[0039] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. In the following description, specific details such as specific configurations and components are provided only to help a comprehensive understanding of the embodiments of the present application. Therefore, those skilled in the art should clearly understand that various changes and modifications can be made to the embodiments described here without departing from the scope and spirit of the present application. In addition, for the sake of clarity and conciseness, the description of known functions and structures is omitted in the embodiments.

[0040] In addition, the present application may repeat reference numerals and / or letters in different examples. This repetition is for the purpose of simplicity and clarity, and does not in itself indicate the relationship between the various embodiments and / or settings discussed.

[0041] The term "and / or" in this article is only a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, B exists alone, and A and B exist simultaneously. The term " / and" in this article describes another association object relationship, indicating that two relationships may exist. For example, A / and B may represent: A exists alone, and A and B exist alone. In addition, the character " / " in this article generally represents that the associated objects before and after are an "or" relationship.

[0042] It should also be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise", or any other variant thereof are intended to cover non-exclusive inclusion.

[0043] Please refer to Figure 1-2 , the present invention provides a technical solution for an intelligent decision-making system and method for fertilization and topdressing based on rice field and empty space perception data: 1. An intelligent decision-making system for fertilization and topdressing based on rice field and empty space perception data, characterized by including the following modules:

[0044] Data acquisition module: used to obtain multi-source agricultural situation data, including UAV remote sensing images, satellite remote sensing data, soil physical and chemical parameters, rice growth parameters and meteorological data collected by ground sensors;

[0045] Data processing and analysis module: perform spatio-temporal fusion, feature extraction and outlier correction on multi-source data to generate a standardized agricultural situation dataset;

[0046] Agricultural situation dynamic prediction module: based on deep learning and time series analysis algorithms, construct a rice growth status prediction model, a pest and disease identification model and a maturity evaluation model, and dynamically output rice growth trend parameters, pest and disease distribution maps and maturity prediction results;

[0047] Prescription map generation module: combine with the agronomic requirement rule base to generate a soil base fertilizer prescription map, a topdressing prescription map, a pest and disease control prescription map and a lodging risk warning map according to the prediction results;

[0048] Intelligent decision-making execution module: link the prescription map with agricultural machinery equipment to achieve variable fertilization, precise pesticide application and agricultural machinery path planning.

[0049] The data acquisition module includes:

[0050] A multi-spectral camera carried by a UAV, used to obtain the spectral reflectance and vegetation index of the rice canopy;

[0051] A ground sensor network, which real-time monitors soil humidity, pH value, conductivity and nitrogen, phosphorus and potassium content;

[0052] A weather station device, which collects temperature, humidity, rainfall and light intensity data.

[0053] The agricultural situation dynamic prediction module adopts a hybrid model combining a convolutional neural network (CNN) and a long short-term memory network (LSTM) to identify the pest and disease risks in the rice growth stage and predict the maturity.

[0054] The prescription map generation module dynamically generates prescription maps for fertilization amount, pesticide application amount and agricultural machinery operation path according to the threshold rules in the agronomic requirement rule base.

[0055] The intelligent decision-making execution module transmits the prescription map to the intelligent agricultural machinery equipment through the Internet of Things interface to achieve precise operation.

[0056] An intelligent decision-making method for fertilization and topdressing based on rice field and airspace perception data, comprising the following steps:

[0057] A. Obtain field and airspace perception data during the rice growth cycle through multi-source data acquisition technology;

[0058] B. Align and fuse the data in time and space, and extract the spectral characteristics of the rice canopy, the distribution characteristics of soil nutrients, and the environmental parameter characteristics;

[0059] C. Use a convolutional neural network (CNN) and a long short-term memory network (LSTM) to construct a time-series dynamic prediction model to identify the risks of diseases, pests and weeds in the rice growth stage and predict the maturity;

[0060] D. Combine the threshold rules in the agronomic knowledge base to dynamically generate prescription maps for fertilization amount, pesticide application amount and agricultural machinery operation path;

[0061] E. Transmit the prescription map to intelligent agricultural machinery equipment through the Internet of Things interface to achieve precise operation.

[0062] The multi-source data acquisition technology includes unmanned aerial vehicle remote sensing images, satellite remote sensing data, soil physical and chemical parameters and meteorological data collected by ground sensors.

[0063] The time-series dynamic prediction model adopts a hybrid model combining a convolutional neural network (CNN) and a long short-term memory network (LSTM) to identify the risks of diseases, pests and weeds in the rice growth stage and predict the maturity.

[0064] The prescription map generation step dynamically generates prescription maps for fertilization amount, pesticide application amount and agricultural machinery operation path according to the threshold rules in the agronomic requirement rule base.

[0065] The precise operation step transmits the prescription map to intelligent agricultural machinery equipment through the Internet of Things interface to achieve variable fertilization, precise pesticide application and agricultural machinery path planning.

[0066] The following is a detailed description in combination with embodiments:

[0067] Embodiment 1 (Multi-source Data Acquisition and Fusion System)

[0068] This embodiment shows the specific implementation method of the data acquisition module:

[0069] 1.1 Configure a DJI M300RTK unmanned aerial vehicle equipped with a RedEdge-MX multispectral camera (band range 475 - 840nm), with a flight altitude of 100m and a resolution of 8cm / pixel to obtain vegetation indices such as NDVI and NDRE;

[0070] 1.2 Deploy a soil sensor network (Decagon 5TE series), arranged in a 20m×20m grid, to monitor the soil humidity (±3% accuracy), pH value (±0.5 accuracy), and electrical conductivity (±5% accuracy) in the soil layer of 0 - 30cm in real time;

[0071] 1.3 Install a weather station (Davis Vantage Pro2) to collect temperature (±0.5℃), rainfall (±5%), and photosynthetically active radiation (0 - 3000 μmol / m 2 / s);

[0072] 1.4 Use the STARFM algorithm to fuse Sentinel - 2 satellite data (10m resolution) and UAV data, and generate a daily - scale farmland data cube after spatio - temporal alignment.

[0073] Example 2 (Precision Fertilization System for Rice)

[0074] This example demonstrates the full - cycle intelligent decision - making process:

[0075] 2.1 Seven days before transplanting: Fuse soil sampling data (0.5 mu per point) and electromagnetic induction instrument (EM38) data through Kriging interpolation method to generate a basal fertilizer prescription map;

[0076] 2.2 Tillering stage: Use the U - Net model to segment UAV thermal infrared images (FLIR A65) to identify low - temperature stress areas and dynamically adjust the tillering fertilizer ratio;

[0077] 2.3 Heading stage: Combine the canopy temperature - transpiration model to calculate nitrogen use efficiency and generate a panicle fertilizer regulation plan;

[0078] 2.4 All prescription maps are used to guide the Kubota SPV75 transplanter to perform variable operations through the AgOpenGPS system.

[0079] The above are only the preferred embodiments of the present invention, and it does not limit the protection scope of the present invention accordingly. For those skilled in the art, the present invention can have various changes and modifications. All changes, modifications, substitutions, integrations, and parameter changes made to these embodiments through conventional substitutions or capable of achieving the same functions without departing from the principle and spirit of the present invention fall within the protection scope of the present invention.

Claims

1. An intelligent decision-making system for fertilization and topdressing based on rice empty space perception data, characterized in that, It includes the following modules: Data acquisition module: used to obtain multi-source agricultural situation data, including UAV remote sensing images, satellite remote sensing data, soil physical and chemical parameters, rice growth parameters and meteorological data collected by ground sensors; Data processing and analysis module: performs spatio-temporal fusion, feature extraction and outlier correction on the multi-source data to generate a standardized agricultural situation dataset; Agricultural situation dynamic prediction module: based on deep learning and time series analysis algorithms, constructs a rice growth status prediction model, a pest and disease identification model and a maturity assessment model, and dynamically outputs rice growth trend parameters, pest and disease distribution maps and maturity prediction results; Prescription map generation module: combines an agronomic requirement rule base to generate soil base fertilizer prescription maps, top dressing prescription maps, pest and disease control prescription maps and lodging risk warning maps according to the prediction results; Intelligent decision-making execution module: links the prescription maps with agricultural machinery equipment to achieve variable fertilization, precise pesticide application and agricultural machinery path planning.

2. The intelligent decision-making system for fertilization and topdressing based on rice empty space perception data according to claim 1, characterized in that: The data acquisition module includes: A multi-spectral camera carried by a UAV, used to obtain rice canopy spectral reflectance and vegetation indices; A ground sensor network that real-time monitors soil humidity, pH value, conductivity and nitrogen, phosphorus and potassium content; Meteorological station equipment that collects temperature, humidity, rainfall and light intensity data.

3. The intelligent decision-making system for fertilization and topdressing based on rice vacant land perception data according to claim 1, characterized in that: The agricultural situation dynamic prediction module adopts a hybrid model combining a convolutional neural network (CNN) and a long short-term memory network (LSTM) to identify the pest and disease risks in the rice growth stage and predict the maturity.

4. The intelligent decision-making system for fertilization and topdressing based on rice empty space perception data according to claim 1, characterized in that: The prescription map generation module dynamically generates prescription maps for fertilization amount, pesticide application amount and agricultural machinery operation path according to the threshold rules in the agronomic requirement rule base.

5. The intelligent decision-making system for fertilization and topdressing based on rice empty space perception data according to claim 1, wherein: The intelligent decision-making execution module transmits the prescription map to intelligent agricultural machinery equipment through an Internet of Things interface to achieve precise operation.

6. An intelligent decision-making method for fertilization and topdressing based on rice empty space perception data, characterized in that: It includes the following steps: A. Obtain air-ground perception data during the rice growth cycle through multi-source data acquisition technology; B. Perform spatio-temporal alignment and fusion on the data, and extract rice canopy spectral characteristics, soil nutrient distribution characteristics and environmental parameter characteristics; C. Use a convolutional neural network (CNN) and a long short-term memory network (LSTM) to construct a time series dynamic prediction model to identify the pest and disease risks in the rice growth stage and predict the maturity; D. Combine the threshold rules in the agronomic knowledge base to dynamically generate prescription maps for fertilization amount, pesticide application amount and agricultural machinery operation path; E. Transmit the prescription map to intelligent agricultural machinery equipment through an Internet of Things interface to achieve precise operation.

7. The intelligent decision-making method for fertilization and topdressing based on rice empty space perception data according to claim 1, characterized in that: The multi-source data acquisition technology includes UAV remote sensing images, satellite remote sensing data, soil physical and chemical parameters collected by ground sensors and meteorological data.

8. The intelligent decision-making method for fertilization and topdressing based on rice empty space perception data according to claim 1, wherein: The time series dynamic prediction model adopts a hybrid model combining a convolutional neural network (CNN) and a long short-term memory network (LSTM) to identify the pest and disease risks in the rice growth stage and predict the maturity.

9. The intelligent decision-making method for fertilization and topdressing based on rice empty space perception data according to claim 1, wherein: The prescription map generation step dynamically generates prescription maps for fertilization amount, pesticide application amount and agricultural machinery operation path according to the threshold rules in the agronomic requirement rule base.

10. The intelligent decision-making method for fertilization and topdressing based on rice vacant land perception data according to claim 1, wherein: The precise operation step transmits the prescription map to intelligent agricultural machinery equipment through an Internet of Things interface to achieve variable fertilization, precise pesticide application and agricultural machinery path planning.

Images