CN-121980286-A - Corn full life cycle monitoring method and system based on air-ground coordination

CN121980286ACN 121980286 ACN121980286 ACN 121980286ACN-121980286-A

Abstract

The invention is applicable to the technical field of agricultural monitoring, and particularly relates to a corn full life cycle monitoring method and system based on air-ground coordination, wherein the method comprises the steps of deducing the diameter of a stalk and the top plane coordinate through unmanned aerial vehicle output; the method comprises the steps of obtaining measured stalk diameter and base plane coordinates thereof through a ground robot, constructing a first plant distribution matrix and a second plant distribution matrix, matching the first plant distribution matrix with the second plant distribution matrix, and associating top monitoring data collected by an unmanned aerial vehicle with side monitoring data collected by the ground robot to the same corn plant according to a matching result. According to the invention, through matching based on the relative position relation between the plant stalk diameter and the group, the problem of data dislocation caused by positioning drift of the unmanned aerial vehicle and the ground robot is solved, expensive high-precision positioning equipment is not needed, a reliable data basis is provided for planting decisions such as variable fertilization and precise pesticide spraying, and the accuracy of precise corn management is improved.

Inventors

LI JIAN
LIU XUE
LIU BO
SHANG XUELING
SI CHANGLIANG
YU WEILIN
ZHANG WEIJIAN
KANG JUNRUI
ZHAO JIAWEI

Assignees

吉林农业大学
吉林省华维智慧农业科技有限公司
吉林省水利科学研究院(吉林省水利科技推广总站、吉林省水利水电工程质量检测中心、吉林省灌溉试验中心站)

Dates

Publication Date: 20260505
Application Date: 20260408

Claims (10)

1. The corn full life cycle monitoring method based on air-ground cooperation is characterized by comprising the following steps of: Collecting multispectral images of corn fields from a top view angle through an unmanned aerial vehicle, detecting and outputting the inferred stalk diameter and the top plane coordinate of each corn plant; Constructing a first plant distribution matrix containing inferred stalk diameters and top plane coordinates of all plants based on data measured by the unmanned aerial vehicle; constructing a second plant distribution matrix containing measured stalk diameters and base plane coordinates of all plants based on data measured by the ground robot; Matching the first plant distribution matrix with the second plant distribution matrix according to the relative position relation between plants based on plane coordinates and deducing the numerical value corresponding relation between the diameter of the stalk and the diameter of the actually measured stalk; And according to the matching result, the top monitoring data collected by the unmanned aerial vehicle and the side monitoring data collected by the ground robot are related to the same corn plant.
2. The method for monitoring the full life cycle of corn based on air-ground coordination according to claim 1, wherein the step of constructing a first plant distribution matrix containing inferred stalk diameters and top plane coordinates of all plants based on unmanned aerial vehicle measurement data comprises the steps of: for unmanned aerial vehicle data, calculating Euclidean distances between each corn plant and coordinates of all other plants by taking top plane coordinates of each corn plant as nodes, and finding out K nearest neighbor plants to form a local topological subgraph taking the plant as a center; for ground robot data, executing the same operation as unmanned plane data, and constructing a local topological subgraph with the same K value for each plant based on the base plane coordinates; In each partial topological sub-graph, the diameter of a central plant, the distance vectors of the central plant and K neighbors and the diameter of the K neighbors are combined into a characteristic vector, and the characteristic vector sets of all plants form a first plant distribution matrix and a second plant distribution matrix.
3. The method for monitoring the full life cycle of corn based on air-ground cooperation according to claim 1, wherein the step of matching the first plant distribution matrix with the second plant distribution matrix according to the relative positional relationship between plants based on plane coordinates and the numerical correspondence between the inferred stalk diameter and the measured stalk diameter specifically comprises: Calculating the comprehensive similarity between any two partial topological sub-graph feature vectors in the first plant distribution matrix and the second plant distribution matrix, wherein the comprehensive similarity is the weighted sum of the central plant diameter difference and the cosine similarity of the two distance vectors; Using a graph matching algorithm, taking the comprehensive similarity as an edge weight, and searching for an optimal matching pair which maximizes the total similarity of the corresponding relations of all plants in the two matrixes; And performing geometric verification on the optimal matching pair, calculating an optimal rigid body transformation matrix for integrating the unmanned aerial vehicle coordinate system to the ground robot coordinate system by using the successfully matched plant pair, and generating a space-ground plant mapping table.
4. The method for monitoring the full life cycle of corn based on air-ground coordination according to claim 1, wherein the step of associating the top monitoring data collected by the unmanned aerial vehicle with the side monitoring data collected by the ground robot to the same corn plant according to the matching result specifically comprises the following steps: According to the space-to-ground plant mapping table, binding multispectral vegetation index data acquired by the unmanned aerial vehicle for each corn plant with ground surface data acquired by the ground robot for the corresponding plant; Correcting the top plane coordinates of the corresponding plants acquired by the unmanned aerial vehicle by using the base plane coordinates acquired by the ground robot as a reference and applying an optimal rigid transformation matrix to obtain the fused geographic positions of the plants; Binding the fused multisource data of each corn strain with the geographic position, and updating the multisource data into the same digital map to form a single-strain-scale full-growth-period monitoring file.
5. The method for monitoring the full life cycle of corn based on air-ground coordination according to claim 1, wherein the ground robot adopts a laser radar to measure the plant diameter.
6. A corn full life cycle monitoring system based on air-ground coordination, the system comprising: the system comprises a data acquisition module, a ground robot, a data acquisition module and a data acquisition module, wherein the data acquisition module is used for acquiring multispectral images of corn fields from a top view angle through the unmanned aerial vehicle, detecting and outputting the inferred stalk diameter and the top plane coordinate of each corn plant; The system comprises a matrix construction module, a ground robot measurement module, a matrix construction module, a first plant distribution matrix, a second plant distribution matrix and a control module, wherein the matrix construction module is used for constructing a first plant distribution matrix comprising inferred stalk diameters and top plane coordinates of all plants based on data measured by the unmanned aerial vehicle; The position matching module is used for matching the first plant distribution matrix with the second plant distribution matrix according to the relative position relation based on plane coordinates among plants and deducing the numerical value corresponding relation between the diameter of the stalk and the diameter of the actually measured stalk; And the data matching module is used for associating the top monitoring data collected by the unmanned aerial vehicle and the side monitoring data collected by the ground robot to the same corn plant according to the matching result.
7. The air-ground collaboration-based corn full life cycle monitoring system of claim 6, wherein the matrix construction module comprises: The aerial data topology unit is used for calculating Euclidean distances between each plant of corn and coordinates of all other plants by taking top plane coordinates of each plant of corn as nodes for unmanned aerial vehicle data, and finding out K nearest neighbor plants of each plant to form a local topology subgraph taking the plant as a center; the ground data topology unit is used for executing the same operation as unmanned aerial vehicle data on ground robot data, and constructing a local topology subgraph with the same K value for each plant based on the base plane coordinates; The matrix generation unit is used for combining the diameter of the central plant, the distance vectors of the central plant and the K neighbors and the diameter of the K neighbors in each partial topological subgraph into a characteristic vector, and the characteristic vector sets of all plants form a first plant distribution matrix and a second plant distribution matrix.
8. The air-ground collaboration-based corn full life cycle monitoring system of claim 6, wherein the location matching module comprises: The similarity calculation unit is used for calculating the comprehensive similarity between any two partial topological sub-graph feature vectors in the first plant distribution matrix and the second plant distribution matrix, wherein the comprehensive similarity is the weighted sum of the central plant diameter difference value and the cosine similarity of the two distance vectors; The plant matching unit is used for searching an optimal matching pair which maximizes the total similarity of the corresponding relations of all plants in the two matrixes by using the comprehensive similarity as an edge weight by using a graph matching algorithm; The mapping table construction unit is used for performing geometric verification on the optimal matching pair, calculating an optimal rigid body transformation matrix which is used for integrating the unmanned aerial vehicle coordinate system to the ground robot coordinate system by utilizing the successfully matched plant pair, and generating the space plant mapping table.
9. The air-ground collaboration-based corn full life cycle monitoring system of claim 6, wherein the data matching module comprises: The data binding unit is used for binding multispectral vegetation index data acquired by the unmanned aerial vehicle for each corn plant with ground surface data acquired by the ground robot for the corresponding plant according to the air-ground plant mapping table; The geographic position conversion unit is used for correcting the top plane coordinates of the corresponding plants acquired by the unmanned aerial vehicle by using the base plane coordinates acquired by the ground robot as a reference and applying an optimal rigid body transformation matrix to acquire the fused geographic position of the plants; And the archive construction unit is used for binding the fused multisource data of each corn strain with the geographic position, and updating the multisource data into the same digital map to form the whole growth period monitoring archive with a single plant size.
10. The air-ground synergy-based corn full life cycle monitoring system of claim 6, wherein the ground robot uses a lidar to measure plant diameter.

Description

Corn full life cycle monitoring method and system based on air-ground coordination Technical Field The invention belongs to the technical field of agricultural monitoring, and particularly relates to a corn full life cycle monitoring method and system based on air-ground coordination. Background The space-ground coordination is a three-dimensional operation mode of unmanned aerial vehicle, ground sensor and intelligent equipment, and the full-closed loop management of wide area sensing, fine diagnosis and accurate execution of a target object is realized by enabling platforms (space base, space base and foundation) with different space dimensions to perform their functions, data intercommunication and intelligent linkage under unified scheduling. The maneuvering high precision of the unmanned aerial vehicle is deeply fused with the real-time verification and execution capacity of the ground equipment, and finally an intelligent system of 'observation-analysis-decision-action' is formed. In the current corn growth monitoring process, the monitoring data of the same corn plant are obtained by the cooperative acquisition of an unmanned aerial vehicle and a ground robot, but the unmanned aerial vehicle or the ground robot has limited positioning precision, position drift and other conditions are easy to occur, so that dislocation occurs in air-ground data matching, the monitoring data of the corn plant are inaccurate, and the follow-up strategy formulation and execution are affected. Disclosure of Invention The invention aims to provide a corn full life cycle monitoring method based on air-ground coordination, which aims to solve the problems that whether an unmanned aerial vehicle or a ground robot is limited in positioning precision, position drift and the like are easy to occur, so that air-ground data matching is misplaced, monitoring data of a corn plant is inaccurate, and follow-up strategy formulation and execution are affected. The invention discloses a corn full life cycle monitoring method based on air-ground coordination, which comprises the following steps: Collecting multispectral images of corn fields from a top view angle through an unmanned aerial vehicle, detecting and outputting the inferred stalk diameter and the top plane coordinate of each corn plant; Constructing a first plant distribution matrix containing inferred stalk diameters and top plane coordinates of all plants based on data measured by the unmanned aerial vehicle; constructing a second plant distribution matrix containing measured stalk diameters and base plane coordinates of all plants based on data measured by the ground robot; Matching the first plant distribution matrix with the second plant distribution matrix according to the relative position relation between plants based on plane coordinates and deducing the numerical value corresponding relation between the diameter of the stalk and the diameter of the actually measured stalk; And according to the matching result, the top monitoring data collected by the unmanned aerial vehicle and the side monitoring data collected by the ground robot are related to the same corn plant. Preferably, the step of constructing a first plant distribution matrix comprising inferred stalk diameters and top plane coordinates of all plants based on data measured by an unmanned aerial vehicle, and constructing a second plant distribution matrix comprising measured stalk diameters and base plane coordinates of all plants based on data measured by a ground robot, specifically comprises: for unmanned aerial vehicle data, calculating Euclidean distances between each corn plant and coordinates of all other plants by taking top plane coordinates of each corn plant as nodes, and finding out K nearest neighbor plants to form a local topological subgraph taking the plant as a center; for ground robot data, executing the same operation as unmanned plane data, and constructing a local topological subgraph with the same K value for each plant based on the base plane coordinates; In each partial topological sub-graph, the diameter of a central plant, the distance vectors of the central plant and K neighbors and the diameter of the K neighbors are combined into a characteristic vector, and the characteristic vector sets of all plants form a first plant distribution matrix and a second plant distribution matrix. Preferably, the step of matching the first plant distribution matrix with the second plant distribution matrix according to the relative positional relationship between plants based on plane coordinates and deducing the numerical correspondence between the diameter of the stalk and the diameter of the actual measured stalk specifically includes: Calculating the comprehensive similarity between any two partial topological sub-graph feature vectors in the first plant distribution matrix and the second plant distribution matrix, wherein the comprehensive similarity is the weighted sum of the central plant