An Aerial Survey Method for UAV-based Agricultural Situation Monitoring Optimized by Spatial Sampling

Abstract

The present invention belongs to the technical field of agricultural situation monitoring, and particularly relates to an unmanned aerial vehicle (UAV) agricultural situation monitoring aerial survey method based on spatial sampling optimization, which includes constructing a digital farmland base model and dividing the smallest agronomic units with independent agronomic attributes; calculating the number of sampling points for each unit based on a sampling model, performing Monte Carlo random sampling and excluding abnormal points in combination with a spatial coordinate verification mechanism to achieve dynamic sampling planning; using a traveling salesman problem optimization algorithm to generate an intelligent flight path; implementing automatic control of the aerial survey task by calculating the flight altitude; dynamically adjusting the sampling density according to the crop growth heterogeneity coefficient to obtain the agricultural situation information of the smallest agronomic units, and integrating and generating a plot-level monitoring report. The UAV agricultural situation monitoring aerial survey method based on spatial sampling optimization according to the present invention can significantly improve the efficiency and accuracy of farmland monitoring and provide precise decision-making support for farmland management through the collaborative design of optimizing the flight path by a multi-objective ant colony algorithm and a dynamic hierarchical sampling mechanism.

Description

Technical Field

[0001] The present invention relates to the technical field of UAV aerial survey and agricultural informatization, and relates to a UAV agricultural situation monitoring aerial survey method based on spatial sampling optimization, specifically a UAV agricultural situation monitoring method based on spatial sampling optimization and intelligent path planning, which is applicable to farmland growth assessment, pest and disease monitoring, and precision agriculture management. Background Art

[0002] With the rapid development of UAV technology, its application in the field of agricultural situation monitoring has become increasingly widespread, bringing significant changes to modern agriculture. Due to its efficient, accurate, and real-time monitoring capabilities, UAVs have become an indispensable important tool in modern agricultural production and are widely used in tasks such as farmland pest and disease monitoring, crop growth monitoring, and soil moisture monitoring. By carrying a variety of remote sensing sensors, such as high-definition digital cameras, multispectral sensors, and thermal imaging sensors, UAVs can obtain multi-dimensional and high-precision farmland information, including the number of plants, vegetation coverage, leaf area index, biomass, plant height, nitrogen nutrition status, water status, etc. These parameters together characterize the growth situation of crops and are ultimately related to yield.

[0003] However, most of the existing UAV agricultural situation data collection and monitoring technologies rely on experimental methods in the scientific research field, and their maturity and applicability are not yet sufficient to meet the needs of large-scale commercial applications. Although the existing UAV agricultural situation monitoring technologies can plan flight routes according to the set flight altitude, flight track overlap rate, and flight speed, there are still many defects in this method in actual farmland management.

[0004] First: The existing technology requires the UAV to cover the entire farmland area. During the flight process, the aerial survey path is planned based on factors such as the aerial survey area, recognition accuracy, resolution of the airborne imaging equipment, flight altitude, and flight track overlap rate, and large-scale image data is obtained according to the set flight line spacing (D) and exposure spacing (B). After these image data are transmitted back to the ground station, complex operations such as image stitching, three-dimensional point cloud computing, and orthophoto generation are required to obtain the orthophoto of the monitored plot. This process places extremely high requirements on data transmission, computer storage space, and computing performance, resulting in poor timeliness of agricultural situation monitoring and high technical thresholds, making it difficult to meet the agricultural situation monitoring needs of actual agricultural production.

[0005] Second: When higher monitoring accuracy is required, the flight altitude needs to be adapted to factors such as the focal length of the airborne imaging equipment, the size of the image sensor, and the sensor resolution (pixel scale). However, the route planning of the existing technology is mainly aimed at the complete area of ​​​​the farmland. In order to ensure the success rate of image stitching, the heading overlap rate is generally required to be ≥65%, and the lateral overlap rate is generally required to be ≥35%. Under the premise of a lower flight altitude, the area of ​​​​the sensor monitoring area is greatly reduced, which will greatly increase the flight route and image acquisition density, resulting in excessive flight time and low flight efficiency. Therefore, there is a prominent contradiction between monitoring accuracy and monitoring efficiency in the existing methods. The growth of crops is often a dynamically changing process, and obtaining data by flying in a way that fully covers the entire farmland area not only wastes a lot of flight time, but also increases the complexity and time cost of later data processing. The route planning method of the existing technology is obviously unable to meet actual needs.

[0006] Third: There is a lack of monitoring methods for spatial management units with independent agronomic management characteristics. For specific minimum agricultural units, the determination of sampling points is often done manually. Whether the number of sampling points can reflect the actual spatial heterogeneity, whether there is redundancy in the sampling points, and the lack of efficient flight route planning and control methods for the sampling point set in a single operation limit the feasibility of high-efficiency drone monitoring of each minimum agricultural unit in the field based on statistical concepts.

[0007] For field production, especially in farmland management decisions, the overall growth information of farmland does not require too high spatial resolution, but more attention is paid to the accuracy and timeliness of monitoring. The existing agricultural data collection methods based on drones have a contradiction between monitoring accuracy and monitoring efficiency, and cannot simultaneously meet the requirements of high accuracy and high timeliness of agricultural data collection. Therefore, how to improve the accuracy and timeliness of agricultural data collection and monitoring in agricultural production is a technical problem that technicians in this field need to solve urgently. Summary of the invention

[0008] In view of the above technical problems, the present invention provides a method for agricultural condition monitoring and aerial surveying by unmanned aerial vehicles (UAVs) based on spatial sampling optimization, which achieves the technical effect of reducing the number of image returns and computer processing pressure, and improving the accuracy and timeliness of agricultural condition data collection.

[0009] In view of this, the present invention provides a method for agricultural condition monitoring and aerial surveying by unmanned aerial vehicle based on spatial sampling optimization, comprising the following steps:

[0010] A method for agricultural condition monitoring and aerial survey using a drone based on spatial sampling optimization includes the following steps:

[0011] S1: Pre-build the GIS model of the monitoring plots, and draw the plot boundaries and the topological structure of the minimum agronomic unit based on the geographic coordinate information;

[0012] S2: Establish a sampling model according to the precision index of agricultural situation monitoring. Based on the minimum agronomic unit area parameter and the crop growth heterogeneity index, calculate the optimal number of sampling points for each minimum agronomic unit.

[0013] S3: Execute the Monte Carlo random sampling algorithm to generate a set of monitoring sampling points that meet the preset quantity and conform to the spatial distribution constraint, and exclude abnormal sampling points that exceed the unit boundary or have overlapping areas.

[0014] S4: Apply the traveling salesman problem optimization algorithm to generate a global route plan. With the minimum number of flight turns, the shortest total flight distance, and the consistent imaging azimuth angle as the multi-objective optimization constraint conditions, intelligently generate the UAV flight route.

[0015] S5: Implement the automatic control of the aerial survey task, perform fixed-point image acquisition at the preset flight altitude based on the RTK positioning technology, and construct a spatio-temporal associated metadata indexing system.

[0016] S6: Deploy a deep learning image analysis model, analyze the crop agricultural situation parameters through feature extraction technology, obtain the agricultural situation information of the minimum agronomic unit by sampling statistics method, and integrate and generate a plot-level monitoring report.

[0017] Further, in step S1, a centimeter-level precision digital farmland base model is constructed through the spatial registration of satellite remote sensing images and UAV aerial survey orthoimages.

[0018] Further, the minimum agronomic unit is a spatial governance unit with independent agronomic management characteristics.

[0019] Further, the division method of the minimum agronomic unit includes at least one of the irrigation system topological structure and the agricultural machinery operation area.

[0020] Further, in step S2, the sampling model of the sampling points satisfies , where n is the number of sampling points, Z is the confidence coefficient, p is the estimated parameter variance, e is the allowable error range.

[0021] Further, establish a spatial coordinate verification mechanism. In step S3, execute the Monte Carlo random sampling and combine the spatial coordinate verification mechanism to exclude abnormal points.

[0022] Further, in step S4, adopt an improved ant colony optimization algorithm, set the pheromone evaporation coefficient ρ = 0.1, the heuristic factor α = 1, the expectation factor β = 5, construct a multi-objective optimization function, and solve the multi-objective traveling salesman problem. Among them, the multi-objective optimization function is:

[0023]

[0024] is the weight coefficient, is the node spacing, is the number of turns, is the preset optimal acquisition azimuth angle, which is parallel to the sowing row azimuth angle.

[0025] Further, in step S4, the improved ant colony algorithm introduces a quantization pheromone decay mechanism. When the path pheromone concentration exceeds the threshold it is dynamically reset according to , where λ is the decay rate and t is the number of iterations.

[0026] Further, in step S5, the method for determining the preset flight altitude includes:

[0027] Based on the spatial resolution requirement of the target agronomic parameters, the flight altitude is calculated through the formula where:

[0028] H is the UAV flight altitude, with the unit of meter;

[0029] GSD is the ground sampling distance, with the unit of centimeter / pixel, and is set to GSD ≤2D, where D is the minimum recognizable size of the target crop feature;

[0030] f is the camera focal length, with the unit of millimeter;

[0031] p x is the camera pixel size, with the unit of micron;

[0032] The flight altitude control error does not exceed ±0.5%H.

[0033] Further, the principle of the sampling and statistical method in step S6 is:

[0034] Perform spatial stratified sampling on each minimum agronomic unit, and dynamically adjust the sampling density according to the crop growth heterogeneity coefficient δ = σ / μ , σ is the standard deviation of historical data, μ is the mean value:

[0035] When δ ≥0.3, adopt a high-density sampling mode, and the number of sampling points n ≥5;

[0036] When 0.1 ≤ δWhen <0.3, a medium-density sampling mode is adopted, 3 ≤ n <5;

[0037] When δ <0.1, a low-density sampling mode is adopted, n = 1;

[0038] The sampling process verifies the representativeness of the samples through spatial autocorrelation analysis to ensure that the Moran index I ∈[-0.2, 0.2].

[0039] Compared with the prior art, the method for unmanned aerial vehicle (UAV) agricultural situation monitoring and aerial survey based on spatial sampling optimization described in the present invention has the following advantages:

[0040] (1) For the method for UAV agricultural situation monitoring and aerial survey based on spatial sampling optimization described in this application, by using a sampling strategy, the farmland is divided into multiple minimum agronomic units. Sampling points are set according to requirements within each operation unit, and flight routes are planned based on these sampling points, reducing redundant data collection, significantly improving flight efficiency, and thus enabling the UAV to fly along the optimal path, reducing unnecessary flights and samplings, optimizing the data collection process, and ensuring that accurate and representative farmland data can still be obtained while improving the monitoring efficiency.

[0041] (2) For the method for UAV agricultural situation monitoring and aerial survey based on spatial sampling optimization described in this application, through the collaborative design of optimizing the flight path with the multi-objective ant colony algorithm and the dynamic hierarchical sampling mechanism, the contradiction between improving the monitoring accuracy and the monitoring efficiency of the UAV is solved, and it is applicable to the rapid and accurate monitoring of agricultural situations in field crop scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 is a schematic diagram of the minimum agronomic unit and UAV flight route planning in the embodiments of the present invention;

[0043] The markings in the figure are represented as:

[0044] Farmland area 1, shelter forest 2, road 3, minimum agronomic unit 4, sampling point 5, flight route 6. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0045] Next, the technical solutions in the embodiments of the present application will be clearly described with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art belong to the scope of protection of the present application.

[0046] In the description of the present application, it should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments of the present application. For the convenience of description, the sizes of the various parts shown in the drawings are not drawn in actual proportional relationship. Technologies, methods, and devices known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, such technologies, methods, and devices should be regarded as part of the authorization specification. In all the examples shown and discussed here, any specific value should be construed as merely exemplary and not as a limitation. Therefore, other examples of the exemplary embodiments may have different values. It should be noted that like reference numerals and letters in the following drawings represent like items, and thus, once an item is defined in one drawing, it does not need to be further discussed in subsequent drawings.

[0047] It should be noted that the terms "first", "second", etc. in the description and claims of the present application are used to distinguish similar objects and are not used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second", etc. are generally of the same category and do not limit the number of objects. For example, the first object can be one or more. In addition, "and / or" in the description and claims means at least one of the connected objects, and the character " / " generally indicates an "or" relationship between the associated objects before and after.

[0048] It should be noted that in the description of the present application, the orientation or positional relationships indicated by the orientation terms such as "front, back, up, down, left, right", "horizontal, vertical, perpendicular, horizontal", and "top, bottom" are generally based on the orientation or positional relationships shown in the drawings. These terms are only for the convenience of describing the present application and simplifying the description. Without contrary description, these orientation terms do not indicate and imply that the device or element referred to must have a specific orientation or be constructed and operated in a specific orientation, and thus should not be construed as limiting the protection scope of the present application; the orientation terms "inside, outside" refer to the inside and outside relative to the contour of each component itself.

[0049] It should be noted that in this application, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article or device including such element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of this application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in the reverse order according to the functions involved. For example, the described methods may be performed in an order different from that described, and various steps may be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0050] The embodiments of this application have been described above in conjunction with the accompanying drawings. Without conflict, the embodiments and features in the embodiments of this application can be combined with each other. This application is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of this application, those of ordinary skill in the art can also make many forms without departing from the purpose of this application and the scope protected by the claims, and all belong to the protection scope of this application.

[0051] Existing UAV agricultural situation monitoring technologies usually need to cover the entire farmland area, obtain a large amount of data through frame-by-frame images, and perform complex stitching and 3D point cloud construction to generate orthophotos. This process places extremely high requirements on computer storage and processing performance, and the stitching and data processing times are long, resulting in poor monitoring timeliness. In addition, the flight path planning of existing technologies is usually based on the complete area of the farmland, with too long flight times and low efficiency, making it difficult to meet the actual farmland management needs. Especially in large-field production, high spatial resolution is not required for monitoring, and more attention is paid to accuracy and timeliness. According to the heterogeneity of the growth of large-field crops, establishing a statistical sampling model can achieve the mean result of the monitoring area based on limited sampling points.

[0052] This application discloses a UAV agricultural situation monitoring aerial survey method based on spatial sampling optimization, including:

[0053] S1: Pre-construct a geographic information system model of the monitored plot, draw the plot boundary and the topological structure of the minimum agronomic unit based on the geographic coordinate information, and establish a spatial planning benchmark;

[0054] S2: Establish a sampling model according to the precision index of agricultural situation monitoring. Based on the area parameter of the minimum agronomic unit and the crop growth heterogeneity index, calculate the optimal number of sampling points for each minimum agronomic unit;

[0055] S3: Execute the Monte Carlo random sampling algorithm to generate a set of monitoring sampling points that meet the preset quantity and conform to the spatial distribution constraints, and establish a spatial coordinate verification mechanism to exclude abnormal sampling points that exceed the unit boundary or have overlapping areas;

[0056] S4: Apply the Traveling Salesman Problem (TSP) optimization algorithm to generate a global route plan, with the minimum number of flight turns, the shortest total flight distance, and the consistent image acquisition azimuth angle as the multi-objective optimization constraints;

[0057] S5: Implement automatic control of aerial survey tasks, perform fixed-point image acquisition at a preset flight altitude based on RTK positioning technology, and construct a spatio-temporal associated metadata indexing system;

[0058] S6: Deploy a deep learning image analysis model, analyze crop agricultural situation parameters through feature extraction technology, obtain agricultural situation information of the minimum agronomic unit by sampling statistics method, and integrate and generate a plot-level monitoring report.

[0059] In farmland management, it is not necessary to conduct comprehensive coverage sampling of the entire farmland. Traditional methods often result in low monitoring efficiency and long processing time due to the huge amount of data. The unmanned aerial vehicle (UAV) agricultural situation monitoring aerial survey method based on spatial sampling optimization disclosed in this application includes the coordinated actions of steps such as geospatial modeling, dynamic sampling planning, intelligent route generation, precise flight altitude control, and adaptive sampling execution. It accurately constructs the boundary of the farmland area 1 and the topology of the minimum agronomic unit 4 through a geographic information system model, providing a benchmark for subsequent spatial planning. According to the precision requirements of agricultural situation monitoring, combined with the area of the minimum agronomic unit and crop growth heterogeneity, scientifically calculate the optimal number of sampling points 5 for each minimum agronomic unit to ensure that the sampling is both representative and not excessive. Then, use the Monte Carlo random sampling algorithm to intelligently generate a set of sampling points that meet the spatial distribution constraints, and eliminate abnormal points through the spatial coordinate verification mechanism to ensure the accuracy of sampling. Subsequently, apply the Traveling Salesman Problem optimization algorithm to generate an efficient global flight route 6 plan with the goals of reducing the number of flight turns, shortening the total flight distance, and keeping the image acquisition azimuth angle consistent. During the aerial survey of the UAV according to the planned flight route 6, precise fixed-point image acquisition is realized based on RTK positioning technology, and a spatio-temporal associated metadata indexing system is established for subsequent data processing. Finally, through the deep learning image analysis model, feature extraction and agricultural situation parameter analysis are performed on the collected image data, and the agricultural situation information of the minimum agronomic unit is obtained by sampling statistics method, and finally a plot-level monitoring report is integrated and generated to achieve precise monitoring of the agricultural situation of the farmland.

[0060] The method for unmanned aerial vehicle (UAV) agricultural situation monitoring aerial survey based on spatial sampling optimization described in this application reduces the monitoring object to the smallest agronomic unit, samples based on the smallest agronomic unit and conducts optimized flight path planning. The optimized flight path planning can accurately control the selection of sampling points, ensure the collection of representative data for each smallest agronomic unit, avoid the generation of excessive useless data, thereby reducing the pressure of data transmission back and subsequent processing, significantly improving the flight efficiency and timeliness of data collection, greatly enhancing the efficiency and accuracy of farmland monitoring, providing precise decision-making support for farmland management, greatly improving the efficiency of agricultural situation monitoring, and meeting the requirements of modern agriculture for rapid and accurate information.

[0061] As a preferred example of this application, in step S1, when drawing the boundary of the farmland area, for the spatial registration of satellite remote sensing images and UAV aerial survey orthophotos, manually draw polygons or use geographic information system (GIS) software to draw the boundary, and construct a digital farmland base model with centimeter-level accuracy. In the example of this application, farmland area 1 generally refers to a whole piece of farmland, usually with a shelter forest 2 or a road 3 as a separation around it. Because the area of a piece of farmland is relatively large, agronomic measures such as irrigation, fertilization, and pesticide spraying cannot be carried out simultaneously, so the smallest agronomic units 4 within the farmland are generated. When determining the smallest agronomic unit 4, first, the border of the farmland area 1 needs to be drawn. A satellite map with geographical coordinates or an orthophoto map synthesized by a UAV with geographical coordinate information can be used. When drawing the farmland boundary, it can be done manually on the remote control or with the help of GIS software such as ArcGIS, and use the shp file to accurately draw the polygon boundary. By using a satellite map or an orthophoto map to draw the farmland boundary in this application, not only the accuracy of the drawing is improved, but also geographical coordinate information can be obtained quickly, providing a reliable basis for subsequent flight planning and data collection. With the support of professional software such as ArcGIS, the boundary of the smallest agronomic unit can be accurately drawn and numbered, making the management of the entire farmland area more scientific and efficient; at the same time, the flexibility of manual drawing or using GIS tools also greatly improves the convenience and adaptability of the operation, enabling the farmland management in different regions to be flexibly adjusted according to the actual situation.

[0062] As a preferred example of this application, the smallest agronomic unit is a spatial governance unit with independent agronomic management characteristics, such as an area where operations such as irrigation, fertilization, and pesticide spraying can be carried out separately. By dividing the farmland into multiple such smallest agronomic units, refined management and monitoring can be carried out independently for each operation unit. For example, the growth status of crops in different operation units may vary, so targeted agronomic measures can be carried out separately in each operation unit, such as adjusting the irrigation volume or fertilization amount according to the needs of the crops. This division method not only considers the actual use situation of the farmland but also provides higher flexibility in farmland management.

[0063] After dividing the farmland into the smallest agronomic units, each operation unit can carry out agronomic measures independently, making farmland management more refined and scientific. Each operation unit can independently collect data and conduct agricultural situation monitoring, avoiding data redundancy caused by large-scale monitoring of the entire farmland area in the traditional method. This not only ensures the accuracy and timeliness of the monitoring data, but also reduces unnecessary data processing work and improves the speed and accuracy of data analysis.

[0064] As a specific example of the present application, according to the irrigation system topology and agricultural machinery operation zoning, the monitored plot is divided into a set of the smallest agronomic units with independent agronomic attributes to form a GIS topology structure. In the example of the present application, the smallest agronomic unit can be divided according to the energy efficiency of agronomic operations. For example, there are multiple wells in the area where the farmland is located, and the smallest agronomic unit is defined according to the maximum operation area when each well works simultaneously, or the smallest agronomic unit is defined according to the operation capacity of the fertilization machinery at each preset time (such as 2h, 4h, 8h, 16h, 24h, etc.). By dividing the farmland into irrigation, fertilization, pesticide spraying or machinery operation zones, each zone can operate and manage independently according to the different needs of the crops, greatly improving the utilization efficiency of resources and the pertinence of operations. For example, in the irrigation zone, the drone can monitor the soil humidity according to the actual water demand of each area, while in the fertilization management zone, the fertilization amount and fertilization method can be adjusted according to the nutrient status of the soil. This refined division of operation units not only improves the efficiency of farmland management, but also reduces the waste of agronomic measures. After the drone conducts precise monitoring according to these zones, it can provide more accurate agricultural situation data, helping the farmland manager to understand the crop growth status of each operation unit in real time and adjust agronomic measures in a timely manner. In the example of the present application, the division of the smallest agronomic unit under different agronomic measures can be the same or different. There are differences in the division methods of the smallest agronomic unit according to the provincial and municipal regions where the farmland is located and the types of farmland crops. Generally, starting from the convenience of farmland management decision-making, a farmland is divided into many small plots with approximately the same size but different shapes, which is called the smallest agronomic unit.

[0065] Through the above precise zoning method, the drone can conduct targeted monitoring within each operation unit, collect the agricultural situation data of each zone, ensure that the agronomic measures in each area are reasonably implemented and timely feedback is obtained, help the farmland manager to understand the crop growth status of each operation unit in real time and adjust agronomic measures in a timely manner. This not only improves the accuracy and timeliness of farmland monitoring, but also ensures the intelligence and high efficiency of farmland management.

[0066] As a preferred example of the present application, in step S2, the number of sampling points within the minimum agronomic unit is dynamically determined according to the area of the minimum agronomic unit and the uniformity of crop growth. In the example of the present application, after determining the boundaries of the farmland area 1 and the minimum agronomic unit 4, the divided minimum agronomic units 4 are numbered, and a unique number is assigned to each minimum agronomic unit according to its position, as well as the area of the minimum agronomic unit and the uniformity of crop growth, to dynamically determine the number of sampling points to be collected within each unit. In this process, the selection of sampling points follows the principle of uniform distribution, and the sampling points can be randomly generated. The sampling point data not close to the boundary of the minimum agronomic unit is selected, and the specific geographical coordinates of each sampling point are determined through precise calculation to ensure that the sampling points can comprehensively and evenly reflect the agricultural conditions within the operation unit.

[0067] By dynamically determining the number of sampling points within the minimum agronomic unit, the present application not only avoids the monitoring blind spots that may be caused by traditional fixed sampling points but also improves the utilization efficiency of monitoring resources. The uniform distribution of sampling points ensures that the monitoring data can comprehensively reflect the crop growth situation within the operation unit, improving the accuracy and reliability of the monitoring results. At the same time, in the monitoring route planning strategy of the present invention, the amount of monitoring data does not increase geometrically with the improvement of monitoring accuracy. The number of sampling points only depends on the spatial heterogeneity of the field, which can fundamentally solve the contradiction that the monitoring accuracy and efficiency of unmanned aerial vehicles cannot be improved synergistically, give full play to the innate advantages of the high spatio-temporal resolution of unmanned aerial vehicles, and accelerate the application of unmanned aerial vehicle monitoring.

[0068] As a preferred example of the present application meeting, in step S2, the sampling model of the sampling points satisfies , where n is the number of sampling points, Z is the confidence coefficient, p is the estimated parameter variance, e is the allowable error range; in step S3, Monte Carlo random sampling is performed and combined with a spatial coordinate verification mechanism to exclude abnormal points.

[0069] By combining the setting of the sampling model formula with Monte Carlo simulation technology, the present application realizes dynamic sampling planning, which can optimize the sampling distribution of farmland operation units within a large range. Using Monte Carlo simulation to perform multiple iterative calculations on the randomness in the sampling process and statistically analyze the error distribution, the coverage and representativeness of sampling points at different spatial positions are maximally guaranteed. At the same time, in the sampling stage, it cooperates with the minimum control strategy of spatial blocks to exclude or re-locate areas with outliers or excessive heterogeneity, thereby ensuring the accuracy and stability of data while taking into account sampling efficiency, providing a reliable data basis for subsequent farmland information analysis and decision-making. This setting reduces unnecessary data collection and analysis workload under the premise of ensuring monitoring accuracy, effectively reducing the overall time and cost investment.

[0070] In the example of the present application, in the schematic diagram, 3 sample points are collected for each minimum agronomic unit. The numbers can be simply 1, 2, 3, or 1-1, 1-2, 1-3 for multiple farmlands. According to the preset precision requirements of agricultural condition monitoring and following the principle of uniform sampling, the positions of these 3 sampling points are automatically generated within the minimum agronomic unit.

[0071] By setting multiple sampling points within the minimum agronomic unit in the present application, the data of multiple sampling points can be mutually verified, reducing the errors and biases that may be brought by a single sampling point. The multiple sampling points are arranged following the principle of uniform sampling, and the function of automatically generating the sampling point positions improves the monitoring efficiency and reduces the time and energy costs of manually arranging sampling points.

[0072] As a preferred example of the present application, in step S4, according to the sampling point positions of all the minimum agronomic units in the farmland, the flight route of the unmanned aerial vehicle (UAV) is planned. Among the monitoring sample points, according to the optimal flight efficiency control strategy, such as strategies with fewer turns, the shortest distance, and consistent sampling azimuth angles, etc., the system can automatically generate an efficient and practical flight route. Different from the traditional route planning method, this solution does not require setting a route overlap rate, but only sets the flight altitude according to the precision requirements of the agricultural condition indicators.

[0073] As a preferred example of the present application, in step S4, the flight route of the UAV adopts an improved ant colony optimization algorithm, with the pheromone evaporation coefficient ρ = 0.1, the heuristic factor α = 1, and the expectation factor β = 5, to solve the multi-objective traveling salesman problem (TSP), and it is intelligently generated with the shortest total flight distance, the fewest number of turns, and consistent image azimuth angles as the optimization constraints. In the example of the present application, the improved ant colony optimization algorithm is adopted, with the pheromone evaporation coefficient ρ = 0.1, the heuristic factor α = 1, and the expectation factor β = 5, to construct a multi-objective optimization function:

[0074]

[0075] where is the weight coefficient, is the node spacing, is the number of turns, is the preset optimal acquisition azimuth angle, which should be parallel to the sowing row azimuth angle;

[0076] By simulating the process of multiple "ant individuals" searching for paths in space through the ant colony algorithm, the optimization of the path is guided based on the set heuristic factor and expected variance. After each ant completes a search, the distribution of pheromones is adjusted according to its path weight and quality, gradually aggregating the optimized flight choices; finally, the algorithm solves the task according to the set quality constraints, realizes the optimal flight path planning, and the system can automatically calculate and intelligently generate the UAV flight path most suitable for farmland monitoring, which maximally improves the flight efficiency and reduces the time and energy consumption.

[0077] The flight path generated by this improved ant colony optimization algorithm in this application can not only minimize the number of sharp turns during flight while maintaining the shortest total flight range, but also ensure a relatively consistent azimuth angle when collecting images, thus shortening the flight range, reducing energy consumption, while reducing the mechanical wear and operation difficulty of the UAV body, and effectively improving the comparability of image data and the efficiency of subsequent stitching processing, realizing the ability to quickly and accurately complete data collection even under large-scale farmland or complex terrain conditions, improving the UAV farmland patrol efficiency and the quality of agricultural situation monitoring, and providing source data with higher accuracy and consistency for subsequent image analysis and agronomic decision-making.

[0078] For the UAV agricultural situation monitoring aerial survey method based on spatial sampling optimization described in this application, in this way, the flight path of the UAV between monitoring sample points is more reasonable and efficient, reducing unnecessary turns and flight distances, thus saving flight time and energy consumption, and while ensuring the monitoring accuracy, completing the monitoring tasks of all sampling points with the optimal flight path, thereby improving the monitoring efficiency and accuracy.

[0079] As a preferred example of this application, the improved ant colony algorithm introduces a quantization pheromone decay mechanism. When the pheromone concentration of the path exceeds the threshold it is reset dynamically according to ( λ is the decay rate, t is the number of iterations). Through this quantization of pheromone decay, the algorithm can find a better solution in fewer iterations, improving the optimization efficiency.

[0080] As a preferred example of this application, in step S5, the unmanned aerial vehicle (UAV) automatically flies according to the set sampling route, acquires point-by-point images of the sampling points, and transmits the image data back. The transmitted image data is sorted and organized according to the smallest agronomic unit where it is located. In a specific example of this application, the UAV is equipped with a high-resolution camera. During flight, when it reaches above each preset sampling point, the camera is automatically triggered to take pictures to acquire the image data of that sampling point. After the shooting is completed, the UAV uses the built-in wireless communication module to transmit the image data to the ground control station in real time. After receiving the image data, the ground control station automatically identifies and determines the smallest agronomic unit to which the image data belongs according to the location information (such as GPS coordinates) contained in the image data. Then, the ground control station classifies and stores the image data according to the smallest agronomic unit, creating corresponding folders or database tables to ensure that the image data of each smallest agronomic unit can be filed in an orderly and clear manner. In this way, users can conveniently search for and view the corresponding image data according to the smallest agronomic unit, providing strong support for subsequent farmland agronomic situation analysis.

[0081] As a preferred example of this application, the method for determining the preset flight altitude in step S5 includes:

[0082] Based on the spatial resolution requirement of the target agronomic parameters, through the formula calculate the flight altitude, where:

[0083] H is the UAV flight altitude (unit: meter);

[0084] GSD is the ground sampling distance (unit: centimeter / pixel), which is set to GSD ≤2D (D is the minimum recognition size of the target crop feature);

[0085] f is the camera focal length (unit: millimeter);

[0086] p x is the camera pixel size (unit: micron);

[0087] The flight altitude control error does not exceed ±0.5%H, and it matches the band penetration characteristics of the spectral sensor, meeting the signal-to-noise ratio requirements of exponential inversion.

[0088] In this application, by using the above height calculation formula, first determine the required spatial resolution requirement according to the crop growth characteristics or monitoring target, and then divide the product of GSD and f by the camera pixel size p xTheoretically, the most suitable flight altitude H is obtained and matched with factors such as camera focal length, lens angle of view, and sensor sensitivity, so that the acquired images have stable optical imaging quality while ensuring resolution, reducing the waste of time and energy caused by excessive or insufficient flight on the premise of ensuring high-resolution acquisition, and effectively avoiding the problems of uneven image clarity and data deviation caused by randomly setting the flight altitude in traditional methods. By introducing an allowable error of ±0.5%H and combining it with ground spectral calibration, environmental interference factors can be quickly excluded during the monitoring process, improving the reliability and consistency of data. Especially when facing large areas of farmland or complex terrain areas, this solution can dynamically adjust the flight altitude according to different crop feature recognition requirements to ensure that key agricultural information can be obtained at the best imaging ratio every time, not only reducing the difficulty of subsequent stitching or feature recognition, but also significantly reducing the burden of data analysis.

[0089] As a preferred example of this application, in step S6, the sampling statistical method adopts an adaptive sampling execution method, and dynamically adjusts the sampling density according to the crop growth heterogeneity coefficient δ = σ / μ ( δ ≥0.3, n ≥5, 0.1≤ δ <0.3, 3≤ n <5, δ <0.1 n =1), and verifies the spatial representativeness through the Moran index (∣ I ∣≤0.2).

[0090] The specific statistical sampling principle in step S6 is as follows:

[0091] Perform spatial stratified sampling on each minimum agronomic unit, and dynamically adjust the sampling density according to the crop growth heterogeneity coefficient δ = σ / μ ( σ is the standard deviation of historical data, μ is the mean value), satisfying:

[0092] When δ ≥0.3, adopt a high-density sampling mode, and the number of sampling points n ≥5;

[0093] When 0.1≤ δ <0.3, adopt a medium-density sampling mode, 3≤ n <5;

[0094] When δ <0.1, adopt a low-density sampling mode, n =1;

[0095] The sampling process verifies the representativeness of the samples through spatial autocorrelation analysis to ensure that the Moran's index I ∈[-0.2, 0.2], achieving the Pareto optimality of sampling error and flight energy consumption.

[0096] This application establishes a dynamic evaluation mechanism for the crop growth heterogeneity coefficient δ = σ / μ Based on the standard deviation σ and the mean μ in the historical farmland data to measure the degree of dispersion of crop growth δ = σ / μ The δ value is divided into three levels: high heterogeneity, medium heterogeneity, and low heterogeneity. Different levels correspond to different numbers of sampling point layout methods. By arranging 5 sampling points in the high heterogeneity area, 3 to 4 sampling points in the medium heterogeneity area, and 1 to 2 sampling points in the low heterogeneity area, the hierarchical monitoring and sampling of crop growth conditions are realized. To further improve the accuracy and representativeness of sampling, the Voronoi diagram can also be used for spatial optimization of point layout. By dividing the farmland into several polygon units, each sampling point can cover the crop area closest to it, so as to effectively capture the differential information of crop growth with fewer point layouts. This application determines the heterogeneity level according to the threshold after obtaining the standard deviation σ and the mean μ , and accurately determines the sampling point positions in combination with the Voronoi grid segmentation method. Finally, a sampling strategy that dynamically adapts to the crop growth distribution characteristics is formed. Through the real-time calculation and evaluation of the δ value, the number and positions of sampling points can be flexibly adjusted at different growth stages of crops, and overall, the accurate control of the farmland conditions and the reasonable allocation of resources are achieved, laying a refined foundation for subsequent farmland data collection and analysis.

[0097] As a preferred example of this application, in step S6, various algorithms are used to extract the agronomic information of the smallest agronomic unit from the image to provide information for farmland management decision-making. After summarizing the agronomic information of all the smallest agronomic units, it is the agronomic information of the entire farmland. In the example of this application, by selecting one or more image analysis algorithms suitable for farmland agronomic monitoring, such as support vector machine (SVM) and random forest in machine learning algorithms, or convolutional neural network (CNN) in deep learning algorithms, the features in the image are automatically learned and classified or regression predicted.

[0098] The method for unmanned aerial vehicle (UAV) - based agricultural situation monitoring and aerial survey optimized by spatial sampling described in this application is aimed at field production. UAVs are used to collect agricultural situation information data. The farmland is divided into multiple minimum agronomic units through a sampling strategy, and multiple sampling points are set within each operation unit. According to the sampling points in all operation units, the optimal flight route is planned. The UAV flies along this route to collect image data within each operation unit, and computer science is used to analyze the images to obtain the agricultural situation of the minimum agronomic units, ultimately providing decision - making support for farmland management. In this way, the traditional method of comprehensively covering the farmland area is changed to monitoring key points within the minimum agronomic units, greatly reducing the scope of farmland monitoring, reducing unnecessary data collection, significantly reducing the flight time and the amount of collected data, thus improving the flight efficiency, reducing the time consumed for image stitching, and enabling rapid and accurate diagnosis of the agricultural situation of the farmland after flight, which conforms to the basic mode of farmland management. The method for UAV - based agricultural situation monitoring and aerial survey optimized by spatial sampling described in this application, through the collaborative design of optimizing the flight path by the multi - objective ant colony algorithm and the dynamic stratified sampling mechanism, solves the contradiction between improving the monitoring accuracy and the monitoring efficiency of UAVs, is applicable to the rapid and accurate monitoring of agricultural situations in field crop scenarios, optimizes the data collection efficiency of UAVs, greatly improves the accuracy and timeliness of monitoring, and also significantly reduces costs, achieving the goal of high - precision and high - timeliness of UAV - based agricultural situation monitoring. Specific embodiments

[0100] This application discloses a method for UAV - based agricultural situation monitoring and aerial survey optimized by spatial sampling, including:

[0101] 1. Geospatial modeling and unit division

[0102] Based on the spatial registration of satellite remote - sensing images and UAV orthophotos, a digital farmland base model with centimeter - level accuracy is constructed;

[0103] According to the topological structure of the irrigation system and the operation areas of agricultural machinery, the monitored plot is divided into a set of minimum agronomic units with independent agronomic attributes, forming a GIS topological structure.

[0104] 2. Construction of an adaptive sampling model

[0105] Establish a dynamic evaluation mechanism of the crop growth heterogeneity coefficient δ = σ / μ and perform spatial stratified sampling according to the heterogeneity level;

[0106] High - heterogeneity area ( δ ≥0.3): Deploy ≥5 sampling points and use Latin hypercube sampling to ensure spatial balance;

[0107] Medium - heterogeneity area (0.1 ≤ δ(<0.3): Set up 3 - 4 sampling points and optimize the minimum distance using the Voronoi diagram;

[0108] Low heterogeneity area ( δ <0.1): Set up 1 representative sampling point;

[0109] Verify the sample space independence through the Moran index (∣ I ∣≤0.2), and achieve a precision control with a sampling error ≤5%.

[0110] 3. Intelligent flight path planning

[0111] Adopt an improved ant colony optimization algorithm, set the pheromone evaporation coefficient ρ =0.1, the heuristic factor α =1, the expected factor β =5, and construct a multi - objective optimization function:

[0112]

[0113] Introduce a quantization pheromone update strategy, and dynamically reset when the path pheromone concentration exceeds the threshold to avoid local optimality.

[0114] 4. Flight altitude - precision coupling control

[0115] Based on the precision requirements for inverting agricultural situation parameters, calculate the flight altitude through the optical diffraction model:

[0116]

[0117] Where GSD ≤2D (D is the target feature size), and the band adaptation error of the multispectral sensor ≤3nm;

[0118] Adaptive sampling execution: Dynamically adjust the sampling density according to the crop growth heterogeneity coefficient ( δ ≥0.3 n ≥5, 0.1≤ δ <0.3 n <5, δ <0.1 n =1), and verify the spatial representativeness through the Moran index (∣ I ∣≤0.2).

[0119] By adopting the UAV - based agricultural situation monitoring aerial survey method based on spatial sampling optimization described in the present invention, it has the following beneficial effects compared with the traditional method:

[0120] 1: Optimization of the number of image acquisitions;

[0121] Traditional methods using full-coverage imagery require the collection of all image data in the area to be surveyed (e.g., for 1000 mu of farmland, more than 2000 images need to be taken), resulting in a data redundancy rate as high as 70% - 80%. Under high-precision requirements (such as GSD ≤ 2 cm), the number of images and resolution are inversely proportional to the square, and the data volume increases exponentially (e.g., when the resolution is doubled, the number of images increases to 4 times).

[0122] However, the UAV-based agricultural situation monitoring aerial survey method based on spatial sampling optimization described in the present invention obtains key-point images only through directional collection at sampling points based on a dynamic sampling model, reducing the number of images by 60% - 80% (e.g., for a 1000 mu area, the number of images is reduced from 2000 to 400 - 800).

[0123] Specific example data: For a 200 mu monitoring area, the number of images is reduced from 1800 (full coverage) to 400 (sampling), and the data volume is reduced by 77.8%.

[0124] In addition, the UAV-based agricultural situation monitoring aerial survey method based on spatial sampling optimization described in the present invention also realizes the function of optimizing spatial distribution.

[0125] By using the Moran index (| I | ≤ 0.2), the spatial independence of sampling points is ensured to avoid repeated coverage. The Voronoi diagram restricts the distance between adjacent points to be ≥ 3 times the image coverage diameter (e.g., if the coverage diameter of a single image is 5 m, then the point spacing ≥ 15 m), further reducing redundant images.

[0126] The UAV-based agricultural situation monitoring aerial survey method based on spatial sampling optimization described in the present invention achieves the following effects in terms of the number of images obtained:

[0127] (1) Reduction in image storage cost: For a 1000 mu area, the data volume is reduced from 500 GB (full coverage) to less than 100 GB.

[0128] (2) Improvement in transmission efficiency: Under a 4G network, the data transmission time is reduced from 3 hours to 40 minutes.

[0129] 2: Compression of image analysis and processing time;

[0130] Traditional methods use image stitching and orthorectification. Stitching 1000 images takes 2 - 3 hours (still 30 - 60 minutes after GPU acceleration), and the accuracy is affected by light and terrain (error ≥ 5%). Analyzing the entire area based on orthoimages (such as calculating NDVI) requires processing hundreds of millions of pixels, with a single analysis taking 1 - 2 hours.

[0131] In the present invention, by adopting a new method of unmanned aerial vehicle (UAV) agricultural situation monitoring aerial survey based on spatial sampling optimization, only the images of sampling points are independently processed, eliminating the splicing and orthorectification links. The analysis time is reduced by more than 90% (in the embodiment, the analysis time for a 200-acre area is reduced from 2 hours to 10 minutes), achieving the purpose of direct analysis without splicing. Moreover, by using a trimmed ResNet-18 model (the number of parameters is compressed from 11M to 3M), the processing time for a single image is reduced from 0.5 s to 0.1 s. The parallel processing framework (such as CUDA acceleration) supports the simultaneous parsing of more than 100 images, and the total processing time for 400 images is ≤1 minute, achieving the purpose of accelerating lightweight models.

[0132] The comparison of the impact analysis efficiency between the traditional method and the new method is as follows:

[0133]

[0134] 3: Reduction of agricultural situation monitoring complexity

[0135] The sources of the complexity of agricultural situation monitoring in the traditional method include:

[0136] Complex data preprocessing: Problems such as image registration error, uneven illumination, and splicing gaps need to be solved;

[0137] High computational load for the entire region: A large amount of pixel data needs to be processed (for example, a 1000-acre orthoimage contains 10^9 pixels), and the algorithm complexity is O(N²);

[0138] Redundant decision-making information: A large amount of irrelevant details are included in the data of the entire region, and additional noise filtering is required.

[0139] In the present invention, by adopting a new method of unmanned aerial vehicle (UAV) agricultural situation monitoring aerial survey based on spatial sampling optimization, stratified sampling is used for dimensionality reduction: the monitoring problem is reduced from "pixel-level analysis of the entire region" to "statistical inference of limited sampling points", and the algorithm complexity is reduced from O(N²) to O(k) (k is the number of sampling points, k N).

[0140] Specific embodiment data: By using the new method of the present invention, the number of analyzed pixels in a 200-acre area is reduced from 200 million to 400 sampling points, and the amount of calculation is reduced by 99.998%. By dynamically focusing on the key area and based on the heterogeneity coefficient δ ( ) resources are dynamically allocated: when the heterogeneity coefficient is in the high heterogeneity area ( δ ≥0.3), dense sampling is performed to capture details; when the heterogeneity coefficient is in the low heterogeneity area ( δ <0.1), single-point sampling is performed to avoid excessive calculation.

[0141] The new method of the UAV-based agricultural situation monitoring aerial survey method proposed in the present invention by adopting spatial sampling optimization can directly output statistical results at the minimum agronomic unit level (such as mean ± variance), replace the traditional full-region heat map, reduce decision-making interference information, and make the decision-making logic clearer.

[0142] The complexity comparison between the traditional method and the new method is as follows:

[0143]

[0144] 4: Comprehensive advantages in high-precision and large-region scenarios

[0145] Under the conditions of large farms of more than 10,000 mu and the requirement of centimeter-level monitoring accuracy, the advantages of the new method of the UAV-based agricultural situation monitoring aerial survey method proposed in the present invention by adopting spatial sampling optimization are further amplified:

[0146] (1) Saving of hardware resources:

[0147] The UAV endurance requirement is reduced from 6 hours (full coverage) to 2 hours (sampling flight path);

[0148] The computing server configuration can be reduced from 32-core GPU to 8-core CPU to meet real-time analysis.

[0149] (2) Real-time monitoring ability:

[0150] It supports the completion of the "aerial survey - analysis - decision-making" closed loop within 3 hours (the traditional method takes 1 - 2 days).

[0151] (3) Enhanced scalability:

[0152] Through distributed sampling (block parallel aerial survey), it supports the monitoring of ultra-large-scale farmland of millions of mu, and the system throughput increases linearly. Combined with the UAV airport system, the monitoring area of a single airport system can be greatly increased, and the system efficiency can be improved.

[0153] Taking 5,000 mu of standardized farmland as an example, compare the costs of the traditional method and the new method of the present application for agricultural situation monitoring:

[0154]

[0155] The UAV-based agricultural situation monitoring aerial survey method proposed in the present application by adopting spatial sampling optimization reduces the number of image acquisitions through dynamic sampling, optimizes the intelligent flight path to reduce the processing time, and simplifies the monitoring complexity through hierarchical analysis, and realizes in scenarios with large monitoring areas and high-precision requirements:

[0156] (1) The amount of image data is reduced by 60% - 80%, and the storage and transmission costs are significantly reduced;

[0157] (2)The processing time efficiency is improved by 98%, supporting near-real-time agricultural situation decision-making;

[0158] (3)The system complexity is reduced by more than 90%, adapting to the deployment of edge computing devices;

[0159] (4)The comprehensive cost is reduced by 75%, having the value of large-scale commercial promotion.

[0160] The method for unmanned aerial vehicle (UAV) agricultural situation monitoring aerial survey based on spatial sampling optimization described in this application redefines the technical paradigm of UAV agricultural situation monitoring, shifting from "full-scale acquisition + post-processing" to "precision sampling + real-time analysis", providing an efficient and scalable technical foundation for smart agriculture.

Claims

1. A method for agricultural monitoring and aerial surveying by unmanned aerial vehicle based on spatial sampling optimization, characterized in that: The steps include: S1: Pre-build a geographic information system model for the monitored plots, and draw the plot boundaries and the topological structure of the minimum agronomic unit based on the geographic coordinate information, wherein the minimum agronomic unit is a spatial governance unit with independent agronomic management characteristics; S2: Establish a sampling model based on the agricultural monitoring accuracy index, calculate the optimal number of sampling points for each minimum agronomic unit based on the minimum agronomic unit area parameter and crop growth heterogeneity index, and the sampling model of the sampling points meets ,in n is the number of sampling points, Z is the confidence coefficient, p is the estimated parameter variance, e is the allowable error range; S3: Execute the Monte Carlo random sampling algorithm to generate a set of monitoring sampling points that meet the preset number and conform to the spatial distribution constraints, and exclude abnormal sampling points that exceed the unit boundary or have overlapping areas; S4: Apply the traveling salesman problem optimization algorithm to generate global route planning, with the minimum number of flight turns, the shortest total flight distance and the consistent image acquisition azimuth as multi-objective optimization constraints, and intelligently generate drone routes; S5: Implement automated control of aerial survey tasks, collect fixed-point images at preset altitudes based on RTK positioning technology, and build a metadata indexing system with temporal and spatial correlation; S6: Deploy a deep learning image analysis model to analyze crop parameters through feature extraction technology, where the sampling density is dynamically adjusted according to the crop growth heterogeneity coefficient δ=σ / μ. σ is the standard deviation of historical data, μ The δ value is divided into three levels: high heterogeneity, medium heterogeneity and low heterogeneity. Different levels correspond to different numbers of sampling point layouts. The sampling statistical method is used to obtain the agricultural information of the minimum agronomic unit, and the integrated monitoring report is generated at the plot level.

2. The method for agricultural condition monitoring and aerial surveying by unmanned aerial vehicle based on spatial sampling optimization according to claim 1 is characterized in that: In step S1, a digital farmland base model with centimeter-level accuracy is constructed through spatial registration of satellite remote sensing images and drone aerial survey orthophotos.

3. The method for agricultural condition monitoring and aerial surveying by unmanned aerial vehicle based on spatial sampling optimization according to claim 1 is characterized in that: The division method of the minimum agronomic unit includes at least one of an irrigation system topology structure and an agricultural machinery operation partition.

4. The method for agricultural condition monitoring and aerial surveying by unmanned aerial vehicle based on spatial sampling optimization according to claim 1 is characterized in that: A spatial coordinate verification mechanism is established. In step S3, Monte Carlo random sampling is performed and combined with the spatial coordinate verification mechanism to eliminate abnormal points.

5. The method for agricultural condition monitoring and aerial surveying by unmanned aerial vehicle based on spatial sampling optimization according to claim 1 is characterized in that: In step S4, an improved ant colony optimization algorithm is used, the pheromone volatility coefficient ρ=0.1, the heuristic factor α=1, the expectation factor β=5 are set, a multi-objective optimization function is constructed, and the multi-objective traveling salesman problem is solved. The multi-objective optimization function is: ； is the weight coefficient, is the node spacing, is the number of turns, To preset the optimal collection azimuth, keep it parallel to the sowing row azimuth.

6. The method for agricultural condition monitoring and aerial surveying by unmanned aerial vehicle based on spatial sampling optimization according to claim 5 is characterized in that: In step S4, the improved ant colony algorithm introduces a quantized pheromone attenuation mechanism. When the path pheromone concentration exceeds the threshold When Dynamic reset, λ is the recession rate, t is the number of iterations.

7. The method for agricultural condition monitoring and aerial surveying by unmanned aerial vehicle based on spatial sampling optimization according to claim 1 is characterized in that: In step S5, the method for determining the preset flight altitude includes: Based on the spatial resolution requirements of the target agricultural parameters, the formula Calculate the flight altitude, where: H is the flight altitude of the drone, in meters; GSD is the ground sampling distance, in centimeters / pixel, which is set according to the monitoring accuracy requirements. GSD ≤2 D , D The minimum recognition size of the target crop feature; f is the focal length of the camera, in millimeters; p x is the camera pixel size in microns; The altitude control error does not exceed ±0.5%H.

8. The method for agricultural condition monitoring and aerial surveying by unmanned aerial vehicle based on spatial sampling optimization according to claim 1 is characterized in that: The principle of the sampling statistical method in step S6 is: Spatial stratified sampling is performed for each minimum agronomic unit, and the heterogeneity coefficient of crop growth within the unit is used to determine the spatial distribution of the crop growth. δ = σ / μ , dynamically adjust the sampling density: when δ ≥0.3, high-density sampling mode is adopted, and the number of sampling points n ≥5; When 0.1≤ δ <0.3, medium density sampling mode is adopted, 3≤ n <5; when δ When <0.1, a low-density sampling mode is used. n =1; The sampling process verifies the representativeness of the sample through spatial autocorrelation analysis to ensure the Moran index I ∈[-0.2, 0.2].

Images