A land data identification method and system based on IoT drone

Abstract

The present invention relates to the technical field of land data identification and processing, and discloses a land data identification method and system based on an Internet of Things UAV, the method comprising the following steps: collecting multi-source land data of agricultural land through an Internet of Things UAV and a sensor and preprocessing; dividing the agricultural land into land blocks, and extracting the irrigation characteristic index of each land block based on the preprocessed multi-source land data; calculating the irrigation factor of each land block based on the irrigation characteristic index; collecting historical data, analyzing and predicting meteorological factors; and formulating an agricultural land irrigation plan based on the results of the analysis and prediction of the irrigation factor and meteorological factors. The present invention improves the efficiency and accuracy of agricultural land land data identification, makes the formulation of irrigation plans more scientific and reasonable, helps promote the development of precision agriculture and smart agriculture, and improves the efficiency and sustainability of agricultural production.

Description

Technical Field

[0001] The present invention relates to the technical field of land data identification and processing, and in particular to a land data identification method and system based on an Internet of Things unmanned aerial vehicle. Background Art

[0002] In recent years, the demand for land data identification for agricultural land has been growing worldwide. Especially in the context of the development of precision agriculture and smart agriculture, the accurate acquisition and analysis of land data has become particularly important. Traditional land data identification methods mainly rely on manual surveys, ground measurements, and satellite remote sensing. These methods have problems such as high cost, long cycle, and slow data update, and it is difficult to meet the needs of modern agriculture for real-time, dynamic, and high-precision identification of land data. The emergence of IoT drones provides a new solution for land data identification for agricultural land. Drones are equipped with various sensors, such as high-resolution cameras, infrared sensors, and lidars, which can quickly acquire a large amount of land data and transmit it to the data processing center in real time for analysis and processing through IoT technology. This method not only improves the efficiency of data acquisition, but also reduces costs and enables real-time data updates.

[0003] Although IoT drones have significant advantages in identifying land data for agricultural land, existing technologies still have some shortcomings. First, due to the complexity and diversity of the agricultural land environment, it is often difficult for a single sensor to obtain land data comprehensively and accurately. Therefore, the fusion and preprocessing of multi-source land data has become an urgent problem to be solved. Secondly, in the process of land data identification, how to effectively extract and utilize the irrigation needs of the land is also a challenge. The existing technology lacks a comprehensive evaluation method that comprehensively considers multiple indicators, which may lead to the formulation of irrigation plans that are not scientific and reasonable enough and cannot meet the needs of actual agricultural production. In addition, the impact of meteorological factors on land data cannot be ignored. Changes in meteorological factors such as temperature and precipitation will directly affect the evaporation of water in the land and the growth of crops, thereby affecting the irrigation demand. However, existing technologies often ignore the analysis and prediction of meteorological factors, resulting in poor accuracy in land data identification and a lack of foresight and pertinence in the formulation of irrigation plans.

[0004] For example, the patent application with publication number CN107292484A discloses a method and system for identifying urban redevelopment land based on urban space big data, which includes: determining a preselected urban area that needs to be identified for urban redevelopment land, gridding the preselected urban area, obtaining multiple land units, and generating a redevelopment land unit set; selecting the candidate redevelopment land unit set to obtain feasible redevelopment land units, and generating a feasible redevelopment land unit set; establishing a multi-objective urban redevelopment land urgency assessment model; and assessing the feasible redevelopment land units in the feasible redevelopment land unit set according to the multi-objective urban redevelopment land urgency assessment model to form an urban redevelopment land unit. The invention can provide objective identification of land at a microscopic scale.

[0005] For example, the patent application with publication number CN116384759A discloses a method and system for identifying the rationality of gridded land use in urban areas, including preprocessing high-resolution land use remote sensing monitoring data to obtain preprocessed urban area land use data; for the preprocessed urban area land use data, using Shannon information entropy to determine the optimal grid segmentation scale to obtain the optimal scale gridded land use data; for the optimal scale gridded land use data, quantitatively estimating the carbon emissions based on urban area land use, the landscape pattern index of the urban area grid and the grid land use conflict comprehensive index, using the objectively weighted entropy weight method to extract the land use coercion index, and identify the rationality of each gridded land use in the urban area. The invention proposes and constructs an automated identification scheme for the rationality of gridded land use in urban areas, which can identify the conflict space between land development and ecological protection.

[0006] The above patents all have the problems raised by this background technology: a single sensor is often difficult to obtain land data comprehensively and accurately, and the problem of integrating multi-source land data has not been solved.

[0007] The information disclosed in this background technology section is only intended to enhance the understanding of the overall background of the invention and should not be regarded as an acknowledgement or any form of suggestion that the information constitutes the prior art already known to ordinary technicians in this field. Summary of the invention

[0008] The technical problem to be solved by the present invention is to overcome the defects of the prior art and provide a land data identification method and system based on the Internet of Things UAV, so as to improve the efficiency and accuracy of land data identification for agricultural land, make the formulation of irrigation plans more scientific, reasonable and forward-looking, promote the development of precision agriculture and smart agriculture, and improve the efficiency and sustainability of agricultural production.

[0009] In order to solve the above technical problems, the present invention provides the following technical solutions: On the one hand, the present invention provides a land data identification method based on an Internet of Things drone, comprising the following steps: S1: Collect and pre-process multi-source land data of agricultural land through IoT drones and sensors; S2: Divide the agricultural land into land blocks and extract the irrigation characteristic indicators of each land block based on the preprocessed multi-source land data; S3: Calculating the irrigation factor of each land block based on the irrigation characteristic index; S4: Collect historical data, analyze and predict meteorological factors; S5: Formulate an irrigation plan for agricultural land based on the analysis and prediction results of the irrigation factors and meteorological factors.

[0010] As a preferred solution of the land data identification method based on the Internet of Things UAV of the present invention, wherein: the multi-source land data includes land images, land infrared radiation data, land elevation information, and soil moisture data; The preprocessing is specifically as follows: spatially aligning multi-source land data; Performing image correction and filtering noise reduction on the land image; Performing radiation calibration and atmospheric correction on the land infrared radiation data to obtain land surface temperature data; Filtering and denoising the land elevation information and establishing a digital elevation model; Outlier detection and replacement are performed on the soil moisture data.

[0011] As a preferred solution of the land data identification method based on the Internet of Things UAV of the present invention, the irrigation characteristic indicators include vegetation coverage, terrain slope, average surface temperature, and average soil moisture; wherein the vegetation coverage is extracted in the following manner: Performing image segmentation on the land image through a machine learning model to obtain vegetation areas and non-vegetation areas; Optimize the land image after image segmentation, including object aggregation and boundary smoothing; Cut and splice the optimized land images to obtain the land images of each land block; The number of pixels in the vegetation area of ​​the land image of each land block is counted and divided by the total number of pixels in the land image to obtain the vegetation coverage of each land block.

[0012] As a preferred solution of the land data identification method based on the Internet of Things UAV of the present invention, the terrain slope is extracted as follows: The digital elevation model is used to extract the elevation and horizontal coordinates of each pixel in the land image of any land block, and the slope of each pixel is calculated. The formula is as follows: ; Among them, s represents the slope of any pixel; Indicates the maximum elevation difference between the pixel and any adjacent pixel in its 8-neighborhood; express The horizontal coordinate difference between the two corresponding pixels, express The difference between the horizontal ordinates of the two corresponding pixels; Represents the arc of a semicircle; The slope of each pixel in the land image of any land block is counted and the average is calculated as the terrain slope of the land block.

[0013] As a preferred solution of the land data identification method based on the Internet of Things UAV of the present invention, the method of extracting the average surface temperature is as follows: The surface temperature data is converted into a surface temperature distribution map, and the surface temperature data of each pixel point in the land image of any land block is obtained by spatial interpolation and the average is calculated as the average surface temperature of the land block; The average soil moisture is extracted as follows: The soil moisture data is converted into a soil moisture distribution map, and the soil moisture data of each pixel in the land image of any land block is obtained by spatial interpolation and the average is calculated as the average soil moisture of the land block.

[0014] As a preferred solution of the land data identification method based on the Internet of Things UAV of the present invention, the calculation formula of the irrigation factor is as follows: ; Where w represents the irrigation factor of any land parcel; Indicates vegetation coverage; represents the terrain slope; t represents the normalized value of the average surface temperature; h represents the average soil moisture; Indicates the soil moisture reference value.

[0015] As a preferred solution of the land data identification method based on the Internet of Things UAV of the present invention, the historical data includes historical precipitation data, historical temperature data, historical irrigation data, and historical growth data; The meteorological factor analysis is specifically to train an irrigation amount reference model and calculate an irrigation amount reference value based on temperature and precipitation; the method is as follows: The historical data are organized into different historical data groups according to the planting cycle of the agricultural land, each historical data group corresponds to the historical data of the same planting cycle; a historical growth data threshold is set, and the historical data groups whose historical growth data are greater than the historical growth data threshold are screened to form a training set for the irrigation amount reference model; The irrigation quantity reference model is constructed, and the equation is as follows: ; Wherein, T represents the average temperature of a planting cycle; P represents the precipitation of a planting cycle; It represents the reference value of irrigation amount in the planting cycle with average temperature T and precipitation P; represents the irrigation efficiency coefficient; represents the precipitation efficiency coefficient; The formula for expressing the water requirement of crops during the planting period with an average temperature of T is as follows: ; Among them, k, a, b, c, and d are all model parameters; The model is trained by the training set, and the model parameters k, a, b, c, d and irrigation efficiency coefficient are determined by parameter fitting method. , precipitation efficiency coefficient The value of ; save the irrigation amount reference model that has completed training.

[0016] As a preferred solution of the land data identification method based on the Internet of Things UAV of the present invention, the meteorological factor prediction specifically includes precipitation prediction and temperature prediction; the method is as follows: Calculate the mean of the historical temperature data in each historical data group, and organize the mean of the historical temperature data of each historical data group into a time series in chronological order as the average temperature series; Extract the historical precipitation data of each historical data group and organize them into a time series in chronological order as the precipitation series; The temperature prediction model is trained based on the average temperature sequence, and the average temperature of the next planting cycle is predicted by the trained temperature prediction model, which is recorded as ; The temperature prediction model is any one of the SARIMA model, the long short-term memory network, and the gated recurrent unit; A precipitation prediction model is trained based on the precipitation sequence, and the precipitation of a future planting cycle is predicted by the trained precipitation prediction model, which is denoted as ; The precipitation prediction model is any one of the SARIMA model, the long short-term memory network, and the gated recurrent unit.

[0017] As a preferred solution of the land data identification method based on the Internet of Things UAV of the present invention, the agricultural land irrigation plan is to plan the irrigation amount of each land block; the method is as follows: The average temperature forecast for the next planting cycle Precipitation forecast for the next planting cycle Input the irrigation amount reference model to obtain the irrigation amount reference value for the next planting cycle, which is recorded as ; Calculate the reference value of irrigation volume for each land parcel, the formula is as follows: ; in, represents the reference value of irrigation volume for the jth land parcel, the value range of j is 1, 2, ..., n, where n is the total number of land parcels; represents the irrigation factor of the jth land parcel; represents the mean value of irrigation factor for all land parcels; Represents the maximum value of the irrigation factor among all land parcels.

[0018] In a second aspect, the present invention provides a land data identification system based on an Internet of Things UAV, including a data acquisition module, a preprocessing module, an irrigation feature module, a meteorological analysis module, a meteorological prediction module, and an irrigation plan module; wherein: The data acquisition module is used to collect multi-source land data of agricultural land; The preprocessing module is used to preprocess multi-source land data of agricultural land; The irrigation feature module is used to extract irrigation feature indicators from multi-source land data and calculate irrigation feature factors; The meteorological analysis module is used to collect historical data and train an irrigation quantity reference model; The weather forecast module is used to forecast precipitation and temperature; The irrigation plan module is used to formulate an irrigation plan for agricultural land.

[0019] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: Through the combination of IoT drones and sensor technology, this application can obtain multi-source land data of agricultural land in real time and efficiently. The flexibility of drones makes data acquisition no longer limited by ground conditions and time, greatly improving the efficiency and real-time nature of data acquisition. Through the fusion and preprocessing of multi-source land data, the physical properties and spatial information of the land can be more comprehensively reflected, making the data more accurate and reliable, providing a solid foundation for the formulation of irrigation plans.

[0020] This application comprehensively considers multiple irrigation characteristic indicators such as vegetation coverage, terrain slope, average surface temperature, average soil moisture, etc., making the extraction of irrigation characteristics more scientific and reasonable, and more accurately reflecting the actual irrigation needs of the land. Based on the analysis results of irrigation factors and meteorological factors, the formulation of irrigation plans is targeted, which can better meet the actual irrigation needs of the land and avoid unnecessary waste of resources. Reduce agricultural production costs, improve the utilization efficiency of water resources, and promote sustainable agricultural development. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them: Figure 1 A flow chart of the land data identification method based on the Internet of Things UAV provided by the present invention; Figure 2 A schematic diagram of the structure of the land data identification system based on the Internet of Things UAV provided by the present invention; Figure 3 This is a structural schematic diagram of the data acquisition module provided by the present invention. DETAILED DESCRIPTION

[0022] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. The embodiments of the present invention and the technical features in the embodiments may be combined with each other unless there is a conflict.

[0023] Example 1

[0024] This embodiment introduces a land data recognition method based on an IoT drone. Figure 1 , the method comprises the following steps: S1: Collect and pre-process multi-source land data of agricultural land through IoT drones and sensors; Multi-source land data include land images, land infrared radiation data, land elevation information, and soil moisture data; The preprocessing is as follows: spatially align the multi-source land data to facilitate the subsequent comprehensive analysis of the multi-source data; Performing image correction and filtering noise reduction on the land image to ensure that the pixel positions in the image are accurate, remove noise in the image, and improve image clarity; The infrared radiation data of the land is subjected to radiation calibration and atmospheric correction to obtain the surface temperature data; water vapor, aerosols, etc. in the atmosphere will affect the propagation of infrared radiation, and atmospheric correction can be performed to obtain the true surface temperature; Filtering and denoising the land elevation information and establishing a digital elevation model; Outlier detection and replacement are performed on the soil moisture data to ensure the accuracy and reliability of the data.

[0025] S2: Agricultural land is divided into land parcels, and irrigation characteristic indicators of each land parcel are extracted based on the preprocessed multi-source land data; Irrigation characteristic indicators include vegetation cover, terrain slope, average surface temperature, and average soil moisture; The vegetation coverage is extracted as follows: Perform image segmentation on the land image through a machine learning model, such as U-Net, FCN, etc., to obtain vegetation areas and non-vegetation areas; The land image after image segmentation is optimized, including object aggregation and boundary smoothing; the goal of object aggregation is to merge those areas that should belong to the same area but are mistakenly divided into multiple small blocks due to the limitations of the segmentation algorithm. Clustering algorithms such as DBSCAN can be used to aggregate objects and merge closely connected similar areas. Boundary smoothing aims to optimize the continuity and naturalness of the boundaries of vegetation areas in the segmentation results and reduce jagged or irregular edges. Morphological dilation and erosion operations are used for boundary smoothing. The dilation operation can help close small gaps, while the erosion operation can remove tiny protrusions. The combination of the two can achieve the effect of boundary smoothing.

[0026] Cut and splice the optimized land images to obtain the land images of each land block; The number of pixels in the vegetation area of ​​the land image of each land block is counted and divided by the total number of pixels in the land image to obtain the vegetation coverage of each land block.

[0027] The terrain slope is extracted as follows: The digital elevation model is used to extract the elevation and horizontal coordinates of each pixel in the land image of any land block, and the slope of each pixel is calculated. The formula is as follows: ; Among them, s represents the slope of any pixel; Indicates the maximum elevation difference between the pixel and any adjacent pixel in its 8-neighborhood; express The horizontal coordinate difference between the two corresponding pixels, express The difference between the horizontal ordinates of the two corresponding pixels; Represents the arc of a semicircle; The slope of each pixel in the land image of any land block is counted and the average is calculated as the terrain slope of the land block.

[0028] The average surface temperature is extracted as follows: The surface temperature data is converted into a surface temperature distribution map, and the surface temperature data of each pixel point in the land image of any land block is obtained by spatial interpolation and the average is calculated as the average surface temperature of the land block; The average soil moisture is extracted as follows: The soil moisture data is converted into a soil moisture distribution map, and the soil moisture data of each pixel in the land image of any land block is obtained by spatial interpolation and the average is calculated as the average soil moisture of the land block.

[0029] The surface temperature distribution map and the soil moisture distribution map represent the surface temperature distribution and soil moisture distribution of agricultural land respectively. Since the surface temperature data and soil moisture data are both discrete data points in space, they cannot cover the surface temperature and soil moisture corresponding to each pixel in the land image. Spatial interpolation, such as Kriging interpolation and spline interpolation, can be used to assign surface temperature and soil moisture to each pixel in the corresponding land image in the surface temperature distribution map and soil moisture distribution map.

[0030] S3: Calculate the irrigation factor of each land block based on the irrigation characteristic index; the formula is as follows: ; Where w represents the irrigation factor of any land parcel; Indicates vegetation coverage; represents the terrain slope; t represents the normalized value of the average surface temperature; h represents the average soil moisture; It represents the soil moisture reference value, which is set by technicians in this field according to actual needs.

[0031] Vegetation coverage can assess the growth density and health of crops. High vegetation coverage usually means that more water is needed to maintain crop growth. The slope of the terrain affects the distribution and loss rate of soil moisture. Slopes will accelerate water runoff, resulting in increased water demand. Infrared data can accurately measure the temperature of the surface. High temperature areas indicate that water evaporates quickly, the soil is prone to drying, and more irrigation is needed. Soil moisture directly affects irrigation demand, and the lower the moisture, the greater the irrigation demand.

[0032] S4: Collect historical data, analyze and predict meteorological factors; The historical data include historical precipitation data, historical temperature data, historical irrigation data, and historical growth data; the historical growth data are data reflecting the growth of crops planted on agricultural land, and may be any one of unit area yield, biological yield, economic yield, and harvest index; wherein, unit area yield is the average crop yield per unit area (such as hectares or mu), biological yield refers to the total amount of organic matter accumulated by crops through photosynthesis during their life cycle; economic yield refers to the part that can be directly used by humans, such as the weight of grain after removing straw; the harvest index is the ratio of the crop's economic yield to its biological yield, reflecting the efficiency of the crop in converting biomass into economic value.

[0033] The meteorological factor analysis is specifically to train an irrigation amount reference model and calculate an irrigation amount reference value based on temperature and precipitation; the method is as follows: The historical data are organized into different historical data groups according to the planting cycle of the agricultural land, each historical data group corresponds to the historical data of the same planting cycle; a historical growth data threshold is set, and the historical data groups whose historical growth data are greater than the historical growth data threshold are screened to form a training set for the irrigation amount reference model; The irrigation quantity reference model is constructed, and the equation is as follows: ; Wherein, T represents the average temperature of a planting cycle; P represents the precipitation of a planting cycle; It represents the reference value of irrigation amount in the planting cycle with average temperature T and precipitation P; It represents the irrigation efficiency coefficient, that is, the proportion of irrigation volume that can be absorbed and utilized by crops; It represents the precipitation efficiency coefficient, that is, the proportion of precipitation that can be absorbed and utilized by crops; precipitation and irrigation cannot be completely converted into water absorption available to crops. Considering factors such as soil absorption capacity, surface runoff and deep infiltration, an efficiency coefficient is set for precipitation and irrigation respectively, which represents the proportion of natural precipitation and irrigation water converted into usable water.

[0034] The formula for expressing the water requirement of crops during the planting period with an average temperature of T is as follows: ; Among them, k, a, b, c, and d are all model parameters; Crops have a theoretical water requirement during their planting cycle, and this water requirement is directly affected by the temperature. On the one hand, suitable temperature conditions can promote the growth of crops, thereby increasing the water requirement of crops. On the other hand, high temperature causes a large amount of water evaporation in the farmland, which will also increase the demand for water of crops. However, too high a temperature may be detrimental to the growth of crops, resulting in a decrease in the demand for water; therefore, the water requirement is positively correlated with the temperature within a certain range, but it is not a linear relationship. Among the above model parameters, k and a determine the maximum value and rising rate of the water requirement-temperature curve, b controls the steepness of the curve, c is the inflection point temperature, that is, the temperature point where the water requirement growth rate begins to decline, and d represents the basic water requirement, which is the minimum water requirement of crops even at the most unsuitable temperature.

[0035] The model is trained by the training set, and the model parameters k, a, b, c, d and irrigation efficiency coefficient are determined by parameter fitting method. , precipitation efficiency coefficient The value of ; save the irrigation amount reference model that has completed training.

[0036] The meteorological factor prediction specifically includes precipitation prediction and temperature prediction; the method is as follows: Calculate the mean of the historical temperature data in each historical data group, and organize the mean of the historical temperature data of each historical data group into a time series in chronological order as the average temperature series; Extract the historical precipitation data of each historical data group and organize them into a time series in chronological order as the precipitation series; The temperature prediction model is trained based on the average temperature sequence, and the average temperature of the next planting cycle is predicted by the trained temperature prediction model, which is recorded as ; The temperature prediction model is any one of the SARIMA model, the long short-term memory network, and the gated recurrent unit; A precipitation prediction model is trained based on the precipitation sequence, and the precipitation of a future planting cycle is predicted by the trained precipitation prediction model, which is denoted as ; The precipitation prediction model is any one of the SARIMA model, the long short-term memory network, and the gated recurrent unit.

[0037] S5: Formulate an irrigation plan for agricultural land based on the analysis and prediction results of the irrigation factors and meteorological factors.

[0038] The agricultural land irrigation plan is to plan the irrigation amount for each land block; the method is as follows: The average temperature forecast for the next planting cycle Precipitation forecast for the next planting cycle Input the irrigation amount reference model to obtain the irrigation amount reference value for the next planting cycle, which is recorded as ; Calculate the reference value of irrigation volume for each land parcel, the formula is as follows: ; in, represents the reference value of irrigation volume for the jth land parcel, the value range of j is 1, 2, ..., n, where n is the total number of land parcels; represents the irrigation factor of the jth land parcel; represents the mean value of irrigation factor for all land parcels; Represents the maximum value of the irrigation factor among all land parcels.

[0039] Example 2 This embodiment is the second embodiment of the present invention; it is based on the same inventive concept as embodiment 1, Figure 2 This embodiment introduces a land data identification system based on an IoT drone, including a data acquisition module, a preprocessing module, an irrigation feature module, a meteorological analysis module, a meteorological prediction module, and an irrigation plan module; wherein: The data acquisition module is used to collect multi-source land data of agricultural land; Figure 3 The module includes a drone unit and a humidity monitoring unit; the drone unit is equipped with a high-resolution camera, an infrared sensor, and a lidar, which can collect images, infrared data, and elevation information of agricultural land; the humidity detection unit is a humidity sensor distributed in the agricultural land, which can collect soil moisture information at different locations.

[0040] The preprocessing module is used to preprocess the multi-source land data of agricultural land, including filtering and denoising the image data and cleaning the numerical data; The irrigation feature module is used to extract irrigation feature indicators from multi-source land data and calculate irrigation feature factors; the module is equipped with tools and formulas such as machine learning models, image processing software, and irrigation factor calculation formulas, and can extract features related to irrigation needs from multi-source land data.

[0041] The meteorological analysis module is used to collect historical data and train an irrigation amount reference model; the historical data includes historical precipitation data, historical temperature data, historical irrigation amount data, and historical growth data; the irrigation amount reference model is trained based on historical data, and calculates the reference value of irrigation amount by considering temperature, precipitation, and crop growth; The meteorological forecast module is used to forecast precipitation and temperature; organize historical precipitation data and historical temperature data into time series, and use the time series forecast model to forecast precipitation and temperature respectively; The irrigation plan module is used to formulate an irrigation plan for agricultural land; the irrigation reference value under future meteorological factors is obtained through the results of precipitation and temperature prediction and the irrigation reference model, and the irrigation reference value is adjusted through the irrigation factor to determine the irrigation amount of agricultural land.

[0042] The specific functions of the above modules are implemented by referring to the relevant contents of the land data identification method based on the Internet of Things UAV described in Example 1, which will not be elaborated here.

[0043] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0044] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the purpose and scope of protection of the present invention, which are all within the protection of the present invention.

Claims

1. A land data identification method based on an Internet of Things drone, characterized by: The following steps are involved: S1: Collect and pre-process multi-source land data of agricultural land through IoT drones and sensors; S2: Divide the agricultural land into land blocks and extract the irrigation characteristic indicators of each land block based on the preprocessed multi-source land data; S3: Calculating the irrigation factor of each land block based on the irrigation characteristic index; S4: Collect historical data, analyze and predict meteorological factors; S5: Formulate an irrigation plan for agricultural land based on the analysis and prediction results of the irrigation factors and meteorological factors.

2. The land data identification method based on the Internet of Things UAV as claimed in claim 1 is characterized by: The multi-source land data includes land images, land infrared radiation data, land elevation information, and soil moisture data; The preprocessing is specifically as follows: spatially aligning multi-source land data; Performing image correction and filtering noise reduction on the land image; Performing radiation calibration and atmospheric correction on the land infrared radiation data to obtain land surface temperature data; Filtering and denoising the land elevation information and establishing a digital elevation model; Outlier detection and replacement are performed on the soil moisture data.

3. The land data identification method based on the Internet of Things UAV as claimed in claim 2 is characterized by: The irrigation characteristic indicators include vegetation coverage, terrain slope, average surface temperature, and average soil moisture; wherein the vegetation coverage is extracted as follows: Performing image segmentation on the land image through a machine learning model to obtain vegetation areas and non-vegetation areas; Optimize the land image after image segmentation, including object aggregation and boundary smoothing; Cut and splice the optimized land images to obtain the land images of each land block; The number of pixels in the vegetation area of ​​the land image of each land block is counted and divided by the total number of pixels in the land image to obtain the vegetation coverage of each land block.

4. The land data identification method based on the Internet of Things UAV as claimed in claim 3 is characterized by: The terrain slope is extracted as follows: The digital elevation model is used to extract the elevation and horizontal coordinates of each pixel in the land image of any land block, and the slope of each pixel is calculated. The formula is as follows: ； Among them, s represents the slope of any pixel; Indicates the maximum elevation difference between the pixel and any adjacent pixel in its 8-neighborhood; express The horizontal coordinate difference between the two corresponding pixels, express The difference between the horizontal ordinates of the two corresponding pixels; Represents the arc of a semicircle; The slope of each pixel in the land image of any land block is counted and the average is calculated as the terrain slope of the land block.

5. The land data identification method based on the Internet of Things UAV as claimed in claim 4 is characterized by: The average surface temperature is extracted as follows: The surface temperature data is converted into a surface temperature distribution map, and the surface temperature data of each pixel point in the land image of any land block is obtained by spatial interpolation and the average is calculated as the average surface temperature of the land block; The average soil moisture is extracted as follows: The soil moisture data is converted into a soil moisture distribution map, and the soil moisture data of each pixel in the land image of any land block is obtained by spatial interpolation and the average is calculated as the average soil moisture of the land block.

6. The land data identification method based on the Internet of Things UAV as claimed in claim 5 is characterized by: The calculation formula of the irrigation factor is as follows: ； Where w represents the irrigation factor of any land parcel; Indicates vegetation coverage; represents the terrain slope; t represents the normalized value of the average surface temperature; h represents the average soil moisture; Indicates the soil moisture reference value.

7. The land data identification method based on the Internet of Things UAV as claimed in claim 6 is characterized by: The historical data include historical precipitation data, historical temperature data, historical irrigation data, and historical growth data; The meteorological factor analysis is specifically to train an irrigation amount reference model and calculate an irrigation amount reference value based on temperature and precipitation; the method is as follows: The historical data are organized into different historical data groups according to the planting cycle of the agricultural land, each historical data group corresponds to the historical data of the same planting cycle; a historical growth data threshold is set, and the historical data groups whose historical growth data are greater than the historical growth data threshold are screened to form a training set for the irrigation amount reference model; The irrigation quantity reference model is constructed, and the equation is as follows: ； Wherein, T represents the average temperature of a planting cycle; P represents the precipitation of a planting cycle; It represents the reference value of irrigation amount in the planting cycle with average temperature T and precipitation P; represents the irrigation efficiency coefficient; represents the precipitation efficiency coefficient; The formula for expressing the water requirement of crops during the planting period with an average temperature of T is as follows: ； Among them, k, a, b, c, and d are all model parameters; The model is trained by the training set, and the model parameters k, a, b, c, d and irrigation efficiency coefficient are determined by parameter fitting method. , precipitation efficiency coefficient The value of ; save the irrigation amount reference model that has completed training.

8. The land data identification method based on the Internet of Things UAV as claimed in claim 7 is characterized by: The meteorological factor prediction specifically includes precipitation prediction and temperature prediction; the method is as follows: Calculate the mean of the historical temperature data in each historical data group, and organize the mean of the historical temperature data of each historical data group into a time series in chronological order as the average temperature series; Extract the historical precipitation data of each historical data group and organize them into a time series in chronological order as the precipitation series; The temperature prediction model is trained based on the average temperature sequence, and the average temperature of the next planting cycle is predicted by the trained temperature prediction model, which is recorded as ; The temperature prediction model is any one of the SARIMA model, the long short-term memory network, and the gated recurrent unit; A precipitation prediction model is trained based on the precipitation sequence, and the precipitation of a future planting cycle is predicted by the trained precipitation prediction model, which is denoted as ; The precipitation prediction model is any one of the SARIMA model, the long short-term memory network, and the gated recurrent unit.

9. The land data identification method based on the Internet of Things UAV as claimed in claim 8, characterized in that: The agricultural land irrigation plan is to plan the irrigation amount for each land block; the method is as follows: The average temperature forecast for the next planting cycle Precipitation forecast for the next planting cycle Input the irrigation amount reference model to obtain the irrigation amount reference value for the next planting cycle, which is recorded as ; Calculate the reference value of irrigation volume for each land parcel, the formula is as follows: ； in, represents the reference value of irrigation volume for the jth land parcel, the value range of j is 1, 2, ..., n, where n is the total number of land parcels; represents the irrigation factor of the jth land parcel; represents the mean value of irrigation factor for all land parcels; Represents the maximum value of the irrigation factor among all land parcels.

10. A land data identification system based on an Internet of Things UAV, which is used to implement a land data identification method based on an Internet of Things UAV according to any one of claims 1 to 9, characterized in that: It includes data acquisition module, preprocessing module, irrigation feature module, meteorological analysis module, meteorological forecast module and irrigation plan module; among which: The data acquisition module is used to collect multi-source land data of agricultural land; The preprocessing module is used to preprocess multi-source land data of agricultural land; The irrigation feature module is used to extract irrigation feature indicators from multi-source land data and calculate irrigation feature factors; The meteorological analysis module is used to collect historical data and train an irrigation quantity reference model; The weather forecast module is used to forecast precipitation and temperature; The irrigation plan module is used to formulate an irrigation plan for agricultural land.

Images