Land use data set processing method and system

Abstract

The present invention relates to the field of data mining technology, and in particular to a land use data set processing method and system. The method comprises the following steps: obtaining ground sensor data of a monitoring area and constructing coordinate system data of the monitoring area; fusing the ground sensor data of the monitoring area to obtain a sensor network of the monitoring area; obtaining remote sensing data of the monitoring area and performing calculations to obtain land vegetation coverage rate data of the monitoring area; performing calculations based on the sensor network of the monitoring area to obtain land vegetation coverage rate error data of the monitoring area; estimating impact factors based on the land vegetation coverage rate data of the monitoring area and the land vegetation coverage rate error data of the monitoring area to obtain resolution impact factors; performing regional land vegetation coverage rate correction on the land vegetation coverage rate data of the monitoring area to obtain standardized regional land vegetation coverage rate data, and performing analysis to obtain regional land vegetation coverage rate trend data. The present invention improves the processing accuracy of land use data.

Description

Technical Field

[0001] The present invention relates to the technical field of data mining, and in particular to a land use data set processing method and system. Background Art

[0002] Scientific management of land use is the key to promoting sustainable development. However, efficient processing, accurate analysis and decision support of land use data rely on the acquisition and integration of a large amount of geospatial data. Therefore, the processing method of land use data set is of great significance to improving the level of land management, optimizing resource allocation and formulating reasonable management strategies. The processing method of land use data set aims to support the formulation of relevant decisions by collecting, analyzing and processing information such as land use type, spatial distribution and utilization intensity. In the past few decades, with the development of remote sensing technology, geographic information system (GIS) and big data processing technology, the collection and processing of land use data sets have gradually become possible. Traditional methods rely on remote sensing images and ground surveys, and the frequency and timeliness of obtaining land use data are low. The update cycle of remote sensing images is often long, and it is difficult to provide real-time land use changes, which poses a great challenge to land management that requires rapid response. Traditional data processing methods often require a lot of manual intervention, and the processing process is cumbersome and inefficient. When faced with large-scale data sets, the computing power and processing speed of traditional methods are difficult to meet the needs. In addition, the complexity of data processing makes it difficult to integrate and compare data sets from different times and sources. Summary of the invention

[0003] Based on this, it is necessary for the present invention to provide a land use dataset processing method and system to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a land use dataset processing method includes the following steps:

[0005] Step S1: Acquire ground sensor data of the monitoring area, and construct a monitoring area coordinate system according to the monitoring area ground sensor data, thereby obtaining the monitoring area coordinate system data; perform sensor data fusion on the monitoring area ground sensor data based on the monitoring area coordinate system data, thereby obtaining a monitoring area sensor network;

[0006] Step S2: Acquire remote sensing data of the monitoring area, and divide the land in the monitoring area into grids based on the remote sensing data of the monitoring area, thereby obtaining land grid data of the monitoring area; calculate the land vegetation coverage rate of the monitoring area according to the land grid data of the monitoring area, thereby obtaining land vegetation coverage rate data of the monitoring area;

[0007] Step S3: collecting real-time sensing data of the monitoring area based on the sensing network of the monitoring area, thereby obtaining real-time sensing data of the monitoring area, and calculating vegetation coverage error based on the real-time sensing data of the monitoring area and the land vegetation coverage data of the monitoring area, thereby obtaining land vegetation coverage error data of the monitoring area;

[0008] Step S4: estimating the resolution impact factor according to the land vegetation coverage rate data of the monitoring area and the land vegetation coverage rate error data of the monitoring area, thereby obtaining the resolution impact factor;

[0009] Step S5: Perform regional land vegetation coverage correction on the land vegetation coverage data of the monitoring area according to the resolution influencing factor and the coordinate system data of the monitoring area, so as to obtain standardized regional land vegetation coverage data, and perform vegetation coverage trend analysis based on the standardized regional land vegetation coverage data, so as to obtain regional land vegetation coverage trend data.

[0010] The acquisition of ground sensor data in the monitoring area of ​​the present invention and the construction of the coordinate system provide a basis for the precise positioning and fusion of subsequent data, making the spatiotemporal information of the monitoring data more accurate, thereby laying a foundation for real-time monitoring of land use. The ground sensor data is fused into a monitoring area sensor network, which can provide comprehensive and high-frequency data acquisition capabilities, ensuring that land use information in different regions and at different times can be quickly updated. This sensor network not only optimizes the efficiency of data acquisition, but also improves the timeliness of data, provides more timely feedback for land managers, and helps them to quickly respond to land changes and management needs. Through the acquisition of remote sensing data and the division of land grids, it is helpful to divide the monitoring area space into regular units, which is convenient for independent analysis of different plots, thereby improving the accuracy and flexibility of data processing. By calculating the land vegetation coverage rate, the ecological and environmental changes of the land can be reflected, providing an important ecological reference basis for the sustainable use of land resources. Compared with traditional methods, remote sensing data can provide a wider range of monitoring capabilities, and has a strong timeliness, and can capture the dynamic changes of land use in a timely manner. The acquisition of real-time sensor data and the combination with land vegetation coverage rate data are helpful for more accurate error calculation. The generation of vegetation coverage error data makes the monitoring results more reliable and avoids the deviation caused by sensor errors or incomplete data collection. This provides higher quality input data for subsequent data correction and analysis, and further improves the credibility of the analysis results. The estimation of resolution impact factor can quantify the impact of spatial resolution on the results during data processing, which provides a more refined adjustment method for land use analysis. Through this factor, data with different resolutions can be optimized to reduce the errors that may be caused by too high or too low resolution, thereby improving the accuracy of the results. Combined with the correction of resolution impact factor, the land vegetation coverage data can be further optimized to have high reliability and consistency at different resolutions. The standardized land vegetation coverage data can be corrected and trend analyzed to help decision makers understand the changing trend of regional land more clearly. This trend analysis not only reflects the current land use status, but also reveals potential change trends, providing a scientific basis for future land management decisions. Through long-term data accumulation and trend prediction, managers can formulate more accurate land use strategies, optimize resource allocation, reduce ecological risks, and promote the sustainable development of land resources. Overall, the effective implementation of this series of steps can significantly improve the processing accuracy and timeliness of land use data, reduce the problems caused by delayed data processing, insufficient accuracy or excessive manual intervention in traditional methods, and greatly improve the scientificity and efficiency of land management decisions through automated data collection, fusion, correction and analysis processes.

[0011] Optionally, step S1 specifically includes:

[0012] Step S11: acquiring ground sensor data in the monitoring area, and extracting ground sensor spatial position features from the ground sensor data in the monitoring area, thereby obtaining ground sensor spatial position data;

[0013] Step S12: constructing a monitoring area coordinate system according to the spatial position data of the ground sensor, thereby obtaining monitoring area coordinate system data;

[0014] Step S13: selecting the monitoring area sensor resolution based on the monitoring area coordinate system data, thereby obtaining the monitoring area sensor resolution data;

[0015] Step S14: resampling the ground sensing data in the monitoring area based on the sensor resolution data in the monitoring area, thereby obtaining the sensing data in the monitoring area;

[0016] Step S15: performing sensor data fusion on the monitoring area sensor data according to the monitoring area coordinate system data, thereby obtaining a monitoring area sensor network.

[0017] The present invention obtains the ground sensor data of the monitoring area and extracts the spatial position feature, and can accurately identify and obtain the spatial distribution information of the sensor, which lays the foundation for the subsequent construction of the coordinate system. By extracting the spatial position of the ground sensor, key data is further provided for the construction of the monitoring area coordinate system, ensuring the data consistency and spatial accuracy of the subsequent operation. In addition, the construction of the monitoring area coordinate system can accurately unify the positions of each sensor in the monitoring area into a common coordinate framework, so that each sensor data can be compared and fused under the same reference system, thereby improving the consistency of data processing. Selecting an appropriate sensor resolution is helpful to carry out fine processing of the monitoring area according to specific monitoring needs. By reasonably selecting the resolution, the accuracy and processing efficiency of the data can be optimized, and the information loss or redundancy caused by too high or too low resolution can be avoided. After the sensor resolution is determined, resampling the sensor data can effectively adjust the granularity of the data, so that the data is more in line with the needs of practical applications at the required resolution, ensuring the quality and reliability of the sensor data. Through sensor data fusion, data from different sensors can be comprehensively processed to further improve the accuracy and comprehensiveness of the data. Data fusion can eliminate the possible errors of a single sensor, enhance the stability of the sensor network and the credibility of the data in the monitoring area, and make the entire sensor network more intelligent and collaborative. Overall, the whole process can provide more accurate and reliable sensor data for the monitoring area, improve the overall performance of the monitoring system, and provide strong data support for subsequent data analysis, decision support and other applications.

[0018] Optionally, step S13 is specifically:

[0019] Step S131: performing sensor spatial distribution statistics on the spatial position data of the ground sensors based on the monitoring area coordinate system data, thereby obtaining sensor dense area data and sensor sparse area data;

[0020] Step S132: extracting sensor resolution features from the ground sensor data in the monitoring area, thereby obtaining ground sensor resolution data;

[0021] Step S133: performing regional sensor data fluctuation statistics on the ground sensor data in the monitoring area, thereby obtaining data of a high sensor data fluctuation area and data of a low sensor data fluctuation area;

[0022] Step S134: performing a region intersection operation on the sensor dense area data and the low sensor data fluctuation area data, so as to obtain the dense sensor area data; performing a region intersection operation on the sensor sparse area data and the high sensor data fluctuation area data, so as to obtain the sparse sensor area data;

[0023] Step S135: performing regional sensor resolution selection on densely sensed area data according to the ground sensor resolution data, thereby obtaining sensor low-resolution area data; performing regional sensor resolution selection on sparsely sensed area data according to the ground sensor resolution data, thereby obtaining sensor high-resolution area data;

[0024] Step S136: performing regional spatial merging on the sensor low-resolution regional data and the sensor high-resolution regional data, thereby obtaining the monitoring area sensor resolution data.

[0025] The present invention can identify the sensor dense area and sparse area in the monitoring area by statistically analyzing the spatial position data of the ground sensor, and provide data support for the subsequent sensor distribution optimization. By extracting the resolution feature of the ground sensor, the resolution information of the sensor can be accurately obtained, which is helpful for the accurate evaluation of the sensor performance in different areas. Statistical analysis of the sensor data fluctuation in the monitoring area helps to identify the high fluctuation area and the low fluctuation area, thereby providing a basis for subsequent regional optimization and data analysis. Based on these data, the intersection operation can accurately determine the position of the dense sensing area and the sparse sensing area, and implement resolution optimization for these two types of areas respectively, avoid waste of resources, and ensure the accuracy and efficiency of data collection in the monitoring area. By merging the data of the low-resolution and high-resolution areas, the overall optimization of the data in the area can be achieved, and the monitoring capability and accuracy of the entire sensor network can be improved. The implementation of these steps not only helps to improve the sensor performance of the monitoring area, but also improves the data quality, thereby enhancing the application value and actual effectiveness of the system.

[0026] Optionally, step S2 specifically includes:

[0027] Step S21: acquiring remote sensing data of the monitoring area, and extracting remote sensing spectral features of the monitoring area from the remote sensing data of the monitoring area, thereby obtaining spectral data of the monitoring area;

[0028] Step S22: performing spectral band combination on the spectral data of the monitoring area, so as to obtain regional vegetation index data, regional water index data and regional building index data;

[0029] Step S23: dividing the spectral data of the monitoring area into vegetation area grids according to the regional vegetation index data, thereby obtaining vegetation area grid data; dividing the spectral data of the monitoring area into water area grids according to the regional water body index data, thereby obtaining water body area grid data; dividing the spectral data of the monitoring area into building area grids according to the regional building index data, thereby obtaining building area grid data;

[0030] Step S24: performing regional grid spatial integration on the vegetation area grid data, the water area grid data and the building area grid data, so as to obtain the monitoring area land grid data;

[0031] Step S25: Calculate the land vegetation coverage rate of the monitoring area according to the land grid data of the monitoring area, so as to obtain the land vegetation coverage rate data of the monitoring area.

[0032] The present invention can provide abundant raw data for subsequent analysis by acquiring remote sensing data of the monitoring area and extracting spectral features. These data reflect the spectral information of various surface materials in the region and provide a basis for accurate classification. By combining spectral bands, further extracting important index data such as regional vegetation, water bodies and buildings, different surface types can be effectively distinguished, and accurate basic data support can be provided for regional environment and resource management. By gridding the spectral data of different regions such as vegetation, water bodies and buildings, the spatial distribution of each type of land object can be accurately delineated, ensuring that the fine-grained spatial information of various types of land objects in the monitoring area can be extracted, which is very important for subsequent analysis and monitoring. This refined grid data provides a scientific basis for subsequent land coverage assessment, and through the spatial integration process, not only different types of regional data can be processed uniformly, but also the overall analysis framework of the monitoring area can be optimized, and the consistency and comparability of the data are enhanced. Based on these spatially integrated land grid data, the vegetation coverage of the monitoring area can be efficiently calculated, providing accurate data support for the changing trend of the regional ecological environment, the vegetation recovery situation and the sustainable assessment of land use, thereby providing a powerful decision-making basis for applications such as environmental monitoring, urban planning, resource management and natural disaster assessment.

[0033] Optionally, step S25 is specifically:

[0034] Step S251: extracting grid vegetation index features from the land grid data in the monitoring area, thereby obtaining land grid vegetation index data;

[0035] Step S252: grid classification of land grid vegetation index data according to a preset vegetation index threshold, thereby obtaining vegetation coverage grid data and non-vegetation coverage grid data;

[0036] Step S253: Calculate the vegetation coverage grid ratio based on the vegetation coverage grid data and the non-vegetation coverage grid data, so as to obtain the land vegetation coverage rate data of the monitoring area.

[0037] The feature extraction method based on vegetation index of the present invention can systematically reflect the distribution state of vegetation on the land surface, and then provide more reliable data support for vegetation change, environmental monitoring and the like. Through the classification processing of vegetation index threshold, the monitoring area can be further accurately divided into vegetation covered and non-vegetation covered areas. This classification not only simplifies the analysis process of vegetation data, but also helps to identify different types of land coverage, provides more detailed regional division information, and is convenient for subsequent land use change research, ecological environment protection and the like. This classification step can effectively eliminate data interference in non-vegetation areas and ensure the specificity and accuracy of data analysis. The calculation step of vegetation coverage grid ratio further quantifies the vegetation coverage and can accurately reflect the land vegetation coverage rate of the monitoring area. This calculation not only simplifies the data processing process, but also provides more intuitive and easy-to-understand indicators, so that the change trend of regional vegetation can be clearly displayed. Through the quantification of this ratio, a more scientific basis can be provided for land resource management, and the sustainable development of ecological environment improvement and land use can be promoted.

[0038] Optionally, step S3 specifically includes:

[0039] Step S31: collecting real-time sensor data of the monitoring area based on the monitoring area sensor network, thereby obtaining real-time sensor data of the monitoring area;

[0040] Step S32: dividing the real-time sensing data of the monitoring area into the monitoring area grid sensing data according to the land grid data of the monitoring area, thereby obtaining the monitoring area grid sensing data;

[0041] Step S33: performing grid light intensity feature extraction and grid soil moisture feature extraction on the grid sensing data of the monitoring area, thereby obtaining grid light intensity data and grid soil moisture data;

[0042] Step S34: estimating the grid land vegetation coverage rate according to the grid light intensity data and the grid soil moisture, thereby obtaining the estimated data of the land vegetation coverage rate in the monitoring area;

[0043] Step S35: Calculate the vegetation coverage error of the estimated data of land vegetation coverage in the monitoring area and the data of land vegetation coverage in the monitoring area, so as to obtain the error data of land vegetation coverage in the monitoring area.

[0044] The present invention can obtain accurate and real-time environmental monitoring data through real-time sensor data collection based on the monitoring area sensor network, and provide effective basic data for further analysis and decision-making. The data are gridded so that the data can be subdivided according to specific geographical areas, which is helpful to achieve more accurate spatial analysis and regional feature extraction. By extracting the characteristics of light intensity and soil moisture, the environmental changes in each grid area can be deeply understood. The light intensity data reflects the sunshine conditions in the area, while the soil moisture data is directly related to the moisture conditions of the soil. These data provide key indicators for the subsequent land vegetation coverage estimation. The estimation of land vegetation coverage helps to evaluate the regional ecological health status, soil conservation capacity and agricultural production conditions, thereby supporting decision-making in multiple fields such as environmental protection, land use planning and agricultural management. Based on the estimation results, by performing error calculation with the actual land vegetation coverage data, the accuracy of the model can be further optimized to provide more reliable monitoring results. This error analysis can not only improve the reliability of the data, but also provide a strong basis for improving the accuracy of subsequent monitoring work, and promote the continuous improvement and perfection of monitoring technology and methods.

[0045] Optionally, step S34 is specifically:

[0046] Step S341: Acquire plant growth characteristic data;

[0047] Step S342: performing grid-suitable plant identification on the grid light intensity data and the grid soil moisture according to the plant growth characteristic data, thereby obtaining grid-suitable plant characteristic data;

[0048] Step S343: extracting grid monitoring video from the grid sensing data of the monitoring area to obtain grid real-time monitoring video, and dividing the grid real-time monitoring video into video frames to obtain grid monitoring video frames;

[0049] Step S344: performing edge detection according to the grid monitoring video frame to obtain edge feature data of the grid monitoring video frame, and performing feature similarity calculation according to the grid suitable plant feature data and the grid monitoring video frame edge feature data to obtain grid plant feature similarity data;

[0050] Step S345: performing vegetation pixel recognition on the grid monitoring video frame according to the grid plant feature similarity data, thereby obtaining a grid vegetation pixel marked frame;

[0051] Step S346: Estimating the grid vegetation coverage pixel ratios of the grid vegetation pixel marking frames and the grid monitoring video frames, thereby obtaining land vegetation coverage estimation data of the monitoring area.

[0052] The present invention can provide an accurate basis for subsequent suitable plant identification by acquiring plant growth characteristic data, ensure that the recommended plant species in each grid match its growth environment, and optimize the ecological environment for plant growth. This not only helps to identify plant species suitable for planting in different regions, but also effectively avoids the problem of poor plant growth caused by unsuitable climate or soil. The comprehensive analysis of grid light intensity and soil moisture data enables the monitoring system to intelligently judge the environmental conditions of each grid area and recommend the best growth conditions for plants in each grid, which has important application value in the fields of precision agriculture, ecological restoration and land use. Through the real-time video data and image processing technology of the monitoring area, the plant growth status and soil moisture change trend can be obtained in real time, ensuring that the system responds quickly to environmental changes, and improving the real-time and dynamic adaptability of the monitoring system. Through edge detection and feature similarity calculation of the monitoring video frame, the system can efficiently extract the features of plants and non-plant areas in the video, providing a high-quality data basis for further vegetation pixel identification. This makes the identification of vegetation coverage areas more accurate, reduces human intervention and errors, and greatly improves the automation of vegetation monitoring. With the marking of vegetation pixels and the estimation of coverage pixel ratios, a more detailed analysis of land vegetation coverage can be carried out, which can not only reveal the vegetation coverage conditions in different regions, but also provide a scientific basis for land management, environmental protection and agricultural development planning.

[0053] Optionally, step S4 is specifically:

[0054] Step S41: performing error grid extraction on the error data of land vegetation coverage in the monitoring area, thereby obtaining vegetation coverage error grid data;

[0055] Step S42: extracting grid vegetation index error from vegetation coverage error grid data, thereby obtaining grid vegetation index error data;

[0056] Step S43: extracting resolution features of the real-time sensing data of the monitoring area and the remote sensing data of the monitoring area, respectively, so as to obtain the sensing resolution data of the monitoring area and the remote sensing resolution data of the monitoring area;

[0057] Step S44: performing error grid resolution difference calculation on the monitoring area sensing resolution data and the monitoring area remote sensing resolution data according to the vegetation coverage error grid data, thereby obtaining error grid resolution difference data;

[0058] Step S45: estimating the resolution impact factor based on the error grid resolution difference data and the grid vegetation index error data, thereby obtaining the resolution impact factor.

[0059] The present invention can more accurately analyze and estimate the impact of resolution on monitoring results by extracting and processing the land vegetation coverage error data, grid vegetation index error data, and sensor and remote sensing resolution data in the monitoring area, thereby improving the accuracy and reliability of monitoring. By extracting the error data of vegetation coverage through the error grid, the error distribution of monitoring data can be deeply analyzed and quantified, providing a basis for subsequent error correction and optimization. By extracting the grid vegetation index error data, the error source of vegetation coverage can be further analyzed, and the vegetation index error can be accurately quantified, enhancing the understanding of vegetation distribution and change trends. By performing resolution feature extraction on real-time sensor data and remote sensing data, the characteristics of data at different resolutions can be identified, thereby providing a theoretical basis for differentiated processing and analysis. The error grid resolution difference calculation helps to reveal the impact of errors on vegetation monitoring results under different resolution conditions, thereby providing support for accurate error correction. Finally, the resolution influencing factor estimated based on the error grid resolution difference data and the grid vegetation index error data can quantify the specific impact of resolution on monitoring accuracy, thereby providing a quantitative basis for the optimization of subsequent monitoring work, helping to select the optimal data fusion strategy between data sources with different resolutions, and ultimately improving the accuracy and application effect of remote sensing and sensor monitoring technology.

[0060] Optionally, step S5 specifically includes:

[0061] Step S51: performing geographic coordinate system spatial error correction on the land vegetation coverage rate data of the monitoring area according to the coordinate system data of the monitoring area, thereby obtaining spatially corrected land vegetation coverage rate data;

[0062] Step S52: standardizing the spatially corrected land vegetation coverage data according to the resolution influencing factor, thereby obtaining standardized regional land vegetation coverage data;

[0063] Step S53: performing time series analysis based on the standardized regional land vegetation coverage data, thereby obtaining regional land vegetation coverage time series data;

[0064] Step S54: Perform vegetation coverage trend analysis based on the regional land vegetation coverage time series data, so as to obtain regional land vegetation coverage trend data.

[0065] The present invention corrects the spatial error of the geographic coordinate system for the coordinate system data of the monitoring area, eliminates the error caused by the coordinate system conversion, makes the final land vegetation coverage data more spatially accurate, and provides reliable basic data for subsequent analysis. The land vegetation coverage data after spatial correction is standardized by vegetation coverage, taking into account the resolution influence of different regions or areas, thereby eliminating the deviation caused by the resolution difference, making the data of different regions comparable after standardization, and avoiding the analysis error that may be caused by inconsistent data quality. By performing time series analysis on the land vegetation coverage data of the standardized area, the trend of vegetation coverage changing over time is revealed, so that researchers can clearly identify the periodicity and long-term trend of land vegetation coverage change, and provide a scientific basis for decision-making in the fields of ecological environment change, agricultural production, etc. According to the vegetation coverage trend analysis, it helps researchers to further extract the key features of vegetation coverage change, obtain the long-term development trend and prediction of regional land vegetation coverage, and can provide theoretical support for relevant departments in ecological protection, land use planning, etc., and promote the realization of sustainable development goals. Overall, this method improves the spatial and temporal consistency of data through precise data correction, standardization and time series analysis, ensuring the accuracy and wide application of vegetation coverage analysis results.

[0066] Optionally, the present specification also provides a land use dataset processing system, which is used to execute the land use dataset processing method as described above, and the land use dataset processing system includes:

[0067] The sensor data fusion module is used to obtain the ground sensor data of the monitoring area, and construct the monitoring area coordinate system according to the ground sensor data of the monitoring area, so as to obtain the monitoring area coordinate system data; based on the monitoring area coordinate system data, the sensor data of the ground sensor data of the monitoring area is fused to obtain the monitoring area sensor network;

[0068] The vegetation coverage calculation module is used to obtain remote sensing data of the monitoring area, and divide the land in the monitoring area into grids based on the remote sensing data of the monitoring area, so as to obtain land grid data of the monitoring area; calculate the land vegetation coverage of the monitoring area according to the land grid data of the monitoring area, so as to obtain the land vegetation coverage data of the monitoring area;

[0069] The vegetation coverage error calculation module is used to collect real-time sensing data of the monitoring area based on the monitoring area sensor network, so as to obtain real-time sensing data of the monitoring area, and to calculate the vegetation coverage error based on the real-time sensing data of the monitoring area and the land vegetation coverage data of the monitoring area, so as to obtain the land vegetation coverage error data of the monitoring area;

[0070] A resolution impact factor estimation module is used to estimate the resolution impact factor based on the land vegetation coverage rate data of the monitoring area and the land vegetation coverage rate error data of the monitoring area, so as to obtain the resolution impact factor;

[0071] The regional land vegetation coverage correction module is used to correct the regional land vegetation coverage data of the monitoring area according to the resolution influencing factor and the monitoring area coordinate system data, so as to obtain standardized regional land vegetation coverage data, and perform vegetation coverage trend analysis based on the standardized regional land vegetation coverage data, so as to obtain regional land vegetation coverage trend data.

[0072] The land use data set processing system of the present invention can implement any land use data set processing method of the present invention, and is used to combine the operations between various modules and the medium of signal transmission to complete the land use data set processing method. The internal modules of the system cooperate with each other to improve the processing accuracy of land use data. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

[0074] Figure 1 A schematic diagram of the steps of the land use data set processing method of the present invention;

[0075] Figure 2 Detailed step flow diagram of step S1 in the present invention;

[0076] Figure 3 Detailed step flow diagram of step S2 in the present invention;

[0077] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0078] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0079] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0080] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0081] To achieve this, please refer to Figures 1 to 3 The present invention provides a land use data set processing method, the method comprising the following steps:

[0082] Step S1: Acquire ground sensor data of the monitoring area, and construct a monitoring area coordinate system according to the monitoring area ground sensor data, thereby obtaining the monitoring area coordinate system data; perform sensor data fusion on the monitoring area ground sensor data based on the monitoring area coordinate system data, thereby obtaining a monitoring area sensor network;

[0083] In this embodiment, real-time ground data in the monitoring area is obtained by installing ground sensors (such as meteorological sensors, soil moisture sensors, temperature sensors, etc.) in the monitoring area. Ground sensors can be deployed in different geographical locations, such as farmland, forests, cities, etc., and collect data. According to these ground sensor data, the location information of the sensor is accurately calibrated by GPS or other positioning systems (such as differential GPS, RTK, etc.), so as to establish a geographical coordinate system of the monitoring area. For example, if GPS data is used, each data point of the sensor will be accompanied by longitude, latitude and altitude information, so as to construct a coordinate system with high accuracy. Then, according to the ground sensor data and the coordinate system data of the monitoring area, the data collected by different sensors are fused through data fusion algorithms (such as Kalman filtering, data weighted averaging, etc.) to improve the accuracy and completeness of the data. The information after data fusion will form a comprehensive sensor network to provide data support for subsequent analysis and monitoring. For example, by fusing meteorological and soil data, accurate monitoring data on the relationship between soil moisture and meteorological conditions can be provided.

[0084] Step S2: Acquire remote sensing data of the monitoring area, and divide the land in the monitoring area into grids based on the remote sensing data of the monitoring area, thereby obtaining land grid data of the monitoring area; calculate the land vegetation coverage rate of the monitoring area according to the land grid data of the monitoring area, thereby obtaining land vegetation coverage rate data of the monitoring area;

[0085] In this embodiment, remote sensing technology is used to obtain large-scale image data of the monitoring area. These remote sensing data can be obtained through high-resolution sensors carried by satellites or drones, such as Landsat or Sentinel satellite images, or commercial remote sensing satellites such as WorldView, GeoIQ and other data. The acquired remote sensing images generally contain data in multiple bands such as red, green, blue (RGB), and near infrared. Use image processing software (such as ENVI or ArcGIS) to pre-process these remote sensing images, including radiation correction, geometric correction, etc., to ensure the accuracy of the data. Then, the land is gridded using remote sensing image data. The specific method is to divide the entire monitoring area into several uniform grid units, each of which can be 10km×10km or finer resolution, determined according to the requirements of monitoring accuracy. These grids will be used for subsequent vegetation coverage calculations. After the grid division is completed, the vegetation index (such as NDVI, NormalizedDifference Vegetation Index) in the remote sensing image is used to calculate the vegetation coverage in each grid unit. The NDVI value can be used to effectively identify the vegetation coverage of different areas, thereby obtaining the vegetation coverage data within each grid. For example, an area with an NDVI value close to 1 indicates dense vegetation, while an area close to -1 indicates no vegetation or bare soil.

[0086] Step S3: collecting real-time sensing data of the monitoring area based on the sensing network of the monitoring area, thereby obtaining real-time sensing data of the monitoring area, and calculating vegetation coverage error based on the real-time sensing data of the monitoring area and the land vegetation coverage data of the monitoring area, thereby obtaining land vegetation coverage error data of the monitoring area;

[0087] In this embodiment, a sensor network (such as deployed meteorological sensors, soil moisture sensors, etc.) in the monitoring area is used for real-time data collection. The sensor periodically collects environmental data and transmits the data to the data center. It is assumed that the sensors used include soil moisture sensors and temperature sensors, and the collection frequency is once an hour. Based on these real-time sensor data and the land vegetation coverage data obtained from the remote sensing image, the error of the vegetation coverage is calculated by a data fusion algorithm. The specific method is to compare and analyze the real-time collected ground sensor data (such as temperature and humidity) with the vegetation coverage in the remote sensing data, and calculate the error between the two. For example, regression analysis or difference analysis methods can be used to calculate the difference between the actual vegetation coverage and the remote sensing data results based on the meteorological and soil conditions, thereby obtaining error data. This error calculation will help identify the deviation between the remote sensing data and the actual ground conditions, and provide a basis for subsequent correction and optimization.

[0088] Step S4: estimating the resolution impact factor according to the land vegetation coverage rate data of the monitoring area and the land vegetation coverage rate error data of the monitoring area, thereby obtaining the resolution impact factor;

[0089] In this embodiment, the influence of the resolution of the remote sensing image on the calculation of the vegetation coverage is estimated based on the land vegetation coverage data and its error data of the monitored area. Specifically, the land vegetation coverage data of the monitored area may have certain errors due to the different resolutions of the remote sensing images. For example, high-resolution images (such as 1m resolution) can better identify the vegetation distribution in a smaller area, while low-resolution images (such as 30m resolution) may merge multiple areas of different vegetation types together, resulting in an increase in the calculation error of the vegetation coverage. By analyzing the vegetation coverage errors at different resolutions, a resolution influence factor can be estimated, which reflects the influence of the image resolution on the calculation accuracy of the vegetation coverage. Assume that through comparative analysis, it is found that the error of the vegetation coverage increases by 5% for every 10 meters reduction in resolution. The estimation of this influence factor is crucial for the subsequent correction process.

[0090] Step S5: Perform regional land vegetation coverage correction on the land vegetation coverage data of the monitoring area according to the resolution influencing factor and the coordinate system data of the monitoring area, so as to obtain standardized regional land vegetation coverage data, and perform vegetation coverage trend analysis based on the standardized regional land vegetation coverage data, so as to obtain regional land vegetation coverage trend data.

[0091] In this embodiment, according to the resolution influencing factor and the coordinate system data of the monitoring area, the land vegetation coverage data is first corrected. The correction method is to adjust the vegetation coverage data in the remote sensing image data according to the resolution influencing factor. For example, if the error calculation shows that the coverage error caused by the low-resolution image is 5%, the error is corrected by weighting to make the vegetation coverage data more accurate. Then, the vegetation coverage trend analysis is performed based on the corrected standardized regional land vegetation coverage data. Trend analysis can use time series analysis methods (such as seasonal decomposition, sliding average method, etc.) to identify the changing trend of vegetation coverage. For example, the fluctuation of vegetation coverage in a monitoring area throughout the year can be analyzed to identify which periods of time the vegetation coverage rate has dropped significantly (possibly due to drought, seasonal changes, etc.) and which periods of time the vegetation coverage rate has rebounded.

[0092] Optionally, step S1 specifically includes:

[0093] Step S11: acquiring ground sensor data in the monitoring area, and extracting ground sensor spatial position features from the ground sensor data in the monitoring area, thereby obtaining ground sensor spatial position data;

[0094] In this embodiment, the ground sensors in the monitoring area obtain sensor data through a wireless sensor network (such as Zigbee, LoRa, etc.). These sensors may include sensors such as temperature, humidity, air pressure, pollutant concentration, and noise. First, the location information of the sensor is obtained through a global positioning system (GPS) or a positioning system based on a ground reference point (such as differential GPS or inertial measurement unit IMU). The location data of the sensor may include its geographic coordinates (longitude, latitude) and coordinates relative to a certain reference point (such as x, y, z coordinates in the local coordinate system). During the data extraction process, the spatial distribution characteristics of the sensor are extracted through algorithms (such as K-means clustering or spatial interpolation) to identify the specific location and spatial distribution of the sensor in the monitoring area. For example, in a farmland monitoring system, the ground sensor determines its precise location in the farmland through a positioning system, and these location characteristics will be used for subsequent data analysis.

[0095] Step S12: constructing a monitoring area coordinate system according to the spatial position data of the ground sensor, thereby obtaining monitoring area coordinate system data;

[0096] In this embodiment, a unified monitoring area coordinate system is constructed based on the acquired ground sensor spatial position data. Assuming that the geographic spatial range of the monitoring area is large, other sensor data in the monitoring area can be converted into local coordinates in the coordinate system by selecting a fixed point in the area (such as the reference position of a sensor or a known geographic coordinate) as the origin of the coordinate system. In specific implementation, GPS coordinates or relative coordinates can be converted into the local coordinate system of the area by using a coordinate conversion algorithm (such as projection conversion, plane coordinate system conversion, etc.). For example, the monitoring area is a large agricultural field. A corner of a field is selected as the origin, and the data of all sensors in the area are uniformly calibrated using a plane coordinate system. In this way, the monitoring data can be integrated according to a fixed standard, laying the foundation for subsequent data processing.

[0097] Step S13: selecting the monitoring area sensor resolution based on the monitoring area coordinate system data, thereby obtaining the monitoring area sensor resolution data;

[0098] In this embodiment, based on the coordinate system data of the monitoring area, a suitable sensor resolution is selected to ensure the accuracy and availability of the monitoring data. The choice of resolution is usually affected by the spatial density of the sensor, the characteristics of the monitoring target and the requirements of the monitoring task. For example, if the monitoring area is a temperature distribution monitoring area of ​​a city, different sensor resolutions can be selected according to different areas of the city (such as the city center, suburbs) and the monitoring accuracy requirements. Specifically, sensors in the city center may require higher spatial resolutions to reflect temperature changes in smaller areas, while sensors with lower resolutions can be used in the suburbs. In addition, factors such as data transmission bandwidth and storage capacity need to be considered. The choice of resolution needs to balance the sensor density and data processing capabilities. Taking agricultural environmental monitoring as an example, if you want to monitor the spatial changes in soil moisture in farmland, you may choose a resolution of 10 meters × 10 meters to capture a wide range of humidity changes, while ensuring that the data collected by the sensor can be effectively transmitted and processed.

[0099] Step S14: resampling the ground sensing data in the monitoring area based on the sensor resolution data in the monitoring area, thereby obtaining the sensing data in the monitoring area;

[0100] In this embodiment, the ground sensor data is resampled based on the resolution data of the sensors in the monitoring area. The purpose of resampling is to scale or interpolate the data according to the predetermined resolution requirements to ensure the consistency of the data at different resolutions. For example, if the resolution of the sensor data is 5 meters × 5 meters, and the new resolution requirement is 10 meters × 10 meters, the original data can be downsampled using downsampling techniques (such as average, maximum or interpolation methods) to ensure that the resampled data is consistent with the monitoring target. In actual implementation, the original sensor data can be processed by interpolation algorithms (such as bilinear interpolation, cubic interpolation) or simplified data processing algorithms (such as mean filtering) to ensure the accuracy and effectiveness of data at different resolutions. For example, in an agricultural irrigation monitoring system, high-resolution soil moisture data can be converted into lower-resolution data as needed through a resampling algorithm to facilitate regional water resource management.

[0101] Step S15: performing sensor data fusion on the monitoring area sensor data according to the monitoring area coordinate system data, thereby obtaining a monitoring area sensor network.

[0102] In this embodiment, the measurement data of different sensors are fused according to the coordinate system data of the monitoring area to form a comprehensive monitoring network. The purpose of data fusion is to improve the monitoring accuracy and eliminate the errors and inconsistencies that may be caused by single sensor data. Commonly used data fusion methods include weighted averaging, Kalman filtering, Bayesian fusion, etc. For example, in urban environmental monitoring, multiple sensors simultaneously monitor air quality data at different locations. Through data fusion, the real-time data of each sensor can be corrected for consistency in space and time to form a unified, high-precision urban air quality monitoring network. When implemented, the weighted averaging method can be used to assign different weights according to the reliability and accuracy of each sensor, or the Kalman filter can be used to predict and fuse dynamically changing data to obtain smoother and more accurate monitoring results. For example, for a temperature monitoring system in a multi-story building, the temperature data of multiple sensors will be fused into an overall internal temperature distribution map of the building, thereby providing more accurate temperature information.

[0103] Optionally, step S13 is specifically:

[0104] Step S131: performing sensor spatial distribution statistics on the spatial position data of the ground sensors based on the monitoring area coordinate system data, thereby obtaining sensor dense area data and sensor sparse area data;

[0105] In this embodiment, spatial distribution statistics are performed in a known coordinate system (such as a UTM coordinate system) based on the spatial position data of all ground sensors in the monitoring area. For example, if the monitoring area is a ground environment of a city, the area can be divided into several small cells (such as 10 meters × 10 meters) by constructing a grid model, and the number of sensors in each cell is counted. Through these statistical information, the area can be divided into a sensor-intensive area (for example, the number of sensors in each cell exceeds a certain threshold, defined as a "dense area") and a sensor-sparse area (for example, the number of sensors in each cell is lower than a certain threshold, defined as a "sparse area"). Assume that a monitoring area of ​​a city is used, the total area of ​​the area is 100 square kilometers, and the number of sensors is 10,000. Through spatial statistical analysis, it is found that there are more sensors distributed in the central area of ​​the city (such as an average of 50 sensors in each 100m × 100m grid), while there are fewer sensors in the suburban area (an average of only 5 sensors in each grid). Based on this statistical result, dense areas and sparse areas can be identified.

[0106] Step S132: extracting sensor resolution features from the ground sensor data in the monitoring area, thereby obtaining ground sensor resolution data;

[0107] In this embodiment, the resolution characteristics of the ground sensor are extracted by analyzing the collected data. The resolution of a sensor generally refers to the minimum change or spatial resolution that it can perceive (for example, the minimum distance or time interval measured). Specifically, the resolution can be determined by comparing the data output of the sensor with a certain standard value. Assume that temperature sensors, humidity sensors, and air quality sensors are installed in the monitoring area, and the collected data of all sensors are stored in the form of time series. By comparing the fluctuation amplitude of each sensor data, its resolution can be calculated. For example, if the minimum change value of the temperature sensor is 0.1°C, then its resolution is 0.1°C. The resolution of the humidity sensor is 1%, and the resolution of the air quality sensor is the minimum detection value of PM2.5. In a specific implementation, it is assumed that the data collected by a temperature sensor does not change by more than 0.5°C within 1 minute, then the resolution of the sensor is 0.5°C. For sensors in the same area, a series of resolution data can be obtained after analysis, and these data will serve as the basis for subsequent analysis.

[0108] Step S133: performing regional sensor data fluctuation statistics on the ground sensor data in the monitoring area, thereby obtaining data of a high sensor data fluctuation area and data of a low sensor data fluctuation area;

[0109] In this embodiment, the collected data of the ground sensor in the monitoring area is subjected to fluctuation statistics, and the amplitude of the data change is calculated. For example, the degree of data fluctuation can be measured by calculating the standard deviation or variance of the sensor's time series data. Areas with large fluctuations can be defined as high fluctuation areas, and areas with small fluctuations can be defined as low fluctuation areas. Assuming that in a certain area, the standard deviation value of the temperature data collected by multiple temperature sensors is large (such as the standard deviation exceeds 2°C), the area is determined to be a high sensor data fluctuation area. On the contrary, if the humidity sensor data in a certain area changes little (such as the standard deviation is less than 0.5%), the area is determined to be a low fluctuation area. Data analysis tools (such as the pandas library in Python) can be used to calculate the volatility of time series data and divide the area by a certain threshold. For example, assuming that the analysis shows that the temperature fluctuation in the central business district in the monitoring area is large, while the temperature change in the remote suburbs is small, then the central business district is identified as a high fluctuation area, and the remote suburbs are low fluctuation areas.

[0110] Step S134: performing a region intersection operation on the sensor dense area data and the low sensor data fluctuation area data, so as to obtain the dense sensor area data; performing a region intersection operation on the sensor sparse area data and the high sensor data fluctuation area data, so as to obtain the sparse sensor area data;

[0111] In this embodiment, a spatial area intersection operation is performed based on the aforementioned dense area and fluctuating area data. Specifically, a geographic information system (GIS) tool can be used to calculate the area intersection through spatial analysis tools (such as buffer analysis and overlapping area analysis). Suppose that area A is identified as a sensor-dense area, area B is identified as a low-fluctuation area, and area C is identified as a high-fluctuation area. Through the intersection operation, the intersection of A and B (that is, the area that belongs to the dense area and has small data fluctuations at the same time) can obtain the dense sensing area data; and the intersection of the sparse area and the high-fluctuation area can obtain the sparse sensing area data.

[0112] Step S135: performing regional sensor resolution selection on densely sensed area data according to the ground sensor resolution data, thereby obtaining sensor low-resolution area data; performing regional sensor resolution selection on sparsely sensed area data according to the ground sensor resolution data, thereby obtaining sensor high-resolution area data;

[0113] In this embodiment, according to the resolution data of the sensor, a suitable resolution is selected and applied to different areas. Because there are more sensors distributed in dense areas, high-resolution sensors are usually not required to increase costs or computing burdens, so a lower resolution (for example, a resolution of 0.5 meters or 1 meter) can be selected. Since there are fewer sensors in sparse areas, a higher resolution is required to ensure the accuracy and validity of the data, so a higher resolution (for example, a resolution of 0.1 meters) can be selected. Assume that in a monitoring area, the sensor resolution in the dense sensing area is 1 meter, while in the sparse sensing area, a resolution of 0.5 meters is selected to improve data accuracy.

[0114] Step S136: performing regional spatial merging on the sensor low-resolution regional data and the sensor high-resolution regional data, thereby obtaining the monitoring area sensor resolution data.

[0115] In this embodiment, the data of the low-resolution area and the high-resolution area are merged to obtain the sensor resolution data of the entire monitoring area. When merging, it is necessary to ensure the spatial alignment of data of different resolutions. Usually, interpolation or data fusion technology is used to merge the data of the high-resolution area into the low-resolution area, or vice versa, to ensure that the final data set meets the accuracy requirements of the monitoring area. Interpolation methods (such as bilinear interpolation or spline interpolation) can be used to smoothly transition data of different resolutions, thereby obtaining a unified resolution level.

[0116] Optionally, step S2 specifically includes:

[0117] Step S21: acquiring remote sensing data of the monitoring area, and extracting remote sensing spectral features of the monitoring area from the remote sensing data of the monitoring area, thereby obtaining spectral data of the monitoring area;

[0118] In this embodiment, remote sensing satellites (such as Landsat 8, Sentinel-2, etc.) are used to obtain remote sensing image data of the target area. Image data can be obtained through remote sensing data providers (such as USGS or ESA), and data types include visible light bands, near infrared bands, etc. Data of different bands are extracted from remote sensing images, and the spectral reflectance value of each pixel is calculated. By analyzing data of different bands (such as red, green, blue, near infrared, etc.), spectral information of vegetation, water bodies, buildings and other characteristics of the area can be extracted. For example, standard vegetation index (NDVI), water body index (NDWI) and other methods are used to extract spectral features for each pixel in the image. NDVI (normalized difference vegetation index) characterizes the health status of vegetation by the ratio of near infrared to red bands. Sentinel-2 satellite data is used to obtain full-color and multispectral data of the monitoring area at a spatial resolution of 20m. By calculating NDVI, vegetation health information of each pixel is extracted. The spectral data obtained by this step will be used for subsequent regional grid division and land use classification.

[0119] Step S22: performing spectral band combination on the spectral data of the monitoring area, so as to obtain regional vegetation index data, regional water index data and regional building index data;

[0120] In this embodiment, multiple spectral bands in the remote sensing data are combined to generate new index data. For example, vegetation index data can calculate NDVI by combining the near infrared band (NIR) and the red band (Red); water index data can calculate NDWI by combining the green band (Green) and the near infrared band (NIR); building index can obtain building feature data by combining the near infrared band and the short-wave infrared band. Vegetation index data (NDVI): NDVI=\frac{NIR-Red}{NIR+Red}; wherein NIR is the near infrared band data, and Red is the red band data. Water index data (NDWI): NDWI=\frac{Green-NIR}{Green+NIR}; wherein Green is the green band, and NIR is the near infrared band. Building index (NDBI): NDBI =\frac{SWIR-NIR}{SWIR+NIR}; wherein SWIR is the short-wave infrared band, and NIR is the near infrared band. Assuming that the monitoring area includes different environments such as cities, forests and water bodies, the red, near-infrared, green and short-wave infrared bands in the Sentinel-2 data are used to calculate the NDVI, NDWI and NDBI data of each pixel to distinguish and extract the different regional characteristics of vegetation, water bodies and buildings.

[0121] Step S23: dividing the spectral data of the monitoring area into vegetation area grids according to the regional vegetation index data, thereby obtaining vegetation area grid data; dividing the spectral data of the monitoring area into water area grids according to the regional water body index data, thereby obtaining water body area grid data; dividing the spectral data of the monitoring area into building area grids according to the regional building index data, thereby obtaining building area grid data;

[0122] In this embodiment, according to the calculated vegetation index (such as NDVI), the pixels in the image are divided into vegetation areas and non-vegetation areas by setting a threshold. For example, pixels with NDVI greater than 0.2 are vegetation areas, and pixels with NDVI less than 0.2 are non-vegetation areas. The vegetation area is divided into several grids, each grid containing a specific vegetation area. Similarly, according to the water body index (such as NDWI), a threshold is set to delineate the water body area. If the NDWI is greater than 0.2, it is a water body area, and if it is lower than this value, it is a non-water body area. According to the distribution of the water body area, gridding is performed to ensure that each grid represents an independent water body block. The image is gridded using a building index (such as NDBI). Areas with higher NDBI values ​​represent building areas, and building area grids are divided by thresholds. Assume that the NDVI value of a city monitoring area is 0.5, the NDWI of the water body area is 0.3, and the NDBI of the building area is 0.4. After dividing the grid according to these values, the vegetation area grid will be composed of the part with NDVI greater than 0.2, the water area grid will be composed of the part with NDWI greater than 0.2, and the building area grid will be composed of the part with NDBI greater than 0.3.

[0123] Step S24: performing regional grid spatial integration on the vegetation area grid data, the water area grid data and the building area grid data, so as to obtain the monitoring area land grid data;

[0124] In this embodiment, different types of regional grids such as vegetation, buildings, and water bodies are integrated according to spatial positions. Each grid should contain its corresponding land use type (vegetation, water body or building), and a comprehensive land grid data set is generated through spatial alignment and overlapping area integration. The data of vegetation regional grids, water body regional grids, building regional grids, etc. are merged into a land grid data file. The attributes of each grid block contain its corresponding land type (such as vegetation, building, water body, etc.). For example, in the monitoring area of ​​a certain city, the vegetation regional grid occupies 40% of the total area, the water body regional grid occupies 20%, the building regional grid occupies 30%, and the remaining 10% is open area. These three types of regional grids are combined to form a land grid data set containing three types of areas.

[0125] Step S25: Calculate the land vegetation coverage rate of the monitoring area according to the land grid data of the monitoring area, so as to obtain the land vegetation coverage rate data of the monitoring area.

[0126] In this embodiment, the proportion of vegetation in the monitoring area is calculated based on the number of vegetation area grids in the land grid data set. The vegetation coverage rate can be determined by the ratio of the area of ​​the vegetation area grid to the total grid area. \text{Vegetation coverage rate} = \frac{\text{Number of vegetation coverage grids}}{\text{Total number of grids}} \times 100\%. Assuming that the total area of ​​the monitoring area is 1,000 square kilometers, of which the area of ​​the vegetation area grid is 600 square kilometers, the vegetation coverage rate is 60%. In the area around the city, by analyzing the vegetation index data of different grids, it is assumed that 60% of the grids are identified as vegetation areas. Through area calculation, it is concluded that the land vegetation coverage rate of the area is 60%, which indicates that the area has a relatively high vegetation coverage rate.

[0127] Optionally, step S25 is specifically:

[0128] Step S251: extracting grid vegetation index features from the land grid data in the monitoring area, thereby obtaining land grid vegetation index data;

[0129] In this embodiment, the land grid data of the monitoring area is obtained from remote sensing images or high-resolution images taken by drones. The area is divided into several grid cells through image processing technology. The size of each grid cell can be set according to actual needs, for example, a square grid with a side length of 50 meters. Then, common vegetation index calculation methods such as normalized vegetation index (NDVI) or enhanced vegetation index (EVI) are used to extract the vegetation characteristics of each grid cell. The specific calculation method is: NDVI=(NIR-RED) / (NIR+RED), where NIR is the reflectivity of the near-infrared band and RED is the reflectivity of the red band. An NDVI value is calculated for each grid cell, representing the vegetation coverage of the area. In this way, the generated grid vegetation index data will reflect the vegetation growth status of the monitoring area.

[0130] Step S252: grid classification of land grid vegetation index data according to a preset vegetation index threshold, thereby obtaining vegetation coverage grid data and non-vegetation coverage grid data;

[0131] In this embodiment, a vegetation index threshold is set by analyzing the distribution of NDVI values. For example, when the NDVI value is greater than 0.2, the grid unit is considered to be a vegetation-covered area; when the NDVI value is less than or equal to 0.2, the grid unit is considered to be a non-vegetation-covered area. According to this standard, the NDVI values ​​of all grid cells are compared with the threshold and classified into two categories: vegetation-covered grids (NDVI values> 0.2) and non-vegetation-covered grids (NDVI values≤ 0.2). After classification, the number of vegetation-covered grids and non-vegetation-covered grids are counted respectively to form corresponding vegetation-covered grid data and non-vegetation-covered grid data. These data can be further used to monitor the distribution and coverage of vegetation in the area.

[0132] Step S253: Calculate the vegetation coverage grid ratio based on the vegetation coverage grid data and the non-vegetation coverage grid data, so as to obtain the land vegetation coverage rate data of the monitoring area.

[0133] In this embodiment, the vegetation coverage rate is calculated by counting the number of vegetation-covered grids and non-vegetation-covered grids. The specific method is: first, the total number of grids in the monitoring area is counted (for example, assuming that there are a total of 1000 grids in the monitoring area), and then the number of vegetation-covered grids is counted (assuming that there are 600 grids), and the number of non-vegetation-covered grids is 400 grids. Next, the vegetation coverage rate is calculated, that is, the ratio of vegetation-covered grids to total grids, and the formula is: \text{vegetation coverage rate} = \frac{\text{Number of vegetation-covered grids}}{\text{Total number of grids}} \times 100\%. In this example, vegetation coverage rate = (600 / 1000) × 100% = 60%. This obtains the land vegetation coverage rate data of the monitoring area. This indicator helps to evaluate the vegetation coverage status of the area and provide data support for environmental protection, land use planning, etc.

[0134] Optionally, step S3 specifically includes:

[0135] Step S31: collecting real-time sensor data of the monitoring area based on the monitoring area sensor network, thereby obtaining real-time sensor data of the monitoring area;

[0136] In this embodiment, the sensor network is composed of several wireless sensor nodes, which are distributed in different locations of the monitoring area. These sensor nodes include temperature and humidity sensors, light sensors, soil moisture sensors and meteorological sensors. Each sensor node collects environmental data in real time wirelessly and transmits the data to the central data collection center through a self-organizing network. The data collection center receives real-time data uploaded by the sensor node every 5 minutes, including parameters such as ambient temperature, humidity, light intensity and soil moisture.

[0137] Step S32: dividing the real-time sensing data of the monitoring area into the monitoring area grid sensing data according to the land grid data of the monitoring area, thereby obtaining the monitoring area grid sensing data;

[0138] In this embodiment, the land in the monitoring area is managed according to a pre-set grid division, for example, the monitoring area is divided into 100m x 100m grid units, each grid containing one or more sensor nodes. Assuming that the total area of ​​the agricultural monitoring area is 100 square kilometers, the grid data will be defined as the center coordinates, area, and storage information of the sensor data in each grid area. During the data processing process, the real-time data collected from the sensor nodes are classified according to the grid division rules, and the sensor data in each grid is integrated to generate a sensor data report for each grid.

[0139] Step S33: performing grid light intensity feature extraction and grid soil moisture feature extraction on the grid sensing data of the monitoring area, thereby obtaining grid light intensity data and grid soil moisture data;

[0140] In this embodiment, for the sensor data in each grid, feature extraction is performed through a specific algorithm. Taking light intensity as an example, the light intensity data collected by the light sensor can be statistically analyzed through features such as average value, maximum value and minimum value, thereby generating light intensity feature data of the grid. In addition, for soil moisture, the data collected by the soil moisture sensor is used, and the sliding window algorithm is used to calculate the average value, change rate and volatility of soil moisture, thereby obtaining soil moisture feature data of each grid. These feature data provide input data for subsequent vegetation coverage estimation.

[0141] Step S34: estimating the grid land vegetation coverage rate according to the grid light intensity data and the grid soil moisture, thereby obtaining the estimated data of the land vegetation coverage rate in the monitoring area;

[0142] In this embodiment, based on the grid light intensity data and soil moisture data, combined with the light and soil moisture data in the grid, the vegetation coverage of each grid is estimated through regression analysis and statistical models. For the different effects of light intensity and soil moisture, the estimation result can be optimized by weight adjustment to obtain the estimated value of vegetation coverage of each grid.

[0143] Step S35: Calculate the vegetation coverage error of the estimated data of land vegetation coverage in the monitoring area and the data of land vegetation coverage in the monitoring area, so as to obtain the error data of land vegetation coverage in the monitoring area.

[0144] In this embodiment, the estimated data of land vegetation coverage rate in the monitoring area is compared with the vegetation coverage of each corresponding grid of the land vegetation coverage rate data in the monitoring area, and the difference obtained by the comparison is the error data of land vegetation coverage rate in the monitoring area.

[0145] Optionally, step S34 is specifically:

[0146] Step S341: Acquire plant growth characteristic data;

[0147] In this embodiment, plant growth characteristic data is obtained through a plant expert experience database (such as a botanical database, an agricultural technology database, etc.), including but not limited to information such as the plant's light requirements, soil moisture adaptation range, temperature suitability, and growth cycle. For example, by querying the plant growth characteristic data in a specific area, the system can obtain local plant species suitable for growth and their growth requirements in different seasons. The data obtained through this database can support subsequent grid light intensity and soil moisture judgments on plant suitability. The plant expert database has the ability to cover a wide range of plant species and detailed growth data, such as growth parameters of different plants under different climatic conditions, soil conditions suitable for growth, etc. And the data needs to be kept updated to reflect the latest research results and practical experience.

[0148] Step S342: performing grid-suitable plant identification on the grid light intensity data and the grid soil moisture according to the plant growth characteristic data, thereby obtaining grid-suitable plant characteristic data;

[0149] In this embodiment, based on the acquired plant growth characteristic data, combined with the grid light intensity and soil moisture data in the actual monitoring area, grid suitable plants are identified. By comparing these data with the data in the plant growth characteristic database, the suitable plant species in each grid are identified. For example, if the light intensity of a grid is 200 lux and the soil moisture is 30%, and the suitable light intensity for a plant species (such as wheat) in the database is 150-250 lux and the soil moisture is 20%-40%, then the grid is suitable for wheat growth.

[0150] Step S343: extracting grid monitoring video from the grid sensing data of the monitoring area to obtain grid real-time monitoring video, and dividing the grid real-time monitoring video into video frames to obtain grid monitoring video frames;

[0151] In this embodiment, a real-time video stream of the area where the grid sensor data is located is obtained by a high-definition camera or drone fixed in the monitoring area. The monitoring area may be large, so the video data is usually divided into multiple grid areas, and the boundary of each grid is determined according to the location where the sensor is deployed. For each grid, the system extracts a monitoring video, such as a video recorded every hour for each grid. Each video will contain the environment, vegetation conditions and other relevant data within the grid. For example, a video may show the growth conditions and vegetation coverage of different plants in a grid.

[0152] Step S344: performing edge detection according to the grid monitoring video frame to obtain edge feature data of the grid monitoring video frame, and performing feature similarity calculation according to the grid suitable plant feature data and the grid monitoring video frame edge feature data to obtain grid plant feature similarity data;

[0153] In this embodiment, image processing technology is used to perform edge detection on each frame of monitoring video image. Common edge detection algorithms include Sobel operator, Canny edge detection, etc. The purpose of edge detection is to extract contour features in the image, especially the edges of vegetation areas, to help identify vegetation pixels later. The video frame is processed by an edge detection algorithm (such as Canny edge detection, Sobel operator, etc.) to extract the plant edge features in the video frame. Through these edge data, the plant morphology can be further analyzed and compared with the known plant growth feature data. Feature similarity calculation is performed based on the grid-appropriate plant feature data (such as plant morphological features, leaf structure, etc.) and the edge feature data of the grid monitoring video frame. The specific method of similarity calculation can use cosine similarity or Euclidean distance to evaluate the similarity between two feature vectors. In practical applications, for each frame of video image, the plant edge feature data in the video frame will be extracted by processing using the Canny operator. Assuming that a video frame of a certain grid shows a piece of grass, the coordinate data of the edge of the grass will be obtained after edge detection as the edge feature data of the video frame.

[0154] Step S345: performing vegetation pixel recognition on the grid monitoring video frame according to the grid plant feature similarity data, thereby obtaining a grid vegetation pixel marked frame;

[0155] In this embodiment, the matching degree between the vegetation in the video frame and the suitable plants is confirmed by calculating the similarity. The edge feature data is matched with the plant growth feature data, and the similarity between them is calculated using an algorithm (such as cosine similarity, Euclidean distance, etc.). If the edge features in the monitoring video frame of a certain grid are similar to the growth feature data of a certain plant in the database, the system will consider the grid to be suitable for the growth of the plant. For example, if the edge detection results show that the edge features in a certain grid area are similar to the growth form of wheat, and the light intensity and soil moisture conditions meet the growth requirements of wheat, then the plant feature similarity of the grid will be calculated to be high.

[0156] Step S346: Estimating the grid vegetation coverage pixel ratios of the grid vegetation pixel marking frames and the grid monitoring video frames, thereby obtaining land vegetation coverage estimation data of the monitoring area.

[0157] In this embodiment, the vegetation coverage rate of each grid is obtained by calculating the ratio of the number of vegetation pixels in the grid vegetation pixel mark frame to the total number of pixels. For example, in a video frame of a certain grid, after pixel recognition, it is found that there are 5,000 vegetation pixels and a total of 10,000 pixels, and the total vegetation coverage rate is 50%. This estimation can help environmental monitoring personnel quickly evaluate the vegetation coverage of the area, and then provide a basis for decisions such as land use planning and crop planting.

[0158] Optionally, step S4 is specifically:

[0159] Step S41: performing error grid extraction on the error data of land vegetation coverage in the monitoring area, thereby obtaining vegetation coverage error grid data;

[0160] In this embodiment, statistical analysis is performed on grids with vegetation coverage errors in each grid, and finally error grid data of vegetation coverage is generated. Each grid contains a corresponding error value as a basis for subsequent analysis.

[0161] Step S42: extracting grid vegetation index error from vegetation coverage error grid data, thereby obtaining grid vegetation index error data;

[0162] In this embodiment, based on the obtained vegetation coverage error grid data, the error amount is further extracted. For example, the normalized difference vegetation index (NDVI) error value is used to reflect the size of the error. Specifically, for each grid, the NDVI value error of all pixels in the grid is first calculated, and then the average of the error values ​​is used as the vegetation index error amount of the grid.

[0163] Step S43: extracting resolution features of the real-time sensing data of the monitoring area and the remote sensing data of the monitoring area, respectively, so as to obtain the sensing resolution data of the monitoring area and the remote sensing resolution data of the monitoring area;

[0164] In this embodiment, resolution analysis is performed on the real-time sensing data and remote sensing data of the monitoring area. For sensor data, resolution feature extraction can be defined by parameters such as the spatial resolution, temporal resolution, and spectral resolution of the sensor. For example, assume that the spatial resolution of the sensor is 1 meter, the temporal resolution is once per hour, and the spectral resolution is 4 bands. Then, the extracted sensor resolution data may contain these key features. For remote sensing data, the spatial resolution of the remote sensing image (for example, 30-meter pixels) can be used for extraction. Through the metadata of the remote sensing image, information such as the spatial resolution, temporal resolution, and spectral resolution of the image is obtained to form "remote sensing resolution data." These data provide a basis for subsequent error analysis.

[0165] Step S44: performing error grid resolution difference calculation on the monitoring area sensing resolution data and the monitoring area remote sensing resolution data according to the vegetation coverage error grid data, thereby obtaining error grid resolution difference data;

[0166] In this embodiment, the obtained vegetation coverage error grid data is differenced with the resolution data in step S43 to obtain error grid resolution difference data. For each grid, the vegetation coverage error of the grid, as well as the sensor resolution and remote sensing resolution corresponding to the grid are first obtained. According to the resolution difference between the sensor and remote sensing data, the resolution difference is calculated. For example, if the spatial resolution of the sensor is 1 meter and the spatial resolution of the remote sensing image is 30 meters, the resolution difference is 29 meters. These difference data will be calculated in each grid unit to obtain "error grid resolution difference data", which can reflect the impact of resolution difference on vegetation coverage error.

[0167] Step S45: estimating the resolution impact factor based on the error grid resolution difference data and the grid vegetation index error data, thereby obtaining the resolution impact factor.

[0168] In this embodiment, a statistical or regression model is used to estimate the impact factor of resolution on vegetation coverage error. Linear regression or multivariate regression analysis is used to establish a relationship model between the error grid resolution difference and the vegetation index error. For example, it is assumed that it can be concluded through regression analysis that for every 10-meter increase in resolution difference, the vegetation coverage error increases by 0.05 (or other statistical coefficients). Based on the model results, the "resolution impact factor" of each grid is calculated, that is, the degree of influence of resolution difference on vegetation coverage error. This impact factor can help determine the impact of remote sensing images or sensor data of different resolutions on the accuracy of vegetation coverage estimation, and then guide data accuracy optimization and improvement.

[0169] Optionally, step S5 specifically includes:

[0170] Step S51: performing geographic coordinate system spatial error correction on the land vegetation coverage rate data of the monitoring area according to the coordinate system data of the monitoring area, thereby obtaining spatially corrected land vegetation coverage rate data;

[0171] In this embodiment, the land vegetation coverage data of the monitoring area is obtained from the remote sensing image data. These data are obtained based on different coordinate systems (such as UTM, WGS84, etc.), so it is necessary to perform spatial coordinate system conversion. Assume that the original data of the monitoring area uses a local coordinate system (such as UTM coordinate system), and needs to be converted into a geographic coordinate system (such as WGS-84 coordinate system). In order to perform spatial error correction, first use a high-precision geographic information system (GIS) tool to select several ground control points with known coordinates in the area as a benchmark for coordinate conversion. Use a geographic information system (such as ArcGIS or QGIS) to import the original data, and perform coordinate system conversion based on the selected control point data to ensure that the original land vegetation coverage data matches the geographic coordinate system and avoid vegetation coverage data deviations caused by coordinate errors. For example, the coordinates of a certain monitoring point may have a deviation of several meters. After correction, its vegetation coverage data will be more consistent with the actual situation.

[0172] Step S52: standardizing the spatially corrected land vegetation coverage data according to the resolution influencing factor, thereby obtaining standardized regional land vegetation coverage data;

[0173] In the present embodiment, the purpose of standardization is to eliminate the influence caused by resolution difference. Vegetation coverage data first need to be standardized according to resolution influence factor. Assume that the spatial resolution of original data is 30 meters, and the vegetation data resolution of some monitoring areas is higher or lower. In this process, first determine the resolution influence factor, which can usually be carried out by comparing the vegetation coverage values ​​of different resolution data sets. Taking 30 meters of resolution as an example, spatial resampling method can be applied by spatial analysis tools (such as ArcGIS or ENVI software), high-resolution data can be adjusted to target resolution, and standardized coverage value can be calculated according to resolution factor. Specifically, the original data value can be normalized, mapped to the interval between 0 and 1, representing the range of coverage. For example, if the original vegetation coverage of a region is 70%, and the data resolution of the region is higher (10 meters), then after its standardization, the standardized value is recalculated.

[0174] Step S53: performing time series analysis based on the standardized regional land vegetation coverage data, thereby obtaining regional land vegetation coverage time series data;

[0175] In this embodiment, based on the standardized vegetation coverage data, this step performs time series analysis, mainly by analyzing the changing trend of the vegetation coverage in the area over a period of time. Assume that there are monthly data for the past five years in the area (for example, from January 2019 to December 2023), and standardized data on land vegetation coverage are available every month. On this basis, firstly, the vegetation coverage data is smoothed using a time series modeling method (such as seasonal trend decomposition method STL, or ARIMA model), seasonal fluctuations and random errors are eliminated, and long-term trends and periodic fluctuations are extracted. Taking a certain area as an example, the analysis results may show that due to climate change in winter, the vegetation coverage rate has decreased, and gradually recovered in spring and autumn. Then, through time series analysis, the vegetation changes in the next few months can be further predicted.

[0176] Step S54: Perform vegetation coverage trend analysis based on the regional land vegetation coverage time series data, so as to obtain regional land vegetation coverage trend data.

[0177] In this embodiment, based on the land vegetation coverage data obtained by time series analysis, trend analysis can be further performed to reveal the long-term change trend of regional vegetation coverage. The main purpose of trend analysis is to find out whether the vegetation coverage shows a trend of continuous increase or decrease, and the relationship between this trend and factors such as climate change and human activities. Use statistical analysis software (such as Pandas, NumPy library in R and Python, or MATLAB, etc.) to fit the trend line. For example, the least squares method is used to perform linear regression analysis on the time series data to calculate the average annual growth rate or decline rate of vegetation coverage. Taking a certain area as an example, it is assumed that the average annual growth rate of vegetation coverage in the past 10 years in this area is 0.5%. According to the trend analysis results, it can be further inferred that the possible change trend of the future vegetation coverage in the region in the next five years, and provide reference for relevant decisions. Through trend analysis, if it is found that the vegetation coverage has declined for a long time, it may be necessary to take corresponding ecological restoration measures, such as strengthening vegetation restoration or controlling over-exploitation.

[0178] Optionally, the present specification also provides a land use dataset processing system, which is used to execute the land use dataset processing method as described above, and the land use dataset processing system includes:

[0179] The sensor data fusion module is used to obtain the ground sensor data of the monitoring area, and construct the monitoring area coordinate system according to the ground sensor data of the monitoring area, so as to obtain the monitoring area coordinate system data; based on the monitoring area coordinate system data, the sensor data of the ground sensor data of the monitoring area is fused to obtain the monitoring area sensor network;

[0180] The vegetation coverage calculation module is used to obtain remote sensing data of the monitoring area, and divide the land in the monitoring area into grids based on the remote sensing data of the monitoring area, so as to obtain land grid data of the monitoring area; calculate the land vegetation coverage of the monitoring area according to the land grid data of the monitoring area, so as to obtain the land vegetation coverage data of the monitoring area;

[0181] The vegetation coverage error calculation module is used to collect real-time sensing data of the monitoring area based on the monitoring area sensor network, so as to obtain real-time sensing data of the monitoring area, and to calculate the vegetation coverage error based on the real-time sensing data of the monitoring area and the land vegetation coverage data of the monitoring area, so as to obtain the land vegetation coverage error data of the monitoring area;

[0182] A resolution impact factor estimation module is used to estimate the resolution impact factor based on the land vegetation coverage rate data of the monitoring area and the land vegetation coverage rate error data of the monitoring area, so as to obtain the resolution impact factor;

[0183] The regional land vegetation coverage correction module is used to correct the regional land vegetation coverage data of the monitoring area according to the resolution influencing factor and the monitoring area coordinate system data, so as to obtain standardized regional land vegetation coverage data, and perform vegetation coverage trend analysis based on the standardized regional land vegetation coverage data, so as to obtain regional land vegetation coverage trend data.

[0184] The land use data set processing system of the present invention can implement any land use data set processing method of the present invention, and is used to combine the operations between various modules and the medium of signal transmission to complete the land use data set processing method. The internal modules of the system cooperate with each other to improve the processing accuracy of land use data.

[0185] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0186] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A land use data set processing method, characterized in that: The following steps are involved: Step S1: Acquire ground sensor data of the monitoring area, and construct a monitoring area coordinate system according to the ground sensor data of the monitoring area, so as to obtain monitoring area coordinate system data; Based on the coordinate system data of the monitoring area, the ground sensor data of the monitoring area is fused to obtain the sensor network of the monitoring area; Step S2: Acquire remote sensing data of the monitoring area, and divide the land in the monitoring area into grids based on the remote sensing data of the monitoring area, thereby obtaining land grid data of the monitoring area; calculate the land vegetation coverage rate of the monitoring area according to the land grid data of the monitoring area, thereby obtaining land vegetation coverage rate data of the monitoring area; Step S3: collecting real-time sensing data of the monitoring area based on the sensing network of the monitoring area, thereby obtaining real-time sensing data of the monitoring area, and estimating the vegetation coverage rate of the grid land according to the real-time sensing data of the monitoring area, thereby obtaining the estimated data of the vegetation coverage rate of the land in the monitoring area; calculating the vegetation coverage rate error of the estimated data of the vegetation coverage rate of the monitoring area and the vegetation coverage rate data of the monitoring area, thereby obtaining the error data of the vegetation coverage rate of the land in the monitoring area; Step S4: estimating the resolution impact factor according to the land vegetation coverage rate data of the monitoring area and the land vegetation coverage rate error data of the monitoring area, thereby obtaining the resolution impact factor; Step S4 is specifically as follows: Step S41: extracting error grids from the vegetation coverage error data of the monitored area, and statistically analyzing the grids with vegetation coverage errors in each grid, thereby obtaining vegetation coverage error grid data; Step S42: extracting grid vegetation index error from the vegetation coverage error grid data, calculating the NDVI error values ​​of all pixels in the grid, and then taking the average of the NDVI error values ​​in the grid as the vegetation index error of the grid, thereby obtaining grid vegetation index error data; Step S43: extracting resolution features of the real-time sensing data of the monitoring area and the remote sensing data of the monitoring area, respectively, so as to obtain the sensing resolution data of the monitoring area and the remote sensing resolution data of the monitoring area; Step S44: performing error grid resolution difference calculation on the monitoring area sensing resolution data and the monitoring area remote sensing resolution data according to the vegetation coverage error grid data, thereby obtaining error grid resolution difference data; Step S45: estimating the resolution impact factor based on the error grid resolution difference data and the grid vegetation index error data, thereby obtaining the resolution impact factor; Step S5: performing regional land vegetation coverage correction on the land vegetation coverage data of the monitoring area according to the resolution influencing factor and the coordinate system data of the monitoring area, thereby obtaining standardized regional land vegetation coverage data, and performing vegetation coverage trend analysis according to the standardized regional land vegetation coverage data, thereby obtaining regional land vegetation coverage trend data; Step S5 is specifically as follows: Step S51: performing geographic coordinate system spatial error correction on the land vegetation coverage rate data of the monitoring area according to the coordinate system data of the monitoring area, thereby obtaining spatially corrected land vegetation coverage rate data; Step S52: standardizing the spatially corrected land vegetation coverage data according to the resolution influencing factor, thereby obtaining standardized regional land vegetation coverage data; Step S53: performing time series analysis based on the standardized regional land vegetation coverage data, thereby obtaining regional land vegetation coverage time series data; Step S54: Perform vegetation coverage trend analysis based on the regional land vegetation coverage time series data, so as to obtain regional land vegetation coverage trend data.

2. The land use dataset processing method according to claim 1, characterized in that: Step S1 is specifically as follows: Step S11: acquiring ground sensor data in the monitoring area, and extracting ground sensor spatial position features from the ground sensor data in the monitoring area, thereby obtaining ground sensor spatial position data; Step S12: constructing a monitoring area coordinate system according to the spatial position data of the ground sensor, thereby obtaining monitoring area coordinate system data; Step S13: selecting the monitoring area sensor resolution based on the monitoring area coordinate system data, thereby obtaining the monitoring area sensor resolution data; Step S14: resampling the ground sensing data in the monitoring area based on the sensor resolution data in the monitoring area, thereby obtaining the sensing data in the monitoring area; Step S15: performing sensor data fusion on the monitoring area sensor data according to the monitoring area coordinate system data, thereby obtaining a monitoring area sensor network.

3. The land use dataset processing method according to claim 2, characterized in that: Step S13 is specifically as follows: Step S131: performing sensor spatial distribution statistics on the spatial position data of the ground sensors based on the monitoring area coordinate system data, thereby obtaining sensor dense area data and sensor sparse area data; Step S132: extracting sensor resolution features from the ground sensor data in the monitoring area, thereby obtaining ground sensor resolution data; Step S133: performing regional sensor data fluctuation statistics on the ground sensor data in the monitoring area, thereby obtaining data of a high sensor data fluctuation area and data of a low sensor data fluctuation area; Step S134: performing a region intersection operation on the sensor dense area data and the low sensor data fluctuation area data, thereby obtaining dense sensor area data; Performing a regional intersection operation on the sensor sparse area data and the high sensor data fluctuation area data to obtain the sparse sensor area data; Step S135: performing regional sensor resolution selection on the densely sensed regional data according to the ground sensor resolution data, thereby obtaining sensor low-resolution regional data; Selecting the regional sensor resolution for the sparse sensing regional data according to the ground sensor resolution data, thereby obtaining the sensor high-resolution regional data; Step S136: performing regional spatial merging on the sensor low-resolution regional data and the sensor high-resolution regional data, thereby obtaining the monitoring area sensor resolution data.

4. The land use dataset processing method according to claim 1, characterized in that: Step S2 is specifically as follows: Step S21: acquiring remote sensing data of the monitoring area, and extracting remote sensing spectral features of the monitoring area from the remote sensing data of the monitoring area, thereby obtaining spectral data of the monitoring area; Step S22: performing spectral band combination on the spectral data of the monitoring area, so as to obtain regional vegetation index data, regional water index data and regional building index data; Step S23: dividing the spectral data of the monitoring area into vegetation area grids according to the regional vegetation index data, thereby obtaining vegetation area grid data; Divide the spectral data of the monitoring area into water body regional grids according to the regional water body index data, so as to obtain water body regional grid data; Divide the spectral data of the monitoring area into building area grids according to the regional building index data, so as to obtain building area grid data; Step S24: performing regional grid spatial integration on the vegetation area grid data, the water area grid data and the building area grid data, so as to obtain the monitoring area land grid data; Step S25: Calculate the land vegetation coverage rate of the monitoring area according to the land grid data of the monitoring area, so as to obtain the land vegetation coverage rate data of the monitoring area.

5. The land use dataset processing method according to claim 4, characterized in that: Step S25 is specifically as follows: Step S251: extracting grid vegetation index features from the land grid data in the monitoring area, thereby obtaining land grid vegetation index data; Step S252: grid classification of land grid vegetation index data according to a preset vegetation index threshold, thereby obtaining vegetation coverage grid data and non-vegetation coverage grid data; Step S253: Calculate the vegetation coverage grid ratio based on the vegetation coverage grid data and the non-vegetation coverage grid data, so as to obtain the land vegetation coverage rate data of the monitoring area.

6. The land use dataset processing method according to claim 1, characterized in that: Step S3 is specifically as follows: Step S31: collecting real-time sensor data of the monitoring area based on the monitoring area sensor network, thereby obtaining real-time sensor data of the monitoring area; Step S32: dividing the real-time sensing data of the monitoring area into the monitoring area grid sensing data according to the land grid data of the monitoring area, thereby obtaining the monitoring area grid sensing data; Step S33: performing grid light intensity feature extraction and grid soil moisture feature extraction on the grid sensing data of the monitoring area, thereby obtaining grid light intensity data and grid soil moisture data; Step S34: estimating the grid land vegetation coverage rate according to the grid light intensity data and the grid soil moisture, thereby obtaining the estimated data of the land vegetation coverage rate in the monitoring area; Step S35: Calculate the vegetation coverage error of the estimated data of land vegetation coverage in the monitoring area and the data of land vegetation coverage in the monitoring area, so as to obtain the error data of land vegetation coverage in the monitoring area.

7. The land use dataset processing method according to claim 6, characterized in that: Step S34 is specifically as follows: Step S341: Acquire plant growth characteristic data; Step S342: performing grid-suitable plant identification on the grid light intensity data and the grid soil moisture according to the plant growth characteristic data, thereby obtaining grid-suitable plant characteristic data; Step S343: extracting grid monitoring video from the grid sensing data of the monitoring area to obtain grid real-time monitoring video, and dividing the grid real-time monitoring video into video frames to obtain grid monitoring video frames; Step S344: performing edge detection according to the grid monitoring video frame to obtain edge feature data of the grid monitoring video frame, and performing feature similarity calculation according to the grid suitable plant feature data and the grid monitoring video frame edge feature data to obtain grid plant feature similarity data; Step S345: performing vegetation pixel recognition on the grid monitoring video frame according to the grid plant feature similarity data, thereby obtaining a grid vegetation pixel marked frame; Step S346: Estimating the grid vegetation coverage pixel ratios of the grid vegetation pixel marking frames and the grid monitoring video frames, thereby obtaining land vegetation coverage estimation data of the monitoring area.

8. A land use data set processing system, characterized in that: Used to execute the land use data set processing method as claimed in claim 1, the land use data set processing system comprises: The sensor data fusion module is used to obtain the ground sensor data of the monitoring area, and construct the monitoring area coordinate system according to the ground sensor data of the monitoring area, so as to obtain the monitoring area coordinate system data; based on the monitoring area coordinate system data, the sensor data of the ground sensor data of the monitoring area is fused to obtain the monitoring area sensor network; The vegetation coverage calculation module is used to obtain remote sensing data of the monitoring area, and divide the land in the monitoring area into grids based on the remote sensing data of the monitoring area, so as to obtain land grid data of the monitoring area; calculate the land vegetation coverage of the monitoring area according to the land grid data of the monitoring area, so as to obtain the land vegetation coverage data of the monitoring area; The vegetation coverage error calculation module is used to collect real-time sensing data of the monitoring area based on the monitoring area sensor network, so as to obtain real-time sensing data of the monitoring area, and to calculate the vegetation coverage error based on the real-time sensing data of the monitoring area and the land vegetation coverage data of the monitoring area, so as to obtain the land vegetation coverage error data of the monitoring area; A resolution impact factor estimation module is used to estimate the resolution impact factor based on the land vegetation coverage rate data of the monitoring area and the land vegetation coverage rate error data of the monitoring area, so as to obtain the resolution impact factor; The regional land vegetation coverage correction module is used to correct the regional land vegetation coverage data of the monitoring area according to the resolution influencing factor and the monitoring area coordinate system data, so as to obtain standardized regional land vegetation coverage data, and perform vegetation coverage trend analysis based on the standardized regional land vegetation coverage data, so as to obtain regional land vegetation coverage trend data.

Images