Land data acquisition system

Abstract

The invention discloses a land data acquisition system, and belongs to the technical field of land data acquisition. Comprising a multi-modal acquisition unit, an acquisition management platform and an interface opening module, the multi-modal acquisition unit comprises a multi-source acquisition terminal, and the multi-source acquisition terminal is connected to an edge computing module; the multi-modal acquisition unit is connected to the acquisition management platform; the acquisition management platform comprises a data area marking unit, a data storage unit, a data fusion unit and a data management unit which are in communication connection with one another; the data fusion unit is connected with a field assignment management unit, and the field assignment management unit is connected with a field assignment library; the data fusion unit is internally provided with a data fusion model, the data fusion model comprises basic fusion and weighted fusion, and the interface opening module is in communication connection with the acquisition management platform; according to the land data acquisition system, accurate acquisition, efficient fusion and cross-field adaptability sharing application of land multi-dimensional data are realized.

Description

Technical Field

[0001] This invention specifically relates to a land data acquisition system, belonging to the field of land data acquisition technology. Background Technology

[0002] Land data is a core foundation for national spatial planning, agricultural production optimization, and ecological environmental protection. The comprehensiveness, accuracy, and utilization efficiency of its collection directly affect the scientific nature of decision-making in related fields. For example, Chinese Patent Publication No. CN120704179A discloses a control system and method for multi-layer soil temperature and humidity sampling in smart agriculture. The system includes a land data acquisition module, an updated data acquisition module, a dataset preprocessing module, a soil moisture prediction module, a decision report generation module, and a report transmission module. It collects land datasets and updated datasets, preprocesses the land datasets and updated datasets to obtain comprehensive feature data. This structure has significant advantages such as high accuracy of sampled soil data, strong decision support capabilities, and good economic benefits. However, existing land data acquisition often adopts a single-terminal acquisition mode, which suffers from incomplete data dimensions and low acquisition efficiency. At the same time, land data from different sources has large differences in format and lacks a unified fusion standard, resulting in poor data correlation and high integration difficulty. In addition, existing systems are mostly closed architectures without open standardized application interfaces, making it difficult for various fields to efficiently call land data and limiting the comprehensive utilization value of land data. Summary of the Invention

[0003] To address the aforementioned issues, this invention proposes a land data acquisition system that enables accurate acquisition, efficient fusion, and cross-domain adaptable sharing and application of multi-dimensional land data.

[0004] The land data acquisition system of the present invention includes: A multimodal acquisition unit includes a multi-source acquisition terminal connected to an edge computing module. The multi-source acquisition terminal and the edge computing module are connected via a 5G / NB-IoT communication protocol, and the edge computing module and the acquisition management platform transmit data via HTTPS protocol to ensure data transmission stability and security. The edge computing module is deployed near the acquisition area and is used to perform noise reduction, outlier removal, and format standardization processing on the raw data acquired by the multi-source acquisition terminal, as detailed below: Noise reduction: Wavelet transform algorithm is used to reduce noise in continuous data collected by soil sensors and meteorological monitoring units to remove noise signals caused by environmental interference. Outlier removal: Based on the 3σ principle, outliers that exceed the normal data range (such as extreme data caused by sensor failure) are identified and removed to ensure data accuracy; Format standardization: Heterogeneous data collected from different terminals (such as numerical data from sensors and image data from remote sensing) are uniformly converted into JSON format, laying the foundation for subsequent uploading and fusion; A data acquisition and management platform is established, to which the multimodal acquisition unit is connected. The platform comprises interconnected data region labeling, data storage, data fusion, and data management units. The data region labeling unit determines the acquired area based on the physical address of the edge computing module and stores the data in the data storage unit after address labeling. The data storage unit uses a distributed storage architecture to store preprocessed raw data and fused data, used for integrating multi-dimensional land data. The data management unit enables data querying, modification, backup, and access control. The data fusion unit is connected to a domain assignment management unit, which in turn is connected to a domain assignment library. The data fusion unit has a built-in data fusion model, which includes basic fusion and weighted fusion. Basic fusion is used for integrating data of the same type, while weighted fusion is used for cross-dimensional fusion of different types of data. The specific details of the basic fusion are as follows: (1) in, The result of fusion of similar types of data, The number of data acquisition terminals of the same type For the first Data collected from each terminal For the first Preset score for the acquisition accuracy of each terminal. The sum of the accuracy scores for data collection from terminals of the same type; The weighted fusion is described in detail below: ; in, For cross-dimensional data fusion results, The number of land data types participating in the integration, For the first The basic fusion result of the data types For the first The weighting coefficients of the data types, and ; An interface opening module is communicatively connected to the data acquisition and management platform. The interface opening module includes an interface management unit, which is connected to a domain request unit, a data region request unit, and a data receiving and output unit.

[0005] Furthermore, the multi-source acquisition terminal adopts a distributed deployment method to cover the target acquisition area, specifically including four types of acquisition units: Soil sensor unit: It adopts buried installation and is distributed in different plots of the collection area to collect core data such as soil moisture, pH value, nitrogen, phosphorus and potassium nutrient content in real time. The collection frequency can be set to 1 time / hour to 1 time / day according to the needs. Topographic surveying unit: integrates GPS positioning module and laser ranging module to collect land elevation, slope and aspect data, with positioning accuracy up to the centimeter level, meeting the needs of topographic analysis; Meteorological monitoring unit: Deployed at the highest point in the data collection area, it collects meteorological data such as temperature, precipitation, wind speed, and sunshine duration to support land ecological assessment; Remote sensing data receiving unit: It periodically receives high-resolution remote sensing image data through a satellite signal receiving antenna to obtain macro data such as land use type and vegetation coverage.

[0006] Furthermore, the data receiving and output unit includes a RESTful API interface and an interface adaptation unit. The interface adaptation unit is used to adapt the output data to the format and match permissions according to the application requirements of different fields. The RESTful API interface is used to provide data query, data download and data subscription services.

[0007] Furthermore, the cloud management platform also includes a data visualization unit, which is used to display the raw data and fused data in the form of charts, heat maps, and 3D models.

[0008] The multi-source acquisition terminal includes a soil sensor unit, a topographic measurement unit, a meteorological monitoring unit, and a remote sensing data receiving unit. The soil sensor unit includes a temperature sensor, a humidity sensor, a pH sensor, and a heavy metal detector, which collects soil physicochemical property data in real time. In use, the soil sensor unit is buried in the soil to be tested to ensure that the sensor is in full contact with the soil. It collects soil temperature, humidity, pH value, and heavy metal content data in real time. The data processing unit preprocesses the raw data to remove noise and outliers.

[0009] Furthermore, the multi-source acquisition terminal includes: An environmental sensing module, which includes a multispectral imaging unit and a miniature weather station; A multimodal acquisition unit is fixed to the adjustment end of a dynamic adjustment mechanism; the multimodal acquisition unit simultaneously acquires soil dielectric constant and topographic features. The controller controls the entire terminal and communicates with the edge computing module. The controller generates environmental baseline data through multispectral scanning; automatically adjusts the sampling depth and frequency according to target parameters; performs synchronous acquisition of multi-source data and edge computing; and preprocesses the raw data, including denoising, normalization and feature extraction.

[0010] Furthermore, the multimodal acquisition unit includes a TOF lidar and a UWB positioning module, a soil dielectric constant sensor and a LIBS spectrometer for completing soil layer acquisition, and a microfluidic chip microbial sensor for completing biolayer acquisition.

[0011] Furthermore, the data acquisition and management platform also includes an intelligent analysis unit. The intelligent analysis unit has a built-in soil assessment model, which is connected to a result database. The soil assessment model performs threshold judgment on real-time data collected by multi-source acquisition terminals, normalizes the threshold differences of various types, and calculates the processed data through weighted fusion. After obtaining fused data, the fused data is used as an index to perform threshold retrieval in the result database and generate a soil quality assessment report.

[0012] Furthermore, the acquisition management platform also includes a prediction model. The prediction model acquires the data types that change on the time axis within the data storage unit, then acquires all data of the corresponding type on the time axis, and uses this data type as the training set and test set to train the prediction model. The trained prediction model then predicts the data at the next time node to obtain the predicted acquisition data.

[0013] Compared with the prior art, the land data acquisition system of the present invention has the following advantages: 1. Multi-source collaborative data acquisition: Integrating multiple data acquisition terminals such as soil, topography, meteorology, and remote sensing, it achieves comprehensive coverage of land data and can be widely used in fields such as precision agriculture, land surveying, ecological environment monitoring, and geological disaster early warning. 2. Efficient data fusion: Data fusion can adapt to the needs of different application scenarios, improve the targeting and accuracy of data fusion, and solve the problems of poor data correlation and high integration difficulty; 3. Cross-domain sharing and application: Open and standardized API interfaces support on-demand data access from various fields, breaking down data silos and enhancing the comprehensive utilization value of land data; 4. Data security and controllability: Data quality is ensured through edge computing preprocessing, and data security is ensured through access control and encrypted transmission, balancing data availability and security. Attached Figure Description

[0014] Figure 1 This is a schematic diagram showing the connections between the modules of the land data acquisition system of the present invention.

[0015] Figure 2 This is a schematic diagram of the data interaction process between the land data acquisition system of the present invention and the external application terminal.

[0016] ‌ Figure 3 This is a schematic diagram of one embodiment of the multi-source acquisition terminal of the present invention.

[0017] ‌ Figure 4 This is a schematic diagram of another embodiment of the multi-source acquisition terminal of the present invention.

[0018] ‌ Figure 5 This is a schematic diagram of the structure of one embodiment of the data acquisition and management platform of the present invention.

[0019] Figure 6 This is a schematic diagram of another embodiment of the data acquisition and management platform of the present invention. Detailed Implementation

[0020] like Figure 1 and Figure 2 The land data acquisition system shown includes: A multimodal acquisition unit includes a multi-source acquisition terminal connected to an edge computing module. The multi-source acquisition terminal and the edge computing module are connected via a 5G / NB-IoT communication protocol, and the edge computing module and the acquisition management platform transmit data via HTTPS protocol to ensure data transmission stability and security. The edge computing module is deployed near the acquisition area and is used to perform noise reduction, outlier removal, and format standardization processing on the raw data acquired by the multi-source acquisition terminal, as detailed below: Noise reduction: Wavelet transform algorithm is used to reduce noise in continuous data collected by soil sensors and meteorological monitoring units to remove noise signals caused by environmental interference. Outlier removal: Based on the 3σ principle, outliers that exceed the normal data range (such as extreme data caused by sensor failure) are identified and removed to ensure data accuracy; Format standardization: Heterogeneous data collected from different terminals (such as numerical data from sensors and image data from remote sensing) are uniformly converted into JSON format, laying the foundation for subsequent uploading and fusion; A data acquisition and management platform is established, to which the multimodal acquisition unit is connected. The platform comprises interconnected data region labeling, data storage, data fusion, and data management units. The data region labeling unit determines the acquired area based on the physical address of the edge computing module and stores the data in the data storage unit after address labeling. The data storage unit uses a distributed storage architecture to store preprocessed raw data and fused data, used for integrating multi-dimensional land data. The data management unit enables data querying, modification, backup, and access control. The data fusion unit is connected to a domain assignment management unit, which in turn is connected to a domain assignment library. The data fusion unit has a built-in data fusion model, which includes basic fusion and weighted fusion. Basic fusion is used for integrating data of the same type, while weighted fusion is used for cross-dimensional fusion of different types of data. The specific details of the basic fusion are as follows: (1) in, The result of fusion of similar types of data, The number of data acquisition terminals of the same type For the first Data collected from each terminal For the first Preset score for the acquisition accuracy of each terminal. The sum of the accuracy scores for data collection from terminals of the same type; The weighted fusion is described in detail below: ; in, For cross-dimensional data fusion results, The number of land data types participating in the integration, For the first The basic fusion result of the data types For the first The weighting coefficients of the data types, and ; An interface opening module is communicatively connected to the data acquisition and management platform. The interface opening module includes an interface management unit, which is connected to a domain request unit, a data region request unit, and a data receiving and output unit.

[0021] The multi-source acquisition terminal adopts a distributed deployment method to cover the target acquisition area, and specifically includes four types of acquisition units: Soil sensor unit: It adopts buried installation and is distributed in different plots of the collection area to collect core data such as soil moisture, pH value, nitrogen, phosphorus and potassium nutrient content in real time. The collection frequency can be set to 1 time / hour to 1 time / day according to the needs. Topographic surveying unit: integrates GPS positioning module and laser ranging module to collect land elevation, slope and aspect data, with positioning accuracy up to the centimeter level, meeting the needs of topographic analysis; Meteorological monitoring unit: Deployed at the highest point in the data collection area, it collects meteorological data such as temperature, precipitation, wind speed, and sunshine duration to support land ecological assessment; Remote sensing data receiving unit: It periodically receives high-resolution remote sensing image data through a satellite signal receiving antenna to obtain macro data such as land use type and vegetation coverage.

[0022] The data receiving and output unit includes a RESTful API interface and an interface adaptation unit. The interface adaptation unit is used to adapt the output data to the format and match permissions according to the application requirements of different fields. The RESTful API interface is used to provide data query, data download and data subscription services.

[0023] The cloud management platform also includes a data visualization unit, which is used to display raw and fused data in the form of charts, heat maps, and 3D models.

[0024] like Figure 3 As shown, the multi-source acquisition terminal includes a soil sensor unit, a topographic measurement unit, a meteorological monitoring unit, and a remote sensing data receiving unit. The soil sensor unit includes a temperature sensor, a humidity sensor, a pH sensor, and a heavy metal detector, which collects soil physicochemical property data in real time. In use, the soil sensor unit is buried in the soil to be tested, ensuring that the sensor is in full contact with the soil. It collects soil temperature, humidity, pH value, and heavy metal content data in real time. The data processing unit preprocesses the raw data to remove noise and outliers.

[0025] like Figure 4 As shown, the multi-source acquisition terminal includes: An environmental sensing module, which includes a multispectral imaging unit and a miniature weather station; A multimodal acquisition unit is fixed to the adjustment end of a dynamic adjustment mechanism; the multimodal acquisition unit simultaneously acquires soil dielectric constant and topographic features. The controller controls the entire terminal and communicates with the edge computing module. The controller generates environmental baseline data through multispectral scanning; automatically adjusts the sampling depth and frequency according to target parameters; performs synchronous acquisition of multi-source data and edge computing; and preprocesses the raw data, including denoising, normalization and feature extraction.

[0026] The multimodal acquisition unit includes a TOF lidar and a UWB positioning module, a soil dielectric constant sensor and a LIBS spectrometer for soil layer acquisition, and a microfluidic chip microbial sensor for biolayer acquisition.

[0027] like Figure 5 As shown, the data acquisition and management platform also includes an intelligent analysis unit. The intelligent analysis unit has a built-in soil assessment model, which is connected to a result database. The soil assessment model performs threshold judgment on real-time data collected by multi-source acquisition terminals, normalizes the threshold differences of various types, and calculates the processed data through weighted fusion. After obtaining the fused data, the fused data is used as an index to perform threshold retrieval in the result database and generate a soil quality assessment report.

[0028] like Figure 6 As shown, the acquisition management platform also includes a prediction model. The prediction model acquires the data types that change on the time axis within the data storage unit, then acquires all data of the corresponding type on the time axis, and uses this data type as the training set and test set to train the prediction model. The trained prediction model then predicts the data at the next time node to obtain the predicted acquisition data.

[0029] The prediction model processing procedure is as follows: First, the prediction model, through the data management unit of the data acquisition and management platform, identifies the target data type to be predicted (data types that change on the time axis, such as single or combined data types like soil moisture, soil nutrient content, topographic subsidence, temperature, precipitation, and remotely sensed vegetation cover). Then, using the time axis as an index, it extracts the full-volume time-series data corresponding to the target data type from the data storage unit, with the time interval being consecutive months or years. Next, the dataset is partitioned: the preprocessed time-series dataset is randomly divided into training and test sets according to a custom partitioning ratio (e.g., 6:4, 7:3, or 8:2), where the training set is used for model parameter learning and the test set is used for model performance verification. Then, the prediction model defaults to using the LSTM (Long Short-Term Memory) model to adapt to the temporal dependency characteristics of land data. It also supports switching the model type based on the characteristics of the target data type; for example, it can switch to ARIM for highly periodic meteorological data. Model A can switch to a GRU (Gated Recurrent Unit) model for multi-dimensional correlated data. During model initialization, preset basic parameters (such as the number of hidden layers, number of neurons, and learning rate of the LSTM model) are automatically loaded, and the input layer dimension is dynamically adjusted according to the time series length of the target data type. Then, the prediction model is iteratively trained using the time series data of the training set as input and the actual data values ​​of the corresponding time nodes as output labels. After training, the test set is input into the model to obtain the prediction results, and the model prediction accuracy is evaluated by calculating the mean absolute error (MAE) and root mean square error (RMSE). If the prediction accuracy does not reach the preset threshold (default MAE ≤ 5%, user-defined is supported), the model parameters (such as learning rate and number of hidden layer neurons) are automatically adjusted or the model type is changed, and retraining is performed until the model accuracy meets the requirements. Finally, the optimized model parameters are stored in the model parameter library of the data acquisition management platform for subsequent prediction calls.

[0030] After the predicted data is generated, it is synchronized to the data storage unit for archiving, marked as predicted data and associated with the corresponding prediction model and accuracy score; on the other hand, it is pushed to the data visualization unit and displayed in the form of actual data trend chart + predicted data annotation, allowing users to view the relationship between predicted data and historical data; at the same time, it can be synchronously output to the associated application end through the RESTful API interface of the interface open module, providing data support for scenarios such as land resource planning, agricultural production scheduling, and disaster early warning.

[0031] I. The multi-source acquisition terminal of the land data acquisition system of the present invention is used for real-time land data acquisition, such as for crop land data monitoring, and for building a smart agriculture monitoring station, including: Environmental sensing module: It adopts a multispectral imager (400-2500nm wavelength range) combined with a micro weather station, which can simultaneously detect 12 environmental parameters, and optimize the sensor layout through ant colony algorithm; Multimodal acquisition unit: Soil layer: dielectric constant sensor (1-3 GHz band) coupled with LIBS laser-induced breakdown spectroscopy; Terrain layer: TOF lidar (range accuracy ±2mm@100m) is integrated with UWB positioning module.

[0032] Hardware configuration: Main control unit: Renesas RZ / G2L processor, supporting Time-Sensitive Networking (TSN); Soil sensor arrays are shown in Table 1: Table 1: Types of Soil Sensors

[0033] Workflow: Multispectral scanning is initiated at regular intervals to generate NDVI vegetation index maps; sampling depth is dynamically adjusted according to the stage (15cm for seedlings, 50cm for mature plants); data is anomaly detected by the edge computing module (running TensorRTLite); encrypted data packets are uploaded daily at 18:00 via the NB-IoT module.

[0034] 2. The multi-source acquisition terminal of the land data acquisition system of the present invention is used for geological and land data monitoring and to construct a geological disaster monitoring instrument; Micro-vibration sensor (frequency response 0.1-1000Hz), inclinometer (resolution 0.0001°). Operating mode: Under normal conditions, micro-seismic background noise is collected every 2 hours; when the acceleration is detected to be >0.01g, continuous recording mode is triggered; the Beidou-3 RDSS module enables data transmission in areas without public network coverage; the built-in AI model can predict landslide risk 72 hours in advance (accuracy 89.7%).

[0035] 3. The multi-source acquisition terminal of the land data acquisition system of the present invention is used for wetland ecological monitoring buoys and is equipped with biosensors. It adopts microfluidic chips and has a detection limit as low as 0.1 ppb. Data acquisition protocol: The dissolved oxygen sensor adopts fluorescence quenching method and does not require electrolyte; Water quality data is uploaded every 15 minutes via LoRaWAN; When a blue-green algae bloom is detected, UAV collaborative sampling is automatically started.

[0036] Example 1: The land data acquisition system of this invention is used for large-scale precision planting decision-making in farmland, as detailed below: Data Acquisition: The application (agricultural production management system) initiates a "precision corn planting" domain request through the domain request unit of the interface open module, and generates precision request data by combining the data area request unit with the number of each contracted plot. After parsing the request, the acquisition management platform retrieves the original data of the corresponding plot from the data storage unit, and then filters it according to the "corn planting" domain requirements to obtain the target data - including soil data such as soil moisture content, nitrogen, phosphorus and potassium content, and salinity of each plot; meteorological data such as temperature, precipitation, and sunshine duration; topographic data such as slope and flatness; and remote sensing data such as crop growth and vegetation coverage. Data fusion processing: The domain assignment management unit uses "precision corn planting" as an index to retrieve appropriate assignment data from the domain assignment library, obtains and sets soil data accuracy scores, meteorological data accuracy scores, etc., and configures weight coefficients (soil data 0.35, meteorological data 0.25, terrain data 0.2, remote sensing data 0.2); the data acquisition management platform calls the data fusion model, first integrates similar target data through basic fusion, and then outputs cross-dimensional fused data through weighted fusion. Data Application: The application communicates with the data receiving and output unit via a RESTful API interface to obtain target data and fused data, and synchronizes the data to the agricultural production management system. The target data allows farmers to view real-time monitoring data of their plots, while the fused data serves as a comprehensive indicator of soil fertility, meteorological conditions, crop growth matching, optimal fertilization rate assessment, and irrigation demand level, providing a direct basis for planting decisions. The system automatically generates differentiated fertilization plans based on the fused data, such as increasing nitrogen fertilizer application and starting drip irrigation for plots with low soil fertility and insufficient rainfall; assisting managers in coordinating and controlling water and fertilizer resources in the field.

[0037] Example 2: The land data acquisition system of this invention is used for urban land and space planning approval applications, as detailed below: Data Acquisition: The application (urban planning management platform) initiates a request in the "Land and Space Planning" field through the interface open module, specifying the administrative division and compilation unit number of the planning plot; after the data collection and management platform parses the request, it retrieves the original data of the plot from the data storage unit, and filters the target data according to the "Residential Land Planning" field—including geological and soil data such as soil bearing capacity, pollution level, and groundwater depth; topographic data such as elevation, slope, and geological structure; meteorological data such as annual average temperature, rainstorm frequency, and prevailing wind direction; and remote sensing supporting data such as surrounding transportation network, public facilities, and green space distribution. Data fusion processing: The domain assignment management unit uses "residential land planning" as an index to retrieve the appropriate accuracy scores and weighted data from the domain assignment library, setting the weights as follows: geological and soil data 0.4, topographic data 0.25, remote sensing supporting data 0.2, and meteorological data 0.15. Through the data fusion model, basic fusion of similar target data and weighted fusion of cross-class data are completed, outputting fused data. The core data includes comprehensive indicators such as land development suitability score, flood control level assessment, satisfaction with supporting facility coverage, and ecological constraint early warning.

[0038] Data Application: Target data and fused data are synchronized to the urban planning management platform through interface adaptation units. Target data is used by planners to verify the original geology and supporting information of the plot, ensuring the authenticity of the approval basis. Fused data directly supports approval decisions—if the fused data shows that the plot's geological carrying capacity score meets the standards, the flood control level meets the requirements, and the supporting facilities have high coverage, it will be given priority for approval. It also assists planners in optimizing residential layouts, such as avoiding areas with large terrain slopes and designing residential clusters near areas with concentrated public facilities, thereby improving the scientific and rational nature of the planning.

[0039] Example 3: The land data acquisition system of this invention is used for ecological acceptance of land reclamation in mining areas, as detailed below: Data Acquisition: The application (ecological environment monitoring system) initiates a request for "mining area ecological restoration monitoring" through the interface open module, specifying the reclamation zone and reclamation period. After parsing, the data acquisition and management platform retrieves the original monitoring data for the corresponding area and filters the target data according to the "ecological acceptance" requirements—including soil remediation data such as soil heavy metal content, fertility level, and permeability; topographic data such as terrain flatness, collapse risk points, and slope stability; meteorological data such as extreme temperatures and sandstorm frequency; and remote sensing change data such as vegetation coverage, species diversity, and land use type.

[0040] Data Fusion: The domain assignment management unit uses "Mining Area Ecological Acceptance" as an index to retrieve appropriate precision scores and weight configurations from the domain assignment library, setting the weights for soil heavy metal data (0.4), vegetation restoration data (0.25), terrain stability data (0.2), and meteorological data (0.15). The data fusion model outputs fused data, which includes core quantitative indicators such as soil remediation compliance rate, vegetation restoration comprehensive index, terrain stability score, and overall ecological restoration compliance level.

[0041] Data application: The target data serves as the original basis for acceptance, allowing acceptance personnel to verify the completeness and authenticity of the monitoring data; the fused data serves as the core assessment basis. If the fused data shows that the soil remediation compliance rate is ≥90%, the vegetation restoration index is ≥0.8, and the terrain stability score is ≥85 points, then the ecological reclamation is deemed to be qualified, providing visual support for the preparation of the acceptance report and providing data reference for the subsequent optimization of the reclamation plan.

[0042] Example 4: The land data acquisition system of this invention is used for land consolidation effectiveness evaluation, as detailed below: Target data acquisition: The application (management platform) initiates a request in the "Rural Land Consolidation" field through the interface open module, specifying the administrative village area and land use type (arable land, homestead, ecological land); after parsing, the data collection and management platform retrieves the original data of the corresponding area, and filters the target data according to the "consolidation effectiveness evaluation" requirements, including soil data such as arable land fertility and topsoil thickness after consolidation, topographic data such as topography, arable land slope, and homestead distribution, meteorological data such as temperature and precipitation, and remote sensing data such as land use type changes, arable land contiguousness, and ecological corridor connectivity.

[0043] Data fusion processing: The domain assignment management unit uses "rural land consolidation" as the index to retrieve the appropriate weight coefficients (0.3 for cultivated land quality data, 0.25 for land use change data, 0.2 for topographic layout data, 0.15 for meteorological data, and 0.1 for ecological corridor data) from the domain assignment library. The data is then output as fused data through the data fusion model. The core data includes evaluation indicators such as cultivated land quality improvement rate, land use efficiency improvement value, ecological corridor connectivity score, and comprehensive level of consolidation effectiveness.

[0044] Data application: Target data is used to verify the differences between the original data before and after the remediation, such as changes in the thickness of the topsoil of cultivated land and the area of ​​homestead reclamation; fused data is used to quantify the effectiveness. If the fused data shows that the improvement rate of cultivated land quality is ≥20%, the improvement value of land use efficiency is ≥15%, and the ecological corridor connectivity score is ≥90 points, then the remediation effect is judged to be excellent.

[0045] The above embodiments are merely preferred embodiments of the present invention. Therefore, all equivalent changes or modifications made to the structure, features and principles described in the claims of the present invention are included within the scope of the present invention.

Claims

1. A land data acquisition system characterized by: The application relates to a multi-source acquisition terminal, an edge computing module, a data acquisition management platform, a data area marking unit, a data storage unit, a data fusion unit, a data management unit, a domain assignment management unit, a domain assignment library, a data fusion model, a basic fusion, a weighted fusion, an interface management unit, a domain request unit, a data area request unit, a data receiving and output unit, a RESTful API interface, an interface adaptation unit, a data visualization unit, a soil sensor unit, a terrain measurement unit, a weather monitoring unit, a remote sensing data receiving unit, an environmental perception module, a multi-modal acquisition unit, a controller and a dynamic adjustment mechanism. The application relates to a multi-source acquisition terminal, an edge computing module, a data acquisition management platform, a data area marking unit, a data storage unit, a data fusion unit, a data management unit, a domain assignment management unit, a domain assignment library, a data fusion model, a basic fusion, a weighted fusion, an interface management unit, a domain request unit, a data area request unit, a data receiving and output unit, a RESTful API interface, an interface adaptation unit, a data visualization unit, a soil sensor unit, a terrain measurement unit, a weather monitoring unit, a remote sensing data receiving unit, an environmental perception module, a multi-modal acquisition unit, a controller and a dynamic adjustment mechanism. The application relates to a multi-source acquisition terminal, an edge computing module, a data acquisition management platform, a data area marking unit, a data storage unit, a data fusion unit, a data management unit, a domain assignment management unit, a domain assignment library, a data fusion model, a basic fusion, a weighted fusion, an interface management unit, a domain request unit, a data area request unit, a data receiving and output unit, a RESTful API interface, an interface adaptation unit, a data visualization unit, a soil sensor unit, a terrain measurement unit, a weather monitoring unit, a remote sensing data receiving unit, an environmental perception module, a multi-modal acquisition unit, a controller and a dynamic adjustment mechanism. The application relates to a multi-source acquisition terminal, an edge computing module, a data acquisition management platform, a data area marking unit, a data storage unit, a data fusion unit, a data management unit, a domain assignment management unit, a domain assignment library, a data fusion model, a basic fusion, a weighted fusion, an interface management unit, a domain request unit, a data area request unit, a data receiving and output unit, a RESTful API interface, an interface adaptation unit, a data visualization unit, a soil sensor unit, a terrain measurement unit, a weather monitoring unit, a remote sensing data receiving unit, an environmental perception module, a multi-modal acquisition unit, a controller and a dynamic adjustment mechanism. （1） wherein, is the same type of data fusion results, is the same type of data collection terminal number, is the collection data of the first terminal, is the collection accuracy preset score of the first terminal, is the total score of the same type of terminal collection accuracy. The application relates to a multi-source acquisition terminal, an edge computing module, a data acquisition management platform, a data area marking unit, a data storage unit, a data fusion unit, a data management unit, a domain assignment management unit, a domain assignment library, a data fusion model, a basic fusion, a weighted fusion, an interface management unit, a domain request unit, a data area request unit, a data receiving and output unit, a RESTful API interface, an interface adaptation unit, a data visualization unit, a soil sensor unit, a terrain measurement unit, a weather monitoring unit, a remote sensing data receiving unit, an environmental perception module, a multi-modal acquisition unit, a controller and a dynamic adjustment mechanism. ； wherein, is a cross-latitude data fusion result, is the number of land data types participating in the fusion, is a basic fusion result of the first class data, is a basic fusion result of the second class data, and ; The application relates to a multi-source acquisition terminal, an edge computing module, a data acquisition management platform, a data area marking unit, a data storage unit, a data fusion unit, a data management unit, a domain assignment management unit, a domain assignment library, a data fusion model, a basic fusion, a weighted fusion, an interface management unit, a domain request unit, a data area request unit, a data receiving and output unit, a RESTful API interface, an interface adaptation unit, a data visualization unit, a soil sensor unit, a terrain measurement unit, a weather monitoring unit, a remote sensing data receiving unit, an environmental perception module, a multi-modal acquisition unit, a controller and a dynamic adjustment mechanism.

2. The land data acquisition system of claim 1, wherein, The application relates to a multi-source acquisition terminal, an edge computing module, a data acquisition management platform, a data area marking unit, a data storage unit, a data fusion unit, a data management unit, a domain assignment management unit, a domain assignment library, a data fusion model, a basic fusion, a weighted fusion, an interface management unit, a domain request unit, a data area request unit, a data receiving and output unit, a RESTful API interface, an interface adaptation unit, a data visualization unit, a soil sensor unit, a terrain measurement unit, a weather monitoring unit, a remote sensing data receiving unit, an environmental perception module, a multi-modal acquisition unit, a controller and a dynamic adjustment mechanism.

3. The land data acquisition system of claim 1, wherein, The application relates to a multi-source acquisition terminal, an edge computing module, a data acquisition management platform, a data area marking unit, a data storage unit, a data fusion unit, a data management unit, a domain assignment management unit, a domain assignment library, a data fusion model, a basic fusion, a weighted fusion, an interface management unit, a domain request unit, a data area request unit, a data receiving and output unit, a RESTful API interface, an interface adaptation unit, a data visualization unit, a soil sensor unit, a terrain measurement unit, a weather monitoring unit, a remote sensing data receiving unit, an environmental perception module, a multi-modal acquisition unit, a controller and a dynamic adjustment mechanism.

4. The land data acquisition system of claim 1, wherein, The application relates to a multi-source acquisition terminal, an edge computing module, a data acquisition management platform, a data area marking unit, a data storage unit, a data fusion unit, a data management unit, a domain assignment management unit, a domain assignment library, a data fusion model, a basic fusion, a weighted fusion, an interface management unit, a domain request unit, a data area request unit, a data receiving and output unit, a RESTful API interface, an interface adaptation unit, a data visualization unit, a soil sensor unit, a terrain measurement unit, a weather monitoring unit, a remote sensing data receiving unit, an environmental perception module, a multi-modal acquisition unit, a controller and a dynamic adjustment mechanism.

5. The land data acquisition system of claim 1, wherein, ​ 6. The land data acquisition system of claim 1, wherein, ​ ​ ​ ​ 7. The land data acquisition system of claim 6, wherein, ​ 8. The land data acquisition system of claim 1, wherein, The collection management platform further comprises an intelligent analysis unit, the intelligent analysis unit is internally provided with a soil evaluation model, the soil evaluation model is connected with a result library, the soil evaluation model performs threshold value judgment on real-time data collected by the multi-source collection terminal, normalizes threshold value differences of various types, obtains fusion data through weighted fusion calculation of processed data, takes the fusion data as an index, performs threshold value retrieval on the result library, and generates a soil quality evaluation report.

9. The land data acquisition system of claim 8, wherein, The collection management platform further comprises a prediction model, the prediction model acquires data types changing on a time axis in the data storage unit, then acquires all data of the corresponding type on the time axis, and takes the data type as a training set and a test set to train the prediction model, the trained prediction model predicts next time node data, and prediction collection data is obtained.

Images