A land surveying and mapping data analysis method based on big data

Abstract

The present invention provides a land surveying and mapping data analysis method based on big data, belonging to the field of data analysis technology. The data analysis method includes: obtaining high-precision remote sensing images containing a target area and preprocessing them to generate quantitative orthorectified images; automatically classifying and identifying land surveying and mapping data to extract ground features; extracting and fusing the classified and identified land surveying and mapping data; using the generated high-level semantic information to make maps and visually express them, and outputting standardized land surveying and mapping analysis map products; by comparing and analyzing the data accuracy before and after processing, evaluating the errors and uncertainties in the entire analysis process, and ensuring the reliability and accuracy of the results; establishing a spatiotemporal database based on big data to store the data generated in the above steps, and using the information in the spatiotemporal database to establish a spatial model to predict future geographic change trends.

Description

Technical Field

[0001] The present application relates to the field of data analysis technology, and specifically to a land surveying and mapping data analysis method based on big data. Background Art

[0002] With the continuous development of science and technology, the acquisition and application of land surveying and mapping data plays an increasingly important role in urban planning, resource management, environmental protection and other fields. However, traditional land surveying and mapping data analysis methods have some problems and limitations, such as low data processing efficiency, low accuracy of analysis results, limited information extraction and utilization, etc. Therefore, there is an urgent need for a new land surveying and mapping data analysis method that can overcome these problems.

[0003] At present, with the rapid development of big data and artificial intelligence technology, its application in the field of geographic information is becoming more and more extensive. Big data technology can process massive amounts of geographic data and extract valuable information from it; while artificial intelligence technology can realize the automatic classification and identification of land features, thereby improving the efficiency and accuracy of data analysis. Therefore, combining big data and artificial intelligence technology to develop a land surveying and mapping data analysis method based on big data has important theoretical and application significance.

[0004] In past studies, some land surveying and mapping data analysis methods based on big data and artificial intelligence have been proposed, but there are still some shortcomings. For example, some methods lack systematicity and standardization in data processing procedures and algorithms, resulting in poor repeatability and comparability of analysis results; some methods have certain limitations in data fusion and high-level semantic information extraction, and cannot meet the needs of complex geographic information; some methods do not handle key links such as error assessment and spatiotemporal database establishment well, which affects the reliability and credibility of analysis results.

[0005] Therefore, in order to solve the above problems, it is necessary to propose a new land surveying and mapping data analysis method, making full use of big data and artificial intelligence technology to realize the automation, efficiency and precision of data analysis. Summary of the invention

[0006] In order to overcome a series of defects in the prior art, the purpose of this patent is to provide a land surveying and mapping data analysis method based on big data for the above problems, including the following steps:

[0007] Step 1, obtain a high-precision remote sensing image containing the target area and perform preprocessing to generate a quantitative orthorectified image;

[0008] Step 2: Automatically classify and identify the land surveying and mapping data to extract land features including buildings, roads, water bodies, vegetation, and land use types;

[0009] Step 3: extract and fuse the classified and identified land surveying and mapping data to generate high-level semantic information including terrain features, landform types, surface cover, geographic entities, and spatial relationships;

[0010] Step 4: Use the generated high-level semantic information to make maps and visual expressions, and output standardized land surveying and mapping analysis map products;

[0011] Step 5: By comparing the data accuracy before and after analysis, evaluate the errors and uncertainties in the entire analysis process to ensure the reliability and accuracy of the results;

[0012] Step 6: Establish a spatiotemporal database based on big data to store the data generated in steps 1-5, and use the information in the spatiotemporal database to establish a spatial model to predict future geographic change trends.

[0013] Furthermore, step 1 includes the following steps:

[0014] Use high-resolution satellite sensors or drones to obtain high-precision remote sensing images of the target area;

[0015] Perform noise reduction, filtering, radiation correction and atmospheric correction on remote sensing images to improve image quality and usability;

[0016] By using the geometric information of digital elevation models and remote sensing images, the distortion caused by terrain undulation and projection deformation in remote sensing images is eliminated, so that the remote sensing images have the properties of orthographic projection, that is, each pixel corresponds to the actual distance of the ground.

[0017] Further, step 2 includes the following steps:

[0018] The processed image is input into the classification and recognition model, and the output of the model is used to automatically classify and recognize various land features in the image, and generate classification and recognition labels and attribute information, including the category, location, area and shape of the land features;

[0019] The classification and recognition results are integrated and analyzed with other data sources to extract more land feature information and form a complete land feature database.

[0020] Furthermore, multi-band data fusion is used to fuse the classification and recognition results with other data sources. The specific formula is: Among them, I fused is the fused image, T is the transformation function, T -1 is the inverse transformation function, I MS is a multispectral image, I PAN is a full-color image, is the fusion operator.

[0021] Further, step 3 includes the following steps:

[0022] Step 3.1: Use the deep learning-based image segmentation model to divide the classified and identified images into different land feature regions, extract the boundaries and topological relationships of each region, and generate a vectorized land feature layer;

[0023] Step 3.2, construct the conceptual model and attribute model of geographic entities and generate a semantic geographic entity layer;

[0024] Step 3.3, perform spatial operations on the geographic entity layer, extract the spatial relationship between geographic entities, and generate a spatial relationship layer;

[0025] Step 3.4, merge the geographic entity layer and the spatial relationship layer to construct a geographic knowledge graph containing geographic entities and spatial relationships, realize high-level semantic representation and reasoning of geographic information, and generate a high-level semantic information layer.

[0026] Further, step 3.2 includes the following steps:

[0027] Design the conceptual model of geographic entities, that is, define the categories and hierarchical structure of geographic entities, as well as the inheritance and inclusion relationships between categories;

[0028] Design the attribute model of geographic entities, that is, define the attributes and value ranges of geographic entities, as well as the data types of each attribute;

[0029] Match the land features with geographic entities, that is, map them to the conceptual model and attribute model of geographic entities according to their categories and characteristics, give the land features the semantic information of geographic entities, and generate a semantic geographic entity layer.

[0030] Further, step 4 includes the following steps:

[0031] Choose appropriate mapping and visualization methods based on the type, content, and purpose of high-level semantic information;

[0032] Convert high-level semantic information into a data format suitable for mapping and visualization;

[0033] Use the selected mapping and visualization methods to map and visualize the converted data to obtain standardized land surveying and mapping analysis map products;

[0034] Evaluate the results of mapping and visualization to check the effect and quality of mapping and visualization;

[0035] Output the mapping and visualization results into the required format for subsequent display and sharing.

[0036] Further, step 5 includes the following steps:

[0037] Select accuracy assessment indicators based on the type, content and purpose of land surveying and mapping data;

[0038] Use the verification point comparison method to compare the land surveying and mapping data before and after processing to obtain the data change amount and change rate;

[0039] Combined with the selected accuracy evaluation index, the variance analysis method is used to conduct error analysis on the change amount and change rate of the data to obtain the size and distribution of the error;

[0040] Combined with the selected accuracy assessment indicators, the sensitivity analysis method is used to evaluate the uncertainty of the data change amount and change rate to obtain the degree and impact of uncertainty;

[0041] The results of error analysis and uncertainty assessment are evaluated to check the reliability and accuracy of the results.

[0042] Further, step 6 includes the following steps:

[0043] Select an appropriate spatiotemporal database system;

[0044] Design a reasonable spatiotemporal data model, and select and optimize it according to the data's changing characteristics, spatiotemporal relationships, and query requirements;

[0045] Using the interfaces and tools provided by the spatiotemporal database system, import the data generated in steps 1-5 into the spatiotemporal database, and perform necessary data cleaning, conversion, and compression;

[0046] Use the index structure and query language provided by the spatiotemporal database system to efficiently retrieve and count spatiotemporal data and extract the characteristics and laws of spatiotemporal data;

[0047] Using the results of spatiotemporal data analysis, combined with spatial analysis methods and tools, a spatial model is constructed to describe the internal logic and correlation of spatiotemporal data;

[0048] By utilizing spatial models, combining historical data with real-time data, and adopting predictive methods and tools, we estimate and extrapolate future spatiotemporal data to obtain the possibility and scope of future geographic changes.

[0049] Compared with the prior art, the present application has at least the following technical effects or advantages.

[0050] This application can improve the processing efficiency and analysis capabilities of land surveying and mapping data through automated processing, high-level semantic information extraction, mapping and visualization, error assessment, and establishment of spatiotemporal databases, providing important support for the application of geographic information systems and geographic decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 This is a flow chart of a land surveying and mapping data analysis method based on big data disclosed in an embodiment of the present application. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical scheme and advantages of the implementation of the present invention clearer, the technical scheme in the embodiment of the present invention will be described in more detail below in conjunction with the drawings in the embodiment of the present invention. In the drawings, the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The described embodiments are part of the embodiments of the present invention, not all of the embodiments.

[0053] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making any creative work shall fall within the scope of protection of the present invention.

[0054] The embodiments and directional terms described below with reference to the accompanying drawings are exemplary and intended to be used to explain the present invention, but should not be construed as limiting the present invention.

[0055] Reference Figure 1 , a land surveying and mapping data analysis method based on big data, comprising the following steps:

[0056] Step 1, obtain high-precision remote sensing images containing the target area and perform preprocessing to generate quantitative orthorectified images; in this step, some remote sensing satellites or drones and other platforms can be used to obtain high-precision remote sensing images containing the target area, which are generally expressed in pixels or resolution. High-precision remote sensing images can provide more geographic information and details; some preprocessing software and tools can be used to perform some necessary processing on the acquired remote sensing images, such as denoising, enhancement, registration, cropping, etc., to improve the quality and availability of remote sensing images; some orthorectified software and tools can be used to perform some necessary transformations on the preprocessed remote sensing images, such as projection, rotation, scaling, etc., to eliminate geometric distortion and perspective deviation in remote sensing images, and generate quantitative orthorectified images, which are generally expressed in meters or scales. Quantitative orthorectified images can provide more accurate geographic locations and sizes;

[0057] Step 2: Automatically classify and identify the land surveying and mapping data, and extract the land features including buildings, roads, water bodies, vegetation, and land use types. In this step, some classification and identification software and tools can be used to perform some necessary analysis on the orthorectified remote sensing images, such as feature extraction, pattern recognition, machine learning, etc., to achieve automatic classification and identification of land surveying and mapping data, generally represented by categories or labels. Automatic classification and identification can improve the organization and management efficiency of land surveying and mapping data. Some extraction and fusion software and tools can be used to perform some necessary operations on the classified and identified land surveying and mapping data, such as boundary extraction, topology construction, attribute assignment, etc., to extract the land features including buildings, roads, water bodies, vegetation, and land use types, generally represented by vectors or rasters. The land features can provide more specific geographic information and attributes.

[0058] Step 3, extract and fuse the classified and identified land surveying and mapping data to generate high-level semantic information including terrain features, landform types, surface cover, geographic entities and spatial relationships; in this step, some extraction and fusion software and tools can be used to perform some necessary analysis on the land features, such as spatial analysis, statistical analysis, network analysis, etc., to extract high-level semantic information including terrain features, landform types, surface cover, geographic entities and spatial relationships, generally expressed in symbols or text, and high-level semantic information can provide more in-depth geographical knowledge and laws; some extraction and fusion software and tools can be used to perform some necessary operations on the high-level semantic information, such as data integration, data conversion, data cleaning, etc., to generate a unified, standardized, complete and accurate high-level semantic information database, generally expressed in a database or file, and the high-level semantic information database can provide more convenient data storage and query;

[0059] Step 4: Use the generated high-level semantic information to make maps and visual expressions, and output standardized land surveying and mapping analysis map products; in this step, some mapping and visualization software and tools can be used to perform some necessary operations on the high-level semantic information database, such as data selection, data symbolization, data layout, etc., so as to use the generated high-level semantic information for mapping and visual expressions, generally represented by graphics or images, and mapping and visualization expressions can provide more intuitive data display and interaction; some mapping and visualization software and tools can be used to perform some necessary operations on mapping and visualization expressions, such as data annotation, data legend, data scale, etc., so as to output standardized land surveying and mapping analysis map products, generally represented by maps or files, and land surveying and mapping analysis map products can provide more standardized data expression and transmission;

[0060] Step 5, by comparing and analyzing the data accuracy before and after processing, evaluate the errors and uncertainties in the entire analysis process to ensure the reliability and accuracy of the results; in this step, some comparative analysis software and tools can be used to perform some necessary operations on the data accuracy before and after processing, such as data comparison, data difference, data error, etc., to compare and analyze the data accuracy before and after processing, generally expressed as a numerical value or percentage, and the data accuracy before and after processing can provide more objective data evaluation and verification; some comparative analysis software and tools can be used to perform some necessary operations on the errors and uncertainties in the entire analysis process, such as error analysis, uncertainty analysis, sensitivity analysis, etc., to evaluate the errors and uncertainties in the entire analysis process, generally expressed as a numerical value or percentage, and evaluating the errors and uncertainties in the entire analysis process can provide more scientific data quality and credibility; some comparative analysis software and tools can be used to perform some necessary operations on the evaluation results, such as error control, uncertainty elimination, sensitivity adjustment, etc., to ensure the reliability and accuracy of the results, generally expressed as a numerical value or percentage, and ensuring the reliability and accuracy of the results can provide better data effects and experience;

[0061] Step 6, establish a spatiotemporal database based on big data to store the data generated in steps 1-5, and use the information in the spatiotemporal database to establish a spatial model to predict future geographic change trends; in this step, some database software and tools can be used to perform some necessary operations on the data generated in steps 1-5, such as data import, data index, data compression, etc., to establish a spatiotemporal database based on big data to store the data generated in steps 1-5, generally represented by a database or file, and the spatiotemporal database based on big data can provide more efficient data storage and management; some database software and tools can be used to perform some necessary operations on the information in the spatiotemporal database, such as data query, data analysis, data mining, etc., to use the information in the spatiotemporal database to establish a spatial model, generally represented by a model or a graph, and the spatial model can provide more dynamic data display and simulation; some database software and tools can be used to perform some necessary operations on the spatial model, such as data input, data output, data update, etc., to predict future geographic change trends, generally represented by predictions or trends, and predicting future geographic change trends can provide more forward-looking data guidance and decision-making.

[0062] Furthermore, step 1 includes the following steps:

[0063] Use high-resolution satellite sensors or drones to obtain high-precision remote sensing images of the target area; in this step, you can select appropriate remote sensing platforms and sensors based on the location, scope, characteristics and needs of the target area to obtain high-precision remote sensing images of the target area. Generally speaking, high-resolution satellite sensors can provide wider coverage and more stable data quality, but they will also be affected by factors such as weather, clouds, and atmosphere, and the data acquisition cycle and cost are relatively high; while the use of drones can provide more flexible flight control and more detailed data details, but they will also be limited by factors such as flight altitude, wind speed, and power, and the difficulty and complexity of data processing are relatively high. You can consider various factors based on your actual situation and choose the remote sensing platform and sensor that best suits you to obtain high-precision remote sensing images of the target area;

[0064] Perform noise reduction, filtering, radiation correction and atmospheric correction operations on remote sensing images to improve the quality and availability of images; in this step, some image processing software and tools can be used to perform some necessary operations on the acquired remote sensing images to improve the quality and availability of images. Generally speaking, remote sensing images will be affected by some noise and interference during the acquisition and transmission process, resulting in a decrease in image quality. Some noise reduction and filtering methods, such as mean filtering, median filtering, Gaussian filtering, etc., can be used to remove noise and interference in the image and improve the clarity and contrast of the image; and remote sensing images will be affected by some radiation and atmosphere during the acquisition process, resulting in changes in the color and brightness of the image. Some radiation correction and atmospheric correction methods, such as dark pixel method, synchronization method, empirical method, etc., can be used to eliminate the influence of radiation and atmosphere in the image and improve the authenticity and consistency of the image;

[0065] Using the geometric information of digital elevation models and remote sensing images, the distortion caused by terrain undulation and projection deformation in remote sensing images is eliminated, so that the remote sensing images have the properties of orthographic projection, that is, the real distance of each pixel corresponding to the ground; in this step, some orthorectification software and tools can be used to perform some necessary operations on the processed remote sensing images to eliminate the distortion in the remote sensing images, so that the remote sensing images have the properties of orthographic projection. Generally speaking, in the process of acquiring remote sensing images, they will be affected by some terrain undulations and projection deformation, resulting in inconsistent distances between pixels in the images and the ground. Some geometric information of digital elevation models and remote sensing images, such as ground control points, relative orientation, absolute orientation, etc., can be used to calculate the correspondence between pixels in the images and the ground, and then some transformations such as translation, rotation, scaling, etc. are performed on the images to eliminate the distortion in the images, so that the remote sensing images have the properties of orthographic projection, that is, the real distance of each pixel corresponding to the ground.

[0066] Further, step 2 includes the following steps:

[0067] The processed image is input into the classification and recognition model, and the output results of the model are used to automatically classify and identify various land features in the image, and generate classification and recognition labels and attribute information, including the category, location, area and shape of the land features; in this step, some classification and recognition models, such as support vector machines, random forests, neural networks, etc., can be used to perform some necessary operations on the processed image, such as data partitioning, data standardization, data enhancement, etc., so as to input the processed image into the classification and recognition model, which is generally represented by a vector or matrix. Inputting the processed image into the classification and recognition model can improve the training and testing efficiency of the model; some classification and recognition models can be used to input Some necessary operations are performed on the input image, such as feature extraction, feature selection, feature fusion, etc., so as to use the output results of the model to automatically classify and identify various land features in the image, which are generally represented by categories or labels. Automatic classification and identification of various land features in the image can improve the recognition and distinction capabilities of land features; some classification and identification models can be used to perform some necessary operations on the output results, such as data parsing, data decoding, data mapping, etc., to generate classification and identification labels and attribute information, including the category, location, area and shape of land features, which are generally represented by text or numbers. Generating classification and identification labels and attribute information can provide more specific information and features of land features.

[0068] The results of classification and identification are integrated and analyzed with other data sources to extract more information about land features and form a complete land feature database. In this step, some integration and analysis software and tools can be used to perform some necessary operations on the results of classification and identification, such as data matching, data alignment, data merging, etc., so as to integrate and analyze the results of classification and identification with other data sources, which are generally represented by data or files. Other data sources can include some geographic information systems, remote sensing databases, socio-economic data, etc., to provide more geographic information and knowledge. The integration and analysis of the results of classification and identification with other data sources can improve the integrity and diversity of land features. Some integration and analysis software and tools can be used to perform some necessary operations on the results after integration and analysis, such as data extraction, data sorting, data optimization, etc., to extract more information about land features and form a complete land feature database, which is generally represented by a database or file. The land feature database can include some basic information, attribute information, spatial information, related information, etc. of land features, provide more convenient data storage and query, extract more information about land features, and form a complete land feature database to improve the availability and credibility of land features.

[0069] Furthermore, multi-band data fusion is used to fuse the classification and recognition results with other data sources. The specific formula is: Among them, I fused is the fused image, T is the transformation function, T -1 is the inverse transformation function, I MS is a multispectral image, I PAN is a full-color image, is the fusion operator.

[0070] In this embodiment, some mathematical and image processing software and tools can be used to perform some necessary definitions and calculations on various variables and functions in the formula to achieve a specific formula for multi-band data fusion, which is generally expressed in formulas or codes. The specific formula is a method for multi-band data fusion based on transformation functions and fusion operators, which can achieve effective fusion of multispectral images and panchromatic images, but also involves some transformation and fusion parameters and conditions; some mathematical and image processing software and tools can be used to perform some necessary explanations and analyses on various variables and functions in the formula to understand the specific formula for multi-band data fusion, which is generally expressed in text or charts. The meaning and function of the specific formula are as follows:

[0071] I fused It is a fused image, which is an image containing multiple bands. The resolution of each band is the same as that of the panchromatic image, and the spectral information of each band is similar to that of the multispectral image. It is the ultimate goal and result of multi-band data fusion;

[0072] T is a transformation function, which is a function that transforms an image from the spatial domain to the frequency domain. It can be Fourier transform, wavelet transform, discrete cosine transform, etc. It can extract information of different frequency bands of the image and is a basic tool and means for multi-band data fusion;

[0073] T -1 It is an inverse transformation function, which is a function that transforms an image from the frequency domain back to the spatial domain. It is the inverse function of the transformation function, which can be an inverse Fourier transform, inverse wavelet transform, inverse discrete cosine transform, etc. It can reconstruct the spatial domain information of the image and is the final step and process of multi-band data fusion;

[0074] I MS It is a multispectral image, which is an image containing multiple bands. The resolution of each band is different from that of the full-color image, and the spectral information of each band is different from that of the full-color image. It is an input and source data for multi-band data fusion;

[0075] I PANIt is a panchromatic image, which is an image containing only one band. The resolution of this band is different from that of the multispectral image, and the spectral information of this band is different from that of the multispectral image. It is another input and source data for multi-band data fusion;

[0076] It is a fusion operator, a function that fuses two or more images. It can be weighted average, maximum value, minimum value, multiplication, division, etc. It can realize the fusion of information in different frequency bands of the image and is the core and key of multi-band data fusion.

[0077] Further, step 3 includes the following steps:

[0078] Step 3.1, use the deep learning-based image segmentation model to divide the classified and identified images into different land feature areas, extract the boundaries and topological relationships of each area, and generate a vectorized land feature layer; In this step, use the deep learning-based image segmentation model to divide the classified and identified images into different land feature areas, extract the boundaries and topological relationships of each area, and generate a vectorized land feature layer: Some deep learning-based image segmentation models, such as FCN, PSPNet, DeepLab, etc., can be used to perform some necessary operations on the classified and identified images, such as data input, data output, data optimization, etc., to divide the classified and identified images into different land feature areas, generally represented by pixels or masks, to divide the classified and identified images into different land feature areas It can improve the segmentation and positioning capabilities of land features, but it will also introduce some errors and uncertainties; some image processing software and tools can be used to perform some necessary operations on the segmented results, such as boundary extraction, topology construction, attribute assignment, etc., to extract the boundaries and topological relationships of each area, which are generally represented by lines or polygons. Extracting the boundaries and topological relationships of each area can provide more accurate information and features of land features, but it will also lose some image information and continuity; some image processing software and tools can be used to perform some necessary operations on the extracted results, such as data conversion, data compression, data storage, etc., to generate vectorized land feature layers, which are generally represented by vectors or files. Vectorized land feature layers can provide more convenient data storage and management, but they will also take up more storage space and computing resources;

[0079] Step 3.2, construct the concept model and attribute model of geographic entities and generate a semantic geographic entity layer; in this step, some knowledge representation software and tools can be used to perform some necessary operations on the feature layer, such as data extraction, data normalization, data ontology, etc., to construct the concept model and attribute model of geographic entities, generally represented by classes or instances. Constructing the concept model and attribute model of geographic entities can provide more abstract geographic knowledge and rules; some knowledge representation software and tools can be used to perform some necessary operations on the constructed results, such as data annotation, data symbolization, data output, etc., to generate a semantic geographic entity layer, generally represented by symbols or text. The semantic geographic entity layer can provide more intuitive data display and interaction, but it will also lose some data accuracy and integrity;

[0080] Step 3.3, perform spatial operations on the geographic entity layer, extract the spatial relationship between geographic entities, and generate a spatial relationship layer; in this step, some spatial analysis software and tools can be used to perform some necessary operations on the geographic entity layer, such as data selection, data operation, data analysis, etc., to perform spatial operations on the geographic entity layer, generally expressed by functions or formulas, and performing spatial operations on the geographic entity layer can provide more dynamic data display and simulation; some spatial analysis software and tools can be used to perform some necessary operations on the results after the operation, such as data extraction, data sorting, data output, etc., to extract the spatial relationship between geographic entities, generally expressed by relationships or graphs, and extracting the spatial relationship between geographic entities can provide more in-depth geographical knowledge and laws; some spatial analysis software and tools can be used to perform some necessary operations on the extracted results, such as data annotation, data symbolization, data storage, etc., to generate a spatial relationship layer, generally expressed by layers or files, and the spatial relationship layer can provide more convenient data storage and query;

[0081] Step 3.4, fuse the geographic entity layer and the spatial relationship layer, construct a geographic knowledge graph containing geographic entities and spatial relationships, realize high-level semantic representation and reasoning of geographic information, and generate a high-level semantic information layer. In this step, you can use some knowledge graph software and tools to perform some necessary operations on the geographic entity layer and the spatial relationship layer, such as data matching, data alignment, data fusion, etc., to fuse the geographic entity layer and the spatial relationship layer, generally represented by data or files. Fusion of the geographic entity layer and the spatial relationship layer can improve the integrity and diversity of geographic information; users can use some knowledge graph software and tools to perform some necessary operations on the fused results, such as data construction, data reasoning, data optimization, etc., to construct a geographic knowledge graph containing geographic entities and spatial relationships, generally represented by a graph or a diagram. Constructing a geographic knowledge graph containing geographic entities and spatial relationships can be To provide more scientific geographic knowledge and reasoning; users can use some knowledge graph software and tools to perform some necessary operations on the constructed results, such as data annotation, data symbolization, data output, etc., to achieve high-level semantic representation and reasoning of geographic information, generally expressed in semantics or text, to provide more intelligent data services and applications; users can use some knowledge graph software and tools to perform some necessary operations on the results after representation and reasoning, such as data conversion, data compression, data storage, etc., to generate high-level semantic information layers, generally expressed in layers or files, high-level semantic information layers can provide more convenient data storage and management.

[0082] Further, step 3.2 includes the following steps:

[0083] Design the conceptual model of geographic entities, that is, define the categories and hierarchical structure of geographic entities, as well as the inheritance and inclusion relationships between categories; in this step, some knowledge engineering software and tools can be used to perform some necessary operations on geographic entities, such as data analysis, data modeling, data ontology, etc., to design the conceptual model of geographic entities, which is generally represented by class diagrams or ontology diagrams. Designing the conceptual model of geographic entities can provide more abstract geographic knowledge and rules; users can use some knowledge engineering software and tools to perform some necessary operations on the designed results, such as data definition, data constraints, data validation, etc., to define the categories and hierarchical structure of geographic entities, as well as the inheritance and inclusion relationships between categories, which are generally represented by text or code. Defining the categories and hierarchical structure of geographic entities, as well as the inheritance and inclusion relationships between categories can provide clearer classification and organization of geographic entities;

[0084] Design the attribute model of geographic entity, that is, define the attributes and value range of geographic entity, as well as the data type of each attribute; in this step, some database software and tools can be used to perform some necessary operations on geographic entity, such as data analysis, data modeling, data normalization, etc., to design the attribute model of geographic entity, generally expressed in tables or schemas, and the design of the attribute model of geographic entity can provide more specific information and features of geographic entity; users can use some database software and tools to perform some necessary operations on the designed results, such as data definition, data constraint, data validation, etc., to define the attributes and value range of geographic entity, as well as the data type of each attribute, generally expressed in text or code, and defining the attributes and value range of geographic entity, as well as the data type of each attribute can provide more accurate data and format of geographic entity;

[0085] The geographical features are matched with the geographic entities, that is, according to the categories and characteristics of the geographical features, they are mapped to the conceptual model and attribute model of the geographic entity, the geographical features are given the semantic information of the geographic entity, and a semantic geographic entity layer is generated; in this step, some data conversion software and tools can be used to perform some necessary operations on the geographical features and the geographic entities, such as data matching, data mapping, data conversion, etc., so as to match the geographical features with the geographic entities, generally represented by mapping or conversion, and matching the geographical features with the geographic entities can improve the semantics and usability of the geographical features; some data conversion software and tools can be used to perform some necessary operations on the corresponding results, such as data annotation, data symbolization, data output, etc., so as to map the geographical features to the conceptual model and attribute model of the geographic entity according to the categories and characteristics of the geographical features, give the geographical features the semantic information of the geographic entity, and generate a semantic geographic entity layer, generally represented by a layer or a file, and generating a semantic geographic entity layer can provide more intuitive data display and interaction.

[0086] Further, step 4 includes the following steps:

[0087] According to the type, content, and purpose of high-level semantic information, select appropriate mapping and visualization methods; in this step, you can use some mapping and visualization software and tools to perform some necessary operations on high-level semantic information, such as data analysis, data classification, data screening, etc., to select appropriate mapping and visualization methods according to the type, content, and purpose of high-level semantic information. Generally, it is expressed as a method or process. Selecting appropriate mapping and visualization methods can improve the efficiency and effect of mapping and visualization; you can use some mapping and visualization software and tools to perform some necessary operations on the selected results, such as data configuration, data adjustment, data preview, etc., to determine the final mapping and visualization method, which is generally expressed as a chart or code. Determining the final mapping and visualization method can improve the accuracy and consistency of mapping and visualization. There are many methods of mapping and visualization, and the more common ones are: bar charts, line charts, pie charts, scatter plots, maps, dashboards, etc.;

[0088] Convert high-level semantic information into a data format suitable for mapping and visualization; in this step, some data conversion software and tools can be used to perform some necessary operations on the high-level semantic information, such as data parsing, data conversion, data standardization, etc., to convert the high-level semantic information into a data format suitable for mapping and visualization, generally represented by a format or file. Converting high-level semantic information into a data format suitable for mapping and visualization can improve the compatibility and usability of mapping and visualization; some data conversion software and tools can be used to perform some necessary operations on the converted results, such as data validation, data optimization, data storage, etc., to generate the final mapping and visualization data, generally represented by data or files, and generating the final mapping and visualization data can provide more convenient data input and output. There are many data formats for mapping and visualization, and the more common ones are: CSV, JSON, XML, Excel, Shapefile, etc.

[0089] Use the selected mapping and visualization method to map and visualize the converted data to obtain standardized land surveying and mapping analysis map products; in this step, some mapping and visualization software and tools can be used to perform some necessary operations on the converted data, such as data input, data processing, data output, etc., so as to use the selected mapping and visualization method to map and visualize the converted data, generally represented by maps or files, and use the selected mapping and visualization method to map and visualize the converted data to provide clearer data display and communication; some mapping and visualization software and tools can be used to perform some necessary operations on the results after mapping and visualization, such as data beautification, data annotation, data export, etc., to obtain standardized land surveying and mapping analysis map products, generally represented by maps or files, and standardized land surveying and mapping analysis map products can provide more professional data analysis and reports. There are many kinds of mapping and visualization software and tools, the more common ones are: ArcGIS, QGIS, Excel, PowerBI, Tableau, etc.

[0090] Evaluate the results of mapping and visualization, and check the effect and quality of mapping and visualization; in this step, some evaluation and inspection software and tools can be used to perform some necessary operations on the results of mapping and visualization, such as data comparison, data evaluation, data feedback, etc., to evaluate the results of mapping and visualization, check the effect and quality of mapping and visualization, generally expressed in scores or reports, evaluate the results of mapping and visualization, and check the effect and quality of mapping and visualization to improve the level and reliability of mapping and visualization; some evaluation and inspection software and tools can be used to perform some necessary operations on the results after evaluation and inspection, such as data modification, data improvement, data update, etc., to optimize the results of mapping and visualization, improve the effect and quality of mapping and visualization, generally expressed in drawings or files, optimize the results of mapping and visualization, and improve the effect and quality of mapping and visualization to provide better data products and services. There are many kinds of evaluation and inspection software and tools, the more common ones are: ArcGIS Data Reviewer, QGIS Data Quality Tools, Excel Data Analysis Tools, etc.

[0091] Output the results of mapping and visualization to the required format for subsequent display and sharing; In this step, some output and sharing software and tools can be used to perform some necessary operations on the results of mapping and visualization, such as data selection, data formatting, data output, etc., to output the results of mapping and visualization to the required format, generally expressed in a format or file. Outputting the results of mapping and visualization to the required format can improve the compatibility and portability of mapping and visualization; some output and sharing software and tools can be used to perform some necessary operations on the output results, such as data upload, data publishing, data sharing, etc., to facilitate subsequent display and sharing, generally expressed in a web page or link, which is convenient for subsequent display and sharing and can improve the influence and value of mapping and visualization. There are many kinds of output and sharing software and tools, the more common ones are: ArcGIS Online, QGIS Cloud, Power BI Service, Tableau Online, etc.

[0092] Further, step 5 includes the following steps:

[0093] According to the type, content and purpose of land surveying and mapping data, select accuracy assessment indicators; in this step, some surveying and mapping theories and methods can be used to perform some necessary operations on land surveying and mapping data, such as data analysis, data classification, data screening, etc., in order to select accuracy assessment indicators according to the type, content and purpose of land surveying and mapping data, generally expressed as indicators or formulas, and selecting accuracy assessment indicators can improve the efficiency and effect of accuracy evaluation of land surveying and mapping data; some surveying and mapping theories and methods can be used to perform some necessary operations on the selected results, such as data definition, data constraint, data verification, etc., in order to determine the final accuracy assessment indicators, generally expressed as text or code, and determining the final accuracy assessment indicators can improve the accuracy and consistency of accuracy evaluation of land surveying and mapping data. There are many kinds of accuracy assessment indicators, the more common ones are: linear accuracy, area accuracy, elevation accuracy, position accuracy, angle accuracy, length accuracy, etc.;

[0094] Use the verification point comparison method to compare the land surveying and mapping data before and after processing to obtain the amount and rate of change of the data; in this step, some data processing software and tools can be used to perform some necessary operations on the land surveying and mapping data, such as data input, data output, data comparison, etc., to compare the land surveying and mapping data before and after processing using the verification point comparison method, generally expressed in tables or graphs, and using the verification point comparison method to compare the land surveying and mapping data before and after processing can provide more intuitive data display and analysis; users can use some data processing software and tools to perform some necessary operations on the results after comparison, such as data calculation, data statistics, data output, etc., to obtain the amount and rate of change of the data, generally expressed in numerical values ​​or percentages, and obtaining the amount and rate of change of the data can provide more objective data evaluation and judgment. The verification point comparison method is a commonly used data comparison method, which is suitable for data from different times or different sources. This method selects some representative verification points, compares their positions or attributes in different data, and calculates their offset distances or difference values, thereby evaluating data changes;

[0095] Combined with the selected precision assessment indicators, the variance analysis method is used to perform error analysis on the amount and rate of change of the data to obtain the size and distribution of the error; in this step, some statistical analysis software and tools can be used to perform some necessary operations on the amount and rate of change of the data, such as data input, data output, data analysis, etc., combined with the selected precision assessment indicators, the variance analysis method is used to perform error analysis on the amount and rate of change of the data, generally expressed as variance or standard deviation, and the use of variance analysis to perform error analysis on the amount and rate of change of the data can provide more scientific data evaluation and reasoning; users can use some statistical analysis software and tools to perform some necessary operations on the results of the analysis, such as data description, data drawing, data output, etc., to obtain the size and distribution of the error, generally expressed as numerical values ​​or graphics, and the size and distribution of the error can provide clearer data display and communication. Variance analysis is a commonly used error analysis method, which is suitable for data with multiple factor comparisons. This method analyzes the variance of the observed data and calculates the contribution of each factor, thereby evaluating the error of the data.

[0096] Combined with the selected precision assessment indicators, the sensitivity analysis method is used to evaluate the uncertainty of the data change amount and change rate, and the degree and impact of uncertainty are obtained; in this step, some uncertainty analysis software and tools can be used to perform some necessary operations on the data change amount and change rate, such as data input, data output, data analysis, etc., to combine the selected precision assessment indicators, use the sensitivity analysis method to evaluate the uncertainty of the data change amount and change rate, generally expressed as sensitivity coefficient or sensitivity index, and use the sensitivity analysis method to evaluate the uncertainty of the data change amount and change rate can provide more in-depth data evaluation and reasoning; some uncertainty analysis software and tools can be used to perform some necessary operations on the results after evaluation, such as data description, data drawing, data output, etc., to obtain the degree and impact of uncertainty, generally expressed as numerical values ​​or graphs, and the degree and impact of uncertainty can provide more comprehensive data display and communication. Sensitivity analysis is a commonly used uncertainty assessment method, which is suitable for multi-parameter data. This method calculates the degree of influence of different parameters on the data, understands the sensitivity of the data, and thus evaluates the uncertainty of the data.

[0097] Evaluate the results of error analysis and uncertainty assessment, and check the reliability and accuracy of the results; in this step, you can use some result evaluation software and tools to perform some necessary operations on the results of error analysis and uncertainty assessment, such as data comparison, data evaluation, data feedback, etc., to evaluate the results of error analysis and uncertainty assessment, and check the reliability and accuracy of the results. Generally, it is expressed in scores or reports. Evaluating the results of error analysis and uncertainty assessment and checking the reliability and accuracy of the results can improve the level and reliability of the precision evaluation of land surveying and mapping data; users can use some result evaluation software and tools to perform some necessary operations on the results after evaluation and inspection, such as data modification, data improvement, data update, etc., to optimize the results of precision evaluation of land surveying and mapping data, and improve the effect and quality of precision evaluation of land surveying and mapping data. Generally, it is expressed in drawings or files, and optimizing the results of precision evaluation of land surveying and mapping data, and improving the effect and quality of precision evaluation of land surveying and mapping data can provide better land surveying and mapping data products and services.

[0098] Further, step 6 includes the following steps:

[0099] Choose a suitable spatiotemporal database system; a spatiotemporal database system is a database system that can store, manage and analyze spatiotemporal data. Spatiotemporal data is data that has both time and space dimensions. The selection of a spatiotemporal database system should consider the following factors: 1) Data type, scale and complexity: Different spatiotemporal database systems may support different data types, such as vector data, raster data, trajectory data, etc. The scale and complexity of the data will also affect the performance and scalability of the database system; 2) Data update frequency and real-time: If the data needs to be updated frequently or processed in real time, the database system should have the ability to efficiently insert, modify and delete data, as well as real-time data synchronization and stream processing capabilities; 3) Data query and analysis requirements: Different spatiotemporal database systems may provide different query languages ​​and analysis functions, such as SQL, OGC standards, spatial topological relationships, spatial aggregation, spatial similarity, spatial pattern mining, etc. Query and analysis requirements should be determined based on the characteristics of the data and the purpose of the application; 4) Data security and reliability: Data security and reliability involve data access control, backup and recovery, fault tolerance mechanisms, data encryption, etc. The database system should be able to ensure data integrity, availability and confidentiality; 5) Data visualization and presentation: Data visualization and presentation is an important part of data analysis, which can help users understand and use data more intuitively. The database system should be able to provide or support various visualization and presentation tools and methods;

[0100] Design a reasonable spatiotemporal data model, and select and optimize it according to the changing characteristics of the data, spatiotemporal relationships and query requirements; the spatiotemporal data model is an abstract representation that describes the structure, semantics and constraints of spatiotemporal data. The design of the spatiotemporal data model should follow the following principles: 1) Effectiveness: The spatiotemporal data model should be able to effectively express the characteristics and laws of spatiotemporal data, such as the shape, position, movement, change, interaction, etc. of spatiotemporal objects. 2) Flexibility: The spatiotemporal data model should be able to adapt to different data types, data sources, data quality, data granularity, etc., as well as different application scenarios and requirements. 3) Consistency: The spatiotemporal data model should be able to ensure the consistency of the data, that is, the representation and interpretation of the data should be consistent within different time and space ranges. 4) Scalability: The spatiotemporal data model should be able to support incremental updates and expansions of data, as well as the addition of new data attributes and relationships. 5) Operability: The spatiotemporal data model should be able to provide or support various operations on the data, such as query, analysis, modification, deletion, etc., as well as the corresponding operation languages ​​and algorithms;

[0101] Using the interfaces and tools provided by the spatiotemporal database system, import the data generated in steps 1-5 into the spatiotemporal database, and perform necessary data cleaning, conversion and compression; the spatiotemporal database system usually provides some interfaces and tools so that users can import data from different data sources and formats into the database, such as files, networks, sensors, etc. In the process of importing data, the following points should be noted: 1) Data cleaning: Data cleaning refers to removing or correcting errors, missing, duplication, inconsistency and other problems in the data to improve the quality and availability of the data. Data cleaning can be performed before or after importing, and can be done manually or automatically, such as data verification, data filling, data matching, data fusion, etc. 2) Data conversion: Data conversion refers to converting data from one format or structure to another format or structure to meet the requirements of the database system or the needs of users. Data conversion can involve changes in data encoding, units, coordinate systems, projections, accuracy, resolution, timestamps, etc. 3) Data compression: Data compression refers to reducing the storage space or transmission bandwidth of data through some algorithms or technologies to improve data efficiency and performance. Data compression can be divided into lossy compression and lossless compression. The former will sacrifice a certain amount of data quality, while the latter will not. Data compression can be performed before or after importing, and spatial domain or frequency domain methods can be used, such as spatial clustering, spatial generalization, spatial indexing, spatial encoding, spatial transformation, etc.

[0102] Using the index structure and query language provided by the spatiotemporal database system, the spatiotemporal data can be efficiently retrieved and counted, and the characteristics and laws of the spatiotemporal data can be extracted; the spatiotemporal database system usually provides some index structures and query languages ​​so that users can efficiently retrieve and count the spatiotemporal data and extract the characteristics and laws of the spatiotemporal data. The index structure is a data structure that can speed up data access and search, such as B-tree, R-tree, quadtree, grid index, etc. The query language is a language that can express the user's query and analysis requirements for data, such as SQL, OGC standards, spatial topological relationships, spatial aggregation, spatial similarity, spatial pattern mining, etc. When using the index structure and query language provided by the spatiotemporal database system, the following points should be noted: 1) Selection of index structure: The selection of index structure should be determined based on the type, scale and complexity of the data, as well as the type, frequency and scope of the query. Different index structures may have different advantages and disadvantages, such as storage space, construction time, update time, query time, etc. Generally speaking, the index structure should be able to support queries in both spatial and temporal dimensions, as well as multi-level and multi-granular queries. 2) Use of query language: The use of query language should be determined according to the characteristics of the data and the purpose of the application. Different query languages ​​may have different syntax, semantics, and functions, such as spatial selection, spatial connection, spatial buffering, spatial overlay, spatial interpolation, spatial prediction, etc. In general, the query language should be able to support complex spatiotemporal query and analysis, as well as multi-source and multi-modal data fusion.

[0103] Using the results of spatiotemporal data analysis, combined with spatial analysis methods and tools, a spatial model is constructed to describe the internal logic and correlation of spatiotemporal data; the results of spatiotemporal data analysis can help users understand the internal logic and correlation of spatiotemporal data more deeply, such as the shape, position, movement, change, interaction, etc. of spatiotemporal objects. A spatial model is an abstract and simplified representation of spatial phenomena and processes, which can be expressed in mathematical formulas, graphic symbols, computer codes, etc. The construction of a spatial model should follow the following principles: 1) Adaptability: The spatial model should be able to adapt to different spatiotemporal data types, data quality, data granularity, etc., as well as different application scenarios and needs. 2) Simplicity: The spatial model should be able to remove irrelevant or secondary information in spatiotemporal data, highlight the main or key information in spatiotemporal data, and avoid overly complex or redundant models. 3) Verifiability: The spatial model should be able to verify its correctness and effectiveness through experiments or observations, or prove its rationality and reliability through theory or logic. 4) Updatability: The spatial model should be able to be modified or improved according to changes in spatiotemporal data or new discoveries, or improved or optimized according to new theories or methods.

[0104] By using spatial models, combining historical data and real-time data, and using prediction methods and tools, we can estimate and infer future spatiotemporal data and obtain the possibility and scope of future geographical changes; spatial models can help users estimate and infer future spatiotemporal data, such as the shape, position, movement, change, interaction, etc. of spatiotemporal objects. Prediction is a process of speculating and anticipating unknown data and information based on known data and information according to certain rules and methods, which can be expressed in terms of probability, interval, trend, etc. Prediction methods and tools include: 1) Spatial interpolation: Spatial interpolation is a method of using known spatial data to infer unknown spatial data according to certain functions or algorithms, such as inverse distance weighted method, kriging method, spline method, etc. 2) Spatial regression: Spatial regression is a method of using known spatial data to analyze the relationship between spatial data according to certain models or equations and predict unknown spatial data, such as linear regression, nonlinear regression, geographically weighted regression, etc. 3) Spatial optimization: Spatial optimization is a method of using known spatial data to find the optimal or best spatial data according to certain goals or constraints, such as linear programming, integer programming, genetic algorithm, etc. 4) Spatial simulation: Spatial simulation is a method that uses known spatial data to simulate the changing process of spatial data according to certain rules or mechanisms, and predicts unknown spatial data, such as cellular automata, artificial neural networks, Markov chains, etc.

Claims

1. A land surveying and mapping data analysis method based on big data, characterized in that: The following steps are involved: Step 1, obtain a high-precision remote sensing image containing the target area and perform preprocessing to generate a quantitative orthorectified image; Step 2: Automatically classify and identify the land surveying and mapping data to extract land features including buildings, roads, water bodies, vegetation, and land use types; Step 3: extract and fuse the classified and identified land surveying and mapping data to generate high-level semantic information including terrain features, landform types, surface cover, geographic entities, and spatial relationships; Step 4: Use the generated high-level semantic information to make maps and visual expressions, and output standardized land surveying and mapping analysis map products; Step 5: By comparing the data accuracy before and after analysis, evaluate the errors and uncertainties in the entire analysis process to ensure the reliability and accuracy of the results; Step 6, establishing a spatiotemporal database based on big data to store the data generated in steps 1-5, and using the information in the spatiotemporal database to establish a spatial model to predict future geographic change trends; Step 2 includes the following steps: inputting the processed image into the classification and recognition model, using the output results of the model to automatically classify and recognize various land features in the image, and generating classification and recognition labels and attribute information, including the category, location, area and shape of the land feature; fusing and analyzing the classification and recognition results with other data sources, extracting more land feature information, and forming a complete land feature database; Multi-band data fusion is used to fuse the classification and recognition results with other data sources. The specific formula is: Among them, I fused is the fused image, T is the transformation function, T -1 is the inverse transformation function, I MS is a multispectral image, I PAN is a full-color image, is the fusion operator; Step 3 includes the following steps: Step 3.1, using a deep learning-based image segmentation model to divide the classified and identified images into different land feature areas, extracting the boundaries and topological relationships of each area, and generating a vectorized land feature layer; Step 3.2, constructing a conceptual model and an attribute model of geographic entities and generating a semantic geographic entity layer; Step 3.3, performing spatial operations on the geographic entity layer, extracting the spatial relationships between geographic entities, and generating a spatial relationship layer; Step 3.4, fusing the geographic entity layer and the spatial relationship layer, constructing a geographic knowledge graph containing geographic entities and spatial relationships, realizing high-level semantic representation and reasoning of geographic information, and generating a high-level semantic information layer; Step 3.2 includes the following steps: designing the conceptual model of geographic entities, that is, defining the categories and hierarchical structure of geographic entities, as well as the inheritance and inclusion relationships between categories; designing the attribute model of geographic entities, that is, defining the attributes and value ranges of geographic entities, as well as the data type of each attribute; matching geographical features with geographic entities, that is, mapping the geographical features to the conceptual model and attribute model of geographic entities according to their categories and characteristics, giving the geographical features the semantic information of geographic entities, and generating a semantic geographic entity layer.

2. The land surveying and mapping data analysis method based on big data according to claim 1 is characterized in that: Step 1 includes the following steps: Use high-resolution satellite sensors or drones to obtain high-precision remote sensing images of the target area; Perform noise reduction, filtering, radiation correction and atmospheric correction on remote sensing images to improve image quality and usability; By using the geometric information of digital elevation models and remote sensing images, the distortion caused by terrain undulation and projection deformation in remote sensing images is eliminated, so that the remote sensing images have the properties of orthographic projection, that is, each pixel corresponds to the actual distance of the ground.

3. The land surveying and mapping data analysis method based on big data according to claim 1 is characterized in that: Step 4 includes the following steps: Select mapping and visualization methods based on the type, content, and purpose of high-level semantic information; Convert high-level semantic information into a data format suitable for mapping and visualization; Use the selected mapping and visualization methods to map and visualize the converted data to obtain standardized land surveying and mapping analysis map products; Evaluate the results of mapping and visualization to check the effect and quality of mapping and visualization; Output the mapping and visualization results into the required format for subsequent display and sharing.

4. The land surveying and mapping data analysis method based on big data according to claim 1 is characterized in that: Step 5 includes the following steps: Select accuracy assessment indicators based on the type, content and purpose of land surveying and mapping data; Use the verification point comparison method to compare the land surveying and mapping data before and after processing to obtain the data change amount and change rate; Combined with the selected accuracy evaluation index, the variance analysis method is used to conduct error analysis on the change amount and change rate of the data to obtain the size and distribution of the error; Combined with the selected accuracy assessment indicators, the sensitivity analysis method is used to evaluate the uncertainty of the data change amount and change rate to obtain the degree and impact of uncertainty; The results of error analysis and uncertainty assessment are evaluated to check the reliability and accuracy of the results.

5. The land surveying and mapping data analysis method based on big data according to claim 1 is characterized in that: Step 6 includes the following steps: Select a spatiotemporal database system; Design a reasonable spatiotemporal data model, and select and optimize it according to the data's changing characteristics, spatiotemporal relationships, and query requirements; Using the interfaces and tools provided by the spatiotemporal database system, import the data generated in steps 1-5 into the spatiotemporal database, and perform data cleaning, conversion, and compression; Use the index structure and query language provided by the spatiotemporal database system to efficiently retrieve and count spatiotemporal data and extract the characteristics and laws of spatiotemporal data; Using the results of spatiotemporal data analysis, combined with spatial analysis methods and tools, a spatial model is constructed to describe the internal logic and correlation of spatiotemporal data; By utilizing spatial models, combining historical data with real-time data, and adopting predictive methods and tools, we estimate and extrapolate future spatiotemporal data to obtain the possibility and scope of future geographic changes.

Images