Method for constructing landform information model based on multi-source geographic data

Abstract

The invention discloses a landform information model construction method based on multi-source geographic data, and aims to efficiently and conveniently integrate and process geographic data with different sources and different formats by designing a new data integration framework. The framework allows an algorithm to automatically identify data types and convert data formats in real time, so that manual participation is reduced, and the efficiency and accuracy of data integration are improved. According to the landform information model construction method based on the multi-source geographic data, a new data integration framework is designed, so that geographic data of different sources and different formats can be efficiently and conveniently integrated and processed. The framework allows an algorithm to automatically identify data types and convert data formats in real time, so that manual participation is reduced, and the efficiency and accuracy of data integration are improved.

Description

Technical Field

[0001] The present invention relates to the field of geographic information technology, and in particular to a method for constructing a terrain and geomorphology information model based on multi-source geographic data. Background Art

[0002] Topography is an important natural feature of the earth's surface and is of great significance to the fields of geographic information systems, geological exploration, urban planning, etc. Traditional methods for obtaining topography information mainly rely on field measurements and analysis of a single data source, which has problems such as high cost, low efficiency, and limited data coverage. With the development of remote sensing technology, geographic information system technology, and computer technology, the application of multi-source geographic data has provided new ideas and methods for the construction of topography information models.

[0003] However, the current multi-source geographic data fusion methods have problems such as poor data compatibility, information loss, and low model accuracy, which limit the application effect of terrain and geomorphic information models. Therefore, a new method and device for constructing terrain and geomorphic information models based on multi-source geographic data is needed to improve the accuracy and reliability of the model.

[0004] The existing technology has the following shortcomings in the process of constructing terrain and geomorphic information models based on multi-source geographic data:

[0005] (1) With the diversification of data sources, various types of geographic data have significant differences in format, accuracy, timeliness, etc., resulting in greater integration difficulties when building models. Existing technologies often rely on traditional data preprocessing methods, which have high requirements for the uniformity of data formats. If the data sources are diverse, the time and labor costs of the integration process are high. At the same time, data inconsistency may reduce the reliability and accuracy of the model.

[0006] (2) Current topographic model construction relies on offline processing and lacks the ability to respond quickly to dynamic changes. Real-time terrain changes, such as changes in terrain after natural disasters, often make it difficult for existing technologies to quickly improve models. In this case, the timeliness of data is severely restricted, which is not conducive to making scientific decisions in a timely manner.

[0007] (3) In the process of integrating multiple data sources, it is easy to cause information loss or error propagation, affecting the accuracy of the final model. Existing technologies often use averaging methods to process data, which causes some detailed information to be ignored, reducing the model's ability to represent complex terrain. Especially when dealing with complex terrain, existing models often cannot provide enough detailed information.

[0008] To this end, the present invention proposes a method for constructing a terrain and geomorphic information model by fusing multi-source geographic data, aiming to solve the problems existing in the existing multi-source geographic data fusion methods, and provide a method for constructing a terrain and geomorphic information model based on multi-source geographic data, so as to realize accurate modeling of terrain and geomorphic information and improve the accuracy and reliability of the model. Summary of the invention

[0009] 1. Technical issues to be resolved

[0010] In view of the shortcomings of the prior art, the present invention provides a method for constructing a terrain and geomorphic information model based on multi-source geographic data, which solves the problems of low efficiency and insufficient precision in the traditional terrain and geomorphic information modeling process. The invention integrates and analyzes geographic data from different sources to achieve efficient extraction and modeling of terrain and geomorphic information, aiming to provide an accurate and comprehensive terrain and geomorphic information model.

[0011] (II) Technical solution

[0012] To achieve the above objectives, the present invention is implemented by the following technical solutions: A method for constructing a topographic and geomorphic information model based on multi-source geographic data, specifically comprising the following steps:

[0013] S1. Data preprocessing: Collect multi-source geographic data including remote sensing images, digital elevation models (DEMs), terrain data, and geological data, perform quality checks on the collected data, remove noise and outliers, convert the formats of data from different data sources, and unify the coordinates for subsequent fusion processing;

[0014] S2. Feature extraction: Extract features from the preprocessed geographic data, including topographic features, geomorphic features, and geological features, and use feature extraction algorithms to improve the accuracy and efficiency of feature extraction;

[0015] S3, model construction: select appropriate topographic information model, use the extracted feature data, adopt machine learning algorithms and statistical methods to build the model, and in the process of model construction, consider the weight and correlation of multi-source geographic data to improve the accuracy and reliability of the model;

[0016] S4. Model optimization: Use cross-validation and regularization methods to optimize the constructed model to prevent overfitting and underfitting, adjust and optimize the model parameters, and improve the performance and generalization ability of the model.

[0017] S5. Result evaluation: Use the mean absolute error (MAE) and correlation coefficient (R) evaluation indicators to evaluate the constructed topographic information model, compare and analyze the evaluation results with the actual topographic data, verify the accuracy and reliability of the model, and improve and optimize the model based on the evaluation results.

[0018] Preferably, in step S2, the terrain features are slope, aspect and roughness, the landform features are valleys, ridges and peaks, and the geological features are strata and lithology.

[0019] Preferably, the data preprocessing in step S1 specifically includes the following steps:

[0020] A1. Data collection;

[0021] A2. Data quality check: Check the quality of the collected geographic data, including the integrity, accuracy, and consistency of the data, remove noise and outliers, use median filtering and mean filtering methods to smooth the data, remove noise interference, fill in missing values ​​in the data, and use interpolation and regression analysis methods to estimate and fill in missing values;

[0022] A3. Data format conversion and coordinate unification: Convert the data from different data sources into a common data format, unify the coordinates of the data from different data sources, and use the geographic coordinate system or projection coordinate system to convert the data to ensure the spatial consistency of the data.

[0023] Preferably, the data collection in step A1 specifically includes the following steps:

[0024] T1. Remote sensing image collection: high-resolution image data acquired through satellite or aerial remote sensing platforms, including images in visible light, infrared, microwave and other bands;

[0025] T2. Digital elevation model (DEM) collection: Digital elevation model data obtained through terrain surveying, radar interferometry and other technologies, reflecting the elevation information of the terrain;

[0026] T3. Topographic data collection: including traditional topographic data such as topographic maps and contour maps, as well as high-precision topographic data obtained through lidar measurement technology;

[0027] T4. Geological data collection: including geological maps, stratigraphic lithology data, and geological structure data, reflecting the impact of geological conditions on topography.

[0028] Preferably, the feature extraction in step S2 specifically includes the following steps:

[0029] B1. Terrain feature extraction: The slope calculation method based on DEM is used to calculate the slope information of the terrain. The slope aspect calculation method based on DEM is used to calculate the slope aspect information of the terrain. The roughness calculation method based on terrain undulation is used to calculate the roughness information of the terrain.

[0030] B2. Geomorphic feature extraction: A method based on terrain morphology is used to extract valley and ridge information from the terrain, and a method based on terrain elevation is used to extract peak information from the terrain.

[0031] Preferably, the model building in step S3 specifically includes the following steps:

[0032] C1. Model selection: Select an appropriate topographic and geomorphic information model based on the topographic and geomorphic characteristics of the study area and application requirements;

[0033] C2. Model construction method: neural network and support vector machine machine learning algorithms are used to build models using the extracted feature data; regression analysis and cluster analysis statistical methods are used to build models using the extracted feature data;

[0034] C3. Multi-source data fusion: In the process of model construction, the weights and correlations of multi-source geographic data are considered, and the weighted average method and principal component analysis method are used to fuse multi-source data.

[0035] Preferably, the model optimization in step S4 specifically includes the following steps:

[0036] D1. Cross-validation: Divide the data set into a training set and a test set, and use the cross-validation method to train and validate the model to prevent overfitting, adjust the model parameters, and improve the performance and generalization ability of the model;

[0037] D2. Structural optimization: A modular design is adopted to allow users to select different model components according to specific needs. The optimization of the model structure will be completed by experimentally comparing the performance of different model components and selecting the optimal combination.

[0038] Preferably, the result evaluation in step S5 specifically includes the following steps:

[0039] E1. Quantitative evaluation: Use accuracy indicators to evaluate the model's terrain classification accuracy, record the model's running time, evaluate the model's computational efficiency, and evaluate the model's robustness to noise and missing data by adding different degrees of noise or missing data to the data;

[0040] E2. Error analysis: Analyze the errors generated by the model and find out the main sources of the errors. The error analysis will combine quantitative indicators and qualitative evaluation results to deeply study the performance of the model on different terrain types and analyze the relationship between errors and data quality, model parameters and algorithm selection.

[0041] (III) Beneficial effects

[0042] The present invention provides a method for constructing a terrain and geomorphology information model based on multi-source geographic data. Compared with the prior art, it has the following beneficial effects:

[0043] (1) This method of constructing a topographic information model based on multi-source geographic data enables efficient and convenient integration and processing of geographic data from different sources and formats by designing a new data integration framework. This framework will allow the algorithm to automatically identify data types and convert data formats in real time, thereby reducing manual participation and improving the efficiency and accuracy of data integration.

[0044] (2) This method of constructing a topographic information model based on multi-source geographic data enables efficient and convenient integration and processing of geographic data from different sources and formats by designing a new data integration framework. This framework will allow the algorithm to automatically identify data types and convert data formats in real time, thereby reducing manual participation and improving the efficiency and accuracy of data integration.

[0045] (3) The method for constructing a topographic and geomorphic information model based on multi-source geographic data adopts a more accurate algorithm model to address the error problem introduced by data processing in the prior art. These algorithms can retain sufficient detail information while integrating data, optimize the expressiveness of the model, and improve the model's level of detail, especially in the analysis of complex terrain. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 The present invention is a flowchart of a method for constructing a terrain and geomorphology information model based on multi-source geographic data. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] See also Figure 1 The present invention provides two technical solutions: a method for constructing a terrain and landform information model based on multi-source geographic data, which specifically includes the following embodiments:

[0049] Embodiment 1: A method for constructing a terrain and geomorphology information model based on multi-source geographic data, specifically comprising the following steps:

[0050] S1. Data preprocessing: Collect multi-source geographic data including remote sensing images, digital elevation models (DEMs), terrain data, and geological data, perform quality checks on the collected data, remove noise and outliers, convert the formats of data from different data sources, and unify the coordinates for subsequent fusion processing;

[0051] S2. Feature extraction: Extract features from the preprocessed geographic data, including topographic features, geomorphic features, and geological features, and use feature extraction algorithms to improve the accuracy and efficiency of feature extraction;

[0052] S3, model construction: select appropriate topographic information model, use the extracted feature data, adopt machine learning algorithms and statistical methods to build the model, and in the process of model construction, consider the weight and correlation of multi-source geographic data to improve the accuracy and reliability of the model;

[0053] S4. Model optimization: Use cross-validation and regularization methods to optimize the constructed model to prevent overfitting and underfitting, adjust and optimize the model parameters, and improve the performance and generalization ability of the model.

[0054] S5. Result evaluation: Use the mean absolute error (MAE) and correlation coefficient (R) evaluation indicators to evaluate the constructed topographic information model, compare and analyze the evaluation results with the actual topographic data, verify the accuracy and reliability of the model, and improve and optimize the model based on the evaluation results.

[0055] In the embodiment of the present invention, in step S2, the terrain features are slope, aspect and roughness, the landform features are valleys, ridges and peaks, and the geological features are strata and lithology.

[0056] In the embodiment of the present invention, the data preprocessing in step S1 specifically includes the following steps:

[0057] A1. Data collection;

[0058] A2. Data quality check: Check the quality of the collected geographic data, including the integrity, accuracy, and consistency of the data, remove noise and outliers, use median filtering and mean filtering methods to smooth the data, remove noise interference, fill in missing values ​​in the data, and use interpolation and regression analysis methods to estimate and fill in missing values;

[0059] A3. Data format conversion and coordinate unification: Convert the data from different data sources into commonly used data formats, such as Geotiff, Shapefile, etc. Unify the coordinates of data from different data sources, and use geographic coordinate systems (such as WGS84) or projection coordinate systems (such as UTM) to convert the data to ensure spatial consistency of the data.

[0060] In the embodiment of the present invention, the data collection in step A1 specifically includes the following steps:

[0061] T1. Remote sensing image collection: high-resolution image data acquired through satellite or aerial remote sensing platforms, including images in visible light, infrared, microwave and other bands;

[0062] T2. Digital elevation model (DEM) collection: Digital elevation model data obtained through terrain surveying, radar interferometry and other technologies, reflecting the elevation information of the terrain;

[0063] T3. Topographic data collection: including traditional topographic data such as topographic maps and contour maps, as well as high-precision topographic data obtained through lidar measurement technology;

[0064] T4. Geological data collection: including geological maps, stratigraphic lithology data, and geological structure data, reflecting the impact of geological conditions on topography.

[0065] In the embodiment of the present invention, the feature extraction in step S2 specifically includes the following steps:

[0066] B1. Terrain feature extraction: Use DEM-based slope calculation methods, such as the maximum slope method and the average slope method, to calculate the slope information of the terrain; use DEM-based slope aspect calculation methods, such as the eight-direction method and the sixteen-direction method, to calculate the slope aspect information of the terrain; use the terrain undulation-based roughness calculation method to calculate the terrain roughness information;

[0067] B2. Geomorphic feature extraction: Use methods based on terrain morphology, such as the valley point method and the ridge point method, to extract valley and ridge information in the terrain. Use methods based on terrain elevation, such as the local maximum method and the contour closure method, to extract peak information in the terrain.

[0068] In the embodiment of the present invention, the model building in step S3 specifically includes the following steps:

[0069] C1. Model selection: According to the topographic and geomorphic characteristics of the study area and application requirements, select appropriate topographic and geomorphic information models, such as DTM, DSM, TFM, etc.;

[0070] C2. Model construction method: neural network and support vector machine machine learning algorithms are used to build models using the extracted feature data; regression analysis and cluster analysis statistical methods are used to build models using the extracted feature data;

[0071] C3. Multi-source data fusion: In the process of model construction, the weights and correlations of multi-source geographic data are considered, and the weighted average method and principal component analysis method are used to fuse multi-source data.

[0072] Embodiment 2: The technical solution of the embodiment of the present invention is different from that of embodiment 1 in that the model optimization in step S4 specifically includes the following steps:

[0073] D1. Cross-validation: Divide the data set into a training set and a test set, and use the cross-validation method to train and validate the model to prevent overfitting. Adjust the model parameters, such as the number of neurons, learning rate, regularization parameters, etc., to improve the performance and generalization ability of the model.

[0074] D2. Structural optimization: In order to improve the expressiveness and adaptability of the model, a modular design is adopted, allowing users to select different model components according to specific needs, such as different feature extraction modules, fusion modules and classification modules. The optimization of the model structure will be completed by experimentally comparing the performance of different model components and selecting the optimal combination.

[0075] In the embodiment of the present invention, the result evaluation in step S5 specifically includes the following steps:

[0076] E1. Quantitative evaluation: Use accuracy indicators to evaluate the model's terrain classification accuracy, record the model's running time, evaluate the model's computational efficiency, and evaluate the model's robustness to noise and missing data by adding different degrees of noise or missing data to the data;

[0077] E2. Error analysis: Analyze the errors generated by the model, find out the main sources of errors, and provide a basis for further optimization of the model. Error analysis will combine quantitative indicators and qualitative evaluation results to deeply study the performance of the model on different terrain types, and analyze the relationship between errors and data quality, model parameters and algorithm selection.

[0078] In summary, the present invention designs a new data integration framework so that geographic data from different sources and in different formats can be integrated and processed efficiently and conveniently. The framework will allow the algorithm to automatically identify the data type and convert the data format in real time, thereby reducing manual participation and improving the efficiency and accuracy of data integration. By designing a new data integration framework, geographic data from different sources and in different formats can be integrated and processed efficiently and conveniently. The framework will allow the algorithm to automatically identify the data type and convert the data format in real time, thereby reducing manual participation and improving the efficiency and accuracy of data integration. In view of the error problem introduced by data processing in the prior art, the present invention adopts a more accurate algorithm model. These algorithms can retain sufficient detail information while integrating data, optimize the expressiveness of the model, and improve the meticulousness of the model, especially in the analysis of complex terrain.

[0079] Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0080] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0081] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for constructing a terrain and geomorphology information model based on multi-source geographic data, characterized by: The specific steps include: S1. Data preprocessing: Collect multi-source geographic data including remote sensing images, digital elevation models (DEMs), terrain data, and geological data, perform quality checks on the collected data, remove noise and outliers, convert the formats of data from different data sources, and unify the coordinates for subsequent fusion processing; S2. Feature extraction: Extract features from the preprocessed geographic data, including topographic features, geomorphic features, and geological features, and use feature extraction algorithms to improve the accuracy and efficiency of feature extraction; S3, model construction: select appropriate topographic information model, use the extracted feature data, adopt machine learning algorithms and statistical methods to build the model, and in the process of model construction, consider the weight and correlation of multi-source geographic data to improve the accuracy and reliability of the model; S4. Model optimization: Use cross-validation and regularization methods to optimize the constructed model to prevent overfitting and underfitting, adjust and optimize the model parameters, and improve the performance and generalization ability of the model. S5. Result evaluation: Use the mean absolute error (MAE) and correlation coefficient (R) evaluation indicators to evaluate the constructed topographic information model, compare and analyze the evaluation results with the actual topographic data, verify the accuracy and reliability of the model, and improve and optimize the model based on the evaluation results.

2. A method for constructing a terrain and geomorphology information model based on multi-source geographic data according to claim 1, characterized in that: In step S2, the terrain features are slope, aspect and roughness, the landform features are valleys, ridges and peaks, and the geological features are strata and lithology.

3. The method for constructing a terrain and geomorphology information model based on multi-source geographic data according to claim 1, characterized in that: The data preprocessing in step S1 specifically includes the following steps: A1. Data collection; A2. Data quality check: Check the quality of the collected geographic data, including the integrity, accuracy, and consistency of the data, remove noise and outliers, use median filtering and mean filtering methods to smooth the data, remove noise interference, fill in missing values ​​in the data, and use interpolation and regression analysis methods to estimate and fill in missing values; A3. Data format conversion and coordinate unification: Convert the data from different data sources into a common data format, unify the coordinates of the data from different data sources, and use the geographic coordinate system or projection coordinate system to convert the data to ensure the spatial consistency of the data.

4. The method for constructing a topographic information model based on multi-source geographic data according to claim 3, characterized in that: The data collection in step A1 specifically includes the following steps: T1. Remote sensing image collection: high-resolution image data acquired through satellite or aerial remote sensing platforms, including images in visible light, infrared, microwave and other bands; T2. Digital elevation model (DEM) collection: Digital elevation model data obtained through terrain surveying, radar interferometry and other technologies, reflecting the elevation information of the terrain; T3. Topographic data collection: including traditional topographic data such as topographic maps and contour maps, as well as high-precision topographic data obtained through lidar measurement technology; T4. Geological data collection: including geological maps, stratigraphic lithology data, and geological structure data, reflecting the impact of geological conditions on topography.

5. The method for constructing a terrain and geomorphology information model based on multi-source geographic data according to claim 1, characterized in that: The feature extraction in step S2 specifically includes the following steps: B1. Terrain feature extraction: The slope calculation method based on DEM is used to calculate the slope information of the terrain. The slope aspect calculation method based on DEM is used to calculate the slope aspect information of the terrain. The roughness calculation method based on terrain undulation is used to calculate the roughness information of the terrain. B2. Geomorphic feature extraction: A method based on terrain morphology is used to extract valley and ridge information from the terrain, and a method based on terrain elevation is used to extract peak information from the terrain.

6. The method for constructing a terrain and geomorphology information model based on multi-source geographic data according to claim 1, characterized in that: The model building in step S3 specifically includes the following steps: C1. Model selection: Select an appropriate topographic and geomorphic information model based on the topographic and geomorphic characteristics of the study area and application requirements; C2. Model construction method: neural network and support vector machine machine learning algorithms are used to build models using the extracted feature data; regression analysis and cluster analysis statistical methods are used to build models using the extracted feature data; C3. Multi-source data fusion: In the process of model construction, the weights and correlations of multi-source geographic data are considered, and the weighted average method and principal component analysis method are used to fuse multi-source data.

7. The method for constructing a terrain and geomorphology information model based on multi-source geographic data according to claim 1, characterized in that: The model optimization in step S4 specifically includes the following steps: D1. Cross-validation: Divide the data set into a training set and a test set, and use the cross-validation method to train and validate the model to prevent overfitting, adjust the model parameters, and improve the performance and generalization ability of the model; D2. Structural optimization: A modular design is adopted to allow users to select different model components according to specific needs. The optimization of the model structure will be completed by experimentally comparing the performance of different model components and selecting the optimal combination.

8. The method for constructing a topographic information model based on multi-source geographic data according to claim 1, characterized in that: The result evaluation in step S5 specifically includes the following steps: E1. Quantitative evaluation: Use accuracy indicators to evaluate the model's terrain classification accuracy, record the model's running time, evaluate the model's computational efficiency, and evaluate the model's robustness to noise and missing data by adding different degrees of noise or missing data to the data; E2. Error analysis: Analyze the errors generated by the model and find out the main sources of the errors. The error analysis will combine quantitative indicators and qualitative evaluation results to deeply study the performance of the model on different terrain types and analyze the relationship between errors and data quality, model parameters and algorithm selection.

Images