An intelligent terrain feature data quality control system and method

Abstract

The application provides a kind of intelligent terrain feature data quality control system and method, it is related to data quality inspection technical field.System includes following module: data acquisition module, data preprocessing module, rule base construction module, quality inspection module, score management module and report generation module.Data acquisition module is used to collect terrain feature dataset;Data preprocessing module carries out noise filtering according to the improved terrain feature maintaining filter algorithm, generates preprocessed dataset;Rule base construction module designs and configures quality inspection rule, to construct rule base;Quality inspection module scans dataset according to rule base, and outputs heat map;Score management module scores data according to heat map, correction trajectory and rule base data, generates score data;Report generation module is used to visualize the score data.The application effectively solves the problems of low efficiency, insufficient accuracy and lack of systematic quality information management in the prior art of terrain feature data quality inspection.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data quality inspection, in particular to an intelligent terrain feature data quality control system and method. BACKGROUND

[0002] Terrain feature data is widely used in geographic information systems (GIS), urban planning, environmental monitoring, and natural disaster assessment. However, with the advent of the big data era, the generation and collection of terrain feature data is extremely large, and data quality problems gradually emerge, mainly in noise, data missing, and logical inconsistency. Traditional quality control methods rely heavily on manual inspection, which is inefficient and prone to subjective bias. In real-world applications, there are often several technical problems: (1) noise interference in data processing, leading to loss or misjudgment of terrain features; (2) lack of flexibility in quality inspection rules, which cannot be adjusted in real time according to data characteristics; (3) the combination of automated quality inspection and manual review is not efficient enough, leading to an extended overall quality inspection period; (4) quality information management methods lack systematization, and quality inspection records are difficult to trace, affecting the credibility of the data. SUMMARY

[0003] In order to overcome the deficiencies of the prior art, the purpose of the present application is to provide an intelligent terrain feature data quality control system and method, which solves the problems of low efficiency, insufficient accuracy, and lack of systematicness in quality information management in the prior art.

[0004] To achieve the above-mentioned purpose, the present application provides the following solutions:

[0005] An intelligent terrain feature data quality control system, comprising:

[0006] a data acquisition module, a data preprocessing module, a rule base construction module, a quality inspection module, a scoring management module, and a report generation module connected in sequence;

[0007] The data acquisition module is configured to acquire a terrain feature data set, the data preprocessing module is configured to filter out noise from the terrain feature data set according to an improved terrain feature preserving filtering algorithm to obtain a preprocessed terrain feature data set, the rule base construction module is configured to design and configure quality inspection rules based on the preprocessed terrain feature data set to construct a rule base, the quality inspection module is configured to identify and scan the preprocessed terrain feature data set according to the rule base to obtain an error spatial distribution heat map, the scoring management module is configured to score the current terrain feature data according to the error spatial distribution heat map, the correction trace data, and the rule base data to obtain scoring data, and the report generation module is configured to visually display the current scoring data.

[0008] Preferably, the data preprocessing module comprises:

[0009] a first data processing submodule, a judgment submodule, a second data processing submodule, a third data processing submodule, and a data integration submodule;

[0010] The first data processing submodule is configured to calculate a terrain feature sensitive factor of the terrain feature dataset, the judgment submodule is configured to perform size judgment on the terrain feature sensitive factor by using a preset terrain feature sensitive factor threshold to obtain a judgment result, wherein the judgment result includes being greater than or equal to the preset terrain feature sensitive factor threshold and being less than the preset terrain feature sensitive factor threshold, the second data processing submodule is configured to perform noise filtering on the terrain feature dataset by using median filtering and gradient direction constraint according to a time corresponding to the judgment result being greater than or equal to the preset terrain feature sensitive factor threshold to obtain a first preprocessed terrain feature data subset, the third data processing submodule is configured to perform smoothing processing on the terrain feature dataset by using an adaptive Gaussian kernel according to a time corresponding to the judgment result being less than the preset terrain feature sensitive factor threshold to obtain a second preprocessed terrain feature data subset, and the data integration submodule is configured to integrate the first preprocessed terrain feature data subset and the second preprocessed terrain feature data subset to obtain a preprocessed terrain feature dataset.

[0011] Preferably, the first data processing submodule comprises:

[0012] an elevation gradient calculation unit, a curvature standard deviation calculation unit, a regional average elevation calculation unit, and a terrain feature sensitive factor calculation unit;

[0013] The elevation gradient calculation unit is configured to calculate an elevation gradient of the terrain feature dataset, the curvature standard deviation calculation unit is configured to calculate a curvature standard deviation of the terrain feature dataset, the regional average elevation calculation unit is configured to calculate a regional average elevation of the terrain feature dataset, and the terrain feature sensitive factor calculation unit is configured to calculate a terrain feature sensitive factor of the terrain feature dataset according to the elevation gradient, the curvature standard deviation, and the regional average elevation;

[0014] The calculation expression of the elevation gradient is as follows:

[0015]

[0016] The calculation expression of the curvature standard deviation is as follows:

[0017]

[0018] The calculation expression of the regional average elevation is as follows:

[0019]

[0020] The calculation expression of the terrain feature sensitive factor is:

[0021]

[0022] wherein, is the elevation change rate in the x direction, is the elevation change rate in the y direction, η x is the terrain strike weight factor in the x direction, η y is the terrain strike weight factor in the y direction, is the numerical value of the elevation gradient, κ i is the single-point curvature value, is the average curvature value in the local window, N is the number of data points in the local window, skew(κ) is the skewness of the curvature distribution in the window, σ curv is the curvature standard deviation, M is the number of elevation zones, Area j is the projection area of the jth elevation zone, Area total is the total area of the region, is the average elevation of all points in the jth elevation zone, is the standard deviation of the elevation in the jth elevation zone, max(σ h ) is the maximum value of the standard deviation of all elevation zones in the entire region, w j is the zoning weight, δ is the terrain feature sensitive factor, and rugosity is the surface roughness.

[0023] Preferably, the second data processing sub-module comprises:

[0024] a median filter unit, a gradient direction constraint unit, a noise discrimination unit, and a data integration unit.

[0025] The median filter unit is configured to perform median filtering on the terrain element dataset to obtain a preliminary data subset after median filtering. The gradient direction constraint unit is configured to apply gradient direction constraints to the preliminary data subset based on elevation gradient direction information to obtain gradient direction constraint data. The noise discrimination unit is configured to classify the gradient direction constraint data to determine high sensitive factor region data and low sensitive factor region data. The data integration unit is configured to perform secondary filtering on the high sensitive factor region data and integrate the filtered data and the low sensitive factor region data to obtain a first preprocessed terrain element data subset.

[0026] Preferably, the third data processing sub-module comprises:

[0027] an adaptive Gaussian kernel smoother unit, a feature maintenance unit, and an accuracy verification unit.

[0028] The adaptive Gaussian kernel smoother unit is configured to perform adaptive Gaussian kernel smoothing on the terrain feature dataset to obtain an adaptively smoothed data subset, the feature maintenance unit is configured to perform feature stabilization on the adaptively smoothed data subset by using a feature maintenance algorithm to obtain a stabilized feature dataset, and the accuracy verification unit is configured to perform accuracy evaluation on the stabilized feature dataset by using historical data and to remove data with an evaluation result lower than an evaluation threshold to obtain a second preprocessed terrain feature data subset.

[0029] Preferably, the rule base construction module comprises:

[0030] a feature analysis unit, a rule design unit, and a rule configuration unit;

[0031] The data feature analysis unit is configured to extract key features and abnormal patterns from the preprocessed terrain feature dataset and generate feature description information, the rule design unit is configured to design quality inspection rules according to the extracted feature description information and in combination with domain knowledge and experience, and the rule configuration unit is configured to parameterize the designed quality inspection rules and store them into a target rule base to construct a final rule base.

[0032] Preferably, the quality inspection module comprises:

[0033] a data recognition unit, a rule execution unit, an error detection unit, and a result feedback unit;

[0034] The data recognition unit is configured to receive the preprocessed terrain feature dataset and recognize elements that need to be subjected to quality inspection, the rule execution unit is configured to perform scanning analysis on the elements that need to be subjected to quality inspection according to the rule base to obtain preliminary quality inspection results, the error detection unit is configured to perform error type analysis and error location analysis on the preliminary quality inspection results to obtain final quality inspection results, and the result feedback unit is configured to compare the elements that need to be subjected to quality inspection with the preprocessed terrain feature dataset to obtain a spatial heat distribution map.

[0035] Preferably, the visualization module is composed of a plurality of screens.

[0036] Preferably, the scoring management module and the report generation module interact with each other through a feedback channel.

[0037] An intelligent terrain feature data quality control method, comprising:

[0038] obtaining a terrain feature dataset;

[0039] performing noise filtering on the terrain feature dataset according to an improved terrain feature maintenance type filtering algorithm to obtain a preprocessed terrain feature dataset;

[0040] design and configure quality inspection rules according to the preprocessed terrain feature data set to build a rule base, and perform identification scanning on the preprocessed terrain feature data set according to the rule base to obtain an error spatial distribution heat map;

[0041] score current terrain feature data according to the error spatial distribution heat map, correction trace data and rule base data to obtain scoring data;

[0042] visually display the current scoring data.

[0043] The present application discloses the following technical effects:

[0044] The present application provides an intelligent terrain feature data quality control system and method, which comprises, in sequence, a data acquisition module, a data preprocessing module, a rule base construction module, a quality inspection module, a scoring management module and a report generation module. The data acquisition module is used to acquire a terrain feature data set, the data preprocessing module is used to remove noise from the terrain feature data set according to an improved terrain feature preserving filtering algorithm to obtain a preprocessed terrain feature data set, the rule base construction module is used to design and configure quality inspection rules according to the preprocessed terrain feature data set to build a rule base, the quality inspection module is used to perform identification scanning on the preprocessed terrain feature data set according to the rule base to obtain an error spatial distribution heat map, the scoring management module is used to score current terrain feature data according to the error spatial distribution heat map, correction trace data and rule base data to obtain scoring data, and the report generation module is used to visually display the current scoring data. The present application effectively integrates various links of data acquisition, processing, analysis and display, not only improves the quality management efficiency and accuracy of terrain feature data, but also provides reliable data support for subsequent application, and realizes intelligent decision-making. BRIEF DESCRIPTION OF DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0046] Figure 1 A structural schematic diagram of an intelligent terrain feature data quality control system according to an embodiment of the present application.

[0047] Explanation of reference signs:

[0048] 1-Data acquisition module, 2-Data preprocessing module, 3-Rule base construction module, 4-Quality inspection module, 5-Scoring management module, 6-Report generation module. Detailed Implementation

[0049] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0050] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0051] like Figure 1 As shown, the present invention provides an intelligent terrain feature data quality control system, comprising:

[0052] The data acquisition module 1, data preprocessing module 2, rule base construction module 3, quality inspection module 4, scoring management module 5, and report generation module 6 are connected in sequence.

[0053] The data acquisition module 1 is used to acquire a terrain feature dataset. The data preprocessing module 2 is used to filter noise from the terrain feature dataset according to an improved terrain feature-preserving filtering algorithm to obtain a preprocessed terrain feature dataset. The rule base construction module 3 is used to design and configure quality inspection rules to build a rule base based on the preprocessed terrain feature dataset. The quality inspection module 4 is used to scan the preprocessed terrain feature dataset according to the rule base to obtain an error spatial distribution heatmap. The scoring management module 5 is used to score the current terrain feature data according to the error spatial distribution heatmap, correction trace data, and rule base data to obtain scoring data. The report generation module 6 is used to visualize the current scoring data.

[0054] Specifically, the terrain feature dataset includes:

[0055] Elevation data: Digital Elevation Model (DEM), Digital Surface Model (DSM), Digital Terrain Model (DTM). These data show the changes in ground elevation and depict the basic features of the terrain.

[0056] Slope and aspect data: This includes the slope and direction of the terrain, which is important for analyzing topographic and hydrological models.

[0057] Terrain curvature data: Includes single-point curvature, average curvature, and elevation rate of change, showing the degree of terrain undulation, which is crucial for analyzing topographic features.

[0058] Surface roughness data: Reflects the roughness of the ground, affecting processes such as water flow and soil erosion.

[0059] Ground element data: Covers the spatial location information and attributes of buildings, roads, water bodies, vegetation, etc. on the ground, providing a basic background for comprehensive analysis.

[0060] Methods for obtaining these terrain element data include remote sensing technology, ground measurement, geographic information system (GIS), and field investigation.

[0061] Methods for obtaining these terrain element data sets, remote sensing technology: Obtain surface information through satellites or drones; ground measurement is to obtain accurate data through actual measurement; geographic information system (GIS) processes and analyzes these data; field investigation ensures the authenticity and accuracy of the data.

[0062] More specifically, quality inspection rules include:

[0063] Elevation data rules: Elevation range rules: Set a reasonable range for elevation data, define the minimum and maximum elevation threshold values according to the natural characteristics of the terrain area (such as mountains, depressions, etc.). Any elevation value outside this range will be marked as abnormal.

[0064] Elevation rate of change rules: Regulate the elevation rate of change between adjacent data points to prevent unreasonable large fluctuations caused by data acquisition errors. For example, if the elevation difference between adjacent points is greater than the pre-set threshold value, it is considered abnormal.

[0065] Slope rules: Slope threshold rules: Set a reasonable range for slope, expressed in numerical form as the maximum slope angle. For example, if the slope exceeds a certain angle, it is marked as an area that needs to be reviewed in detail.

[0066] Slope rate of change rules: Detect the change in slope within a continuous area and set a threshold for the slope rate of change to avoid abnormal slope changes caused by errors in the data collection process.

[0067] Curvature rules: Curvature range rules: For curvature data, set a reasonable range of curvature values to detect potential data errors. Excessive or insufficient curvature values may indicate data inaccuracy.

[0068] Curvature change rules: Analyze the curvature change within adjacent areas and set features with excessively prominent changes as abnormal to ensure the reasonable presentation of topographic features.

[0069] Data completeness rule: missing value rule: identify and record the situation of missing values in the dataset, set the tolerance and processing strategy of missing values.

[0070] Data sparsity rule: define the rule for processing sparse terrain data points, set the threshold of point density to ensure the continuity and integrity of terrain features, and avoid feature distortion caused by data sparsity.

[0071] Configuration of quality inspection rules

[0072] In the rule library construction module 3, each quality inspection rule will be configured according to the following aspects:

[0073] Parameterized setting: define specific numerical thresholds and sensitive factors for each rule to adapt to the characteristics of different datasets.

[0074] Priority arrangement: set the inspection priority of different rules to ensure that key issues can be given priority and attention.

[0075] Rule documentation: record the definition, scope of application and use conditions of the rule in the form of document, for subsequent query and application.

[0076] Further, the data preprocessing module 2 comprises:

[0077] The first data processing submodule, the judgment submodule, the second data processing submodule, the third data processing submodule and the data integration submodule;

[0078] The first data processing submodule is used to calculate the terrain feature sensitive factor of the terrain feature data set, the judgment submodule is used to judge the size of the terrain feature sensitive factor by using the preset terrain feature sensitive factor threshold, and the judgment result is obtained, wherein the judgment result includes greater than or equal to the preset terrain feature sensitive factor threshold and less than the preset terrain feature sensitive factor threshold, the second data processing submodule is used to filter out noise from the terrain feature data set by using median filtering and gradient direction constraint according to the time corresponding to the judgment result greater than or equal to the preset terrain feature sensitive factor threshold, and obtain the first preprocessed terrain feature data subset, the third data processing submodule is used to process the time corresponding to the judgment result less than the preset terrain feature sensitive factor threshold, and the terrain feature data set is smoothed by using adaptive Gaussian kernel, and the second preprocessed terrain feature data subset is obtained, and the data integration submodule is used to integrate the first preprocessed terrain feature data subset and the second preprocessed terrain feature data subset, and obtain the preprocessed terrain feature data set.

[0079] Further, the first data processing submodule comprises:

[0080] a height gradient calculation unit, a curvature standard deviation calculation unit, a regional average height calculation unit, and a terrain feature sensitive factor calculation unit;

[0081] The height gradient calculation unit is configured to calculate a height gradient of the terrain feature data set, the curvature standard deviation calculation unit is configured to calculate a curvature standard deviation of the terrain feature data set, the regional average height calculation unit is configured to calculate a regional average height of the terrain feature data set, and the terrain feature sensitive factor calculation unit is configured to calculate a terrain feature sensitive factor of the terrain feature data set according to the height gradient, the curvature standard deviation, and the regional average height.

[0082] The calculation expression of the height gradient is as follows:

[0083]

[0084] The calculation expression of the curvature standard deviation is as follows:

[0085]

[0086] The calculation expression of the regional average height is as follows:

[0087]

[0088] The calculation expression of the terrain feature sensitive factor is as follows:

[0089]

[0090] wherein, is a height variation rate in the x direction, is a height variation rate in the y direction, x is a terrain strike weight factor in the x direction, y is a terrain strike weight factor in the y direction, is a numerical value of the height gradient, i is a single-point curvature value, is a local window mean curvature, N is a number of data points in the local window, skew(κ) is a skewness of the curvature distribution in the window, and σ curv is the curvature standard deviation, M is a number of height zoning, and Area j is a projection area of the jth height zoning, Area total is a total area of the region, is an average height of all points in the jth height zoning, is a standard deviation of the height in the jth height zoning, and max(σ h ) is a maximum value of the standard deviations of all height zonings in the entire region, and w jwhere δ is a terrain feature sensitivity factor, and rugosity is a surface roughness.

[0091] Specifically, the terrain orientation weight factor indicates that in the terrain transformation, some directions can be more important than others, and the weight factor enables this difference to be reflected in the analysis process;

[0092] The skewness of the curvature distribution in the window is an index that describes the shape of the curvature value distribution in the window. Skewness can determine the symmetry and deviation of the data distribution.

[0093] Specifically, η x η y Adaptive adjustment based on terrain orientation → solve anisotropic gradient distortion;

[0094] w j Correlation area and elevation variation → prevent flat areas from masking mountain features;

[0095] Skewness (κ) amplifies cliff signals → distinguishes noise random fluctuations from real micro-topography skewness distribution;

[0096] log(1+rugosity) enhances rough terrain weight → enhances gully / avalanche detection sensitivity;

[0097] Zoning weight normalization → ensures that the elevation representative is not affected by the area scale.

[0098] Further, the second data processing sub-module comprises:

[0099] A median filter unit, a gradient direction constraint unit, a noise discrimination unit, and a data integration unit;

[0100] The median filter unit is used for median filtering algorithm to process the terrain feature data set to obtain a preliminary data subset after median filtering, the gradient direction constraint unit is used for applying gradient direction constraint to the preliminary data subset based on elevation gradient direction information to obtain gradient direction constrained data, the noise discrimination unit is used for classifying the gradient direction constrained data to determine high sensitivity factor area data and low sensitivity factor area data, and the data integration unit is used for secondary filtering the high sensitivity factor area data and integrating the filtered data and the low sensitivity factor area data to obtain a first pre-processed terrain feature data subset.

[0101] Specifically, the specific steps of the classification process include:

[0102] The input is data processed by the gradient direction constraint, which contains elevation information and its corresponding gradient direction data;

[0103] Key features for classification are extracted from the gradient direction constrained data, including:

[0104] Gradient magnitude: Calculate the gradient magnitude of each data point to assess the steepness of the terrain.

[0105] Gradient direction: Refer to the direction of elevation change of each data point to determine its importance in classification.

[0106] Local curvature: Calculate the local curvature around each point to further judge the complexity of the terrain.

[0107] Based on the previously defined sensitive factor threshold, determine the criteria for high and low sensitive factors;

[0108] The classification results can be corrected by existing correction trace data and quality inspection rules to ensure classification accuracy; by comparing the data and subsequent results of high sensitive areas, the sensitive factor threshold can be adjusted to improve future classification accuracy.

[0109] Further, the third data processing sub-module includes:

[0110] Adaptive Gaussian kernel smoother unit, feature maintenance unit, accuracy verification unit;

[0111] The adaptive Gaussian kernel smoother unit is used for adaptive Gaussian kernel smoothing processing of the terrain feature data set to obtain an adaptive smoothed data subset, the feature maintenance unit is used for feature stability processing of the adaptive smoothed data subset by using a feature maintenance algorithm to obtain a stable feature data set, and the accuracy verification unit is used for accuracy evaluation of the stable feature data set by using historical data and removing data with evaluation results lower than an evaluation threshold to obtain a second preprocessed terrain feature data subset.

[0112] Specifically, the historical data is previous project data: based on terrain data in similar areas or adjacent time periods, which can effectively reflect the terrain changes and data stability.

[0113] Remote sensing image: use satellite or aerial remote sensing image data of different periods for comparison to provide a long-term perspective of terrain changes.

[0114] Ground measurement data: accurate elevation data obtained through ground survey, especially reference data in high sensitive areas.

[0115] Further, the rule base construction module 3 includes:

[0116] According to the feature analysis unit, the rule design unit and the rule configuration unit;

[0117] The data feature analysis unit is used to extract key features and abnormal patterns from the pre-processed terrain feature dataset and generate feature description information; the rule design unit designs quality inspection rules according to the extracted feature description information, combined with domain knowledge and experience; the rule configuration unit is used to parameterize the designed quality inspection rules and store them into the target rule library to construct the final rule library.

[0118] Specifically, the mean and standard deviation of the elevation, slope, slope direction and curvature information are calculated, and the frequency distribution is evaluated to determine the feature description information; the existing statistical analysis is used to determine the outliers and abnormal patterns in the dataset;

[0119] The rule design combined with domain knowledge and experience includes:

[0120] Standardized rules: use industry standards and regulations to design quality inspection rules, for example, define the reasonable value range of elevation data and slope;

[0121] Abnormal processing rules: design specific processing measures according to abnormal patterns, for example, set correction strategies for data points exceeding the elevation range;

[0122] Customized rules: design personalized rules for specific regions or projects according to data features to ensure that the rules can be applied to diversified datasets;

[0123] Rule documentation: all designed quality inspection rules should be recorded in detail, including rule purpose, application instructions, design basis and how to apply to the dataset.

[0124] Parameterization configuration:

[0125] Parameter setting: define necessary parameter values for each rule based on specific application scenarios and data features, such as elevation tolerance, slope and curvature adjustment threshold.

[0126] Priority management: set priority for rules to ensure that key rules are executed first in the quality inspection process.

[0127] Rule storage and management: store the configured rules in a standardized form in the database, for example, use structured query language (SQL) to define rule storage table, including rule ID, type, description, parameters, priority, etc.

[0128] Further, the quality inspection module 4 includes:

[0129] Data recognition unit, rule execution unit, error detection unit and result feedback unit;

[0130] The data identification unit is configured to receive the preprocessed terrain feature dataset and identify elements that need to be subjected to quality inspection; the rule execution unit is configured to obtain preliminary quality inspection results by scanning and analyzing the elements that need to be subjected to quality inspection according to the rule library; the error detection unit is configured to obtain final quality inspection results by performing error type analysis and error location analysis on the preliminary quality inspection results; and the result feedback unit is configured to compare the elements that need to be subjected to quality inspection with the preprocessed terrain feature dataset to obtain a spatial heat distribution map.

[0131] Specifically, data elements that need to be subjected to quality inspection, such as elevation points, slope points or feature areas, are identified based on specific conditions such as data validity, integrity, etc.

[0132] Marking elements that need to be subjected to quality inspection: marking the identified data elements that need to be subjected to quality inspection.

[0133] Heat map generation: generating a spatial heat distribution map according to the comparison results to show the error density and distribution of different areas.

[0134] The color gradient is used to represent the error density, with red representing high-frequency errors and green representing low-frequency errors.

[0135] Further, the visualization module is composed of several screens.

[0136] Further, the scoring management module 5 and the report generation module 6 interact with each other through a feedback channel.

[0137] The embodiment also provides an intelligent terrain feature data quality control method, which comprises:

[0138] Obtaining a terrain feature dataset;

[0139] Filtering noise from the terrain feature dataset according to an improved terrain feature preserving filtering algorithm to obtain a preprocessed terrain feature dataset;

[0140] Designing and configuring quality inspection rules according to the preprocessed terrain feature dataset to build a rule library; and identifying and scanning the preprocessed terrain feature dataset according to the rule library to obtain an error spatial distribution heat map;

[0141] Scoring the current terrain feature data according to the error spatial distribution heat map, the correction trace data and the rule library data to obtain scoring data;

[0142] Visualizing the current scoring data.

[0143] The various embodiments described in this specification are presented for the purpose of illustrating the principles of the present application and its best mode of operation. Each of the embodiments described in this specification has been provided for the purpose of illustration and is not intended to limit the application.

[0144] The principles and implementations of the present application have been described in the above embodiments, which are only used to help understand the method of the present application and its core idea. Meanwhile, for those skilled in the art, the specific implementation and application range of the present application can be changed according to the idea of the present application. In summary, the content of the specification should not be understood as a limitation of the present application.

Claims

1. An intelligent terrain feature data quality control system, characterized by, The application relates to a terrain feature data quality inspection method and device. The method comprises the following steps: sequentially connecting a data acquisition module, a data preprocessing module, a rule base construction module, a quality inspection module, a scoring management module and a report generation module; the data acquisition module is used for acquiring a terrain feature data set; the data preprocessing module is used for filtering noise of the terrain feature data set according to an improved terrain feature maintaining filtering algorithm to obtain a preprocessed terrain feature data set; the rule base construction module is used for designing and configuring quality inspection rules according to the preprocessed terrain feature data set to construct a rule base; the quality inspection module is used for identifying and scanning the preprocessed terrain feature data set according to the rule base to obtain an error spatial distribution heat map; the scoring management module is used for scoring current terrain feature data according to the error spatial distribution heat map, correction trace data and rule base data to obtain scoring data; and the report generation module is used for visually displaying the current scoring data. The data preprocessing module comprises a first data processing submodule, a judgment submodule, a second data processing submodule, a third data processing submodule and a data integration submodule. The first data processing submodule is used for calculating terrain feature sensitive factors of the terrain feature data set; the judgment submodule is used for judging the size of the terrain feature sensitive factors by using a preset terrain feature sensitive factor threshold to obtain a judgment result, wherein the judgment result comprises being greater than or equal to the preset terrain feature sensitive factor threshold and being less than the preset terrain feature sensitive factor threshold; the second data processing submodule is used for filtering noise of the terrain feature data set by using median filtering and gradient direction constraint according to a time corresponding to the judgment result greater than or equal to the preset terrain feature sensitive factor threshold to obtain a first preprocessed terrain feature data subset; the third data processing submodule is used for processing a time corresponding to the judgment result less than the preset terrain feature sensitive factor threshold to smooth the terrain feature data set by using an adaptive Gaussian kernel to obtain a second preprocessed terrain feature data subset; and the data integration submodule is used for integrating the first preprocessed terrain feature data subset and the second preprocessed terrain feature data subset to obtain the preprocessed terrain feature data set. The first data processing submodule comprises an elevation gradient calculation unit, a curvature standard deviation calculation unit, a regional average elevation calculation unit and a terrain feature sensitive factor calculation unit. The elevation gradient calculation unit is used for calculating the elevation gradient of the terrain feature data set; the curvature standard deviation calculation unit is used for calculating the curvature standard deviation of the terrain feature data set; the regional average elevation calculation unit is used for calculating the regional average elevation of the terrain feature data set; and the terrain feature sensitive factor calculation unit is used for calculating the terrain feature sensitive factors of the terrain feature data set according to the elevation gradient, the curvature standard deviation and the regional average elevation. The calculation expression of the elevation gradient is as follows: The calculation expression of the curvature standard deviation is as follows: The calculation expression of the regional average elevation is as follows: ​ ； ​ ； ​ ； The formula for calculating the terrain feature sensitivity factor is: ； wherein, is the rate of elevation change in the x-direction, is the rate of elevation change in the y-direction, is the terrain aspect weight factor in the x-direction, is the terrain aspect weight factor in the y-direction, is the value of the elevation gradient, is the single point curvature value, is the mean curvature value within the local window, is the number of data points within the local window, is the skewness of the curvature distribution within the window, is the curvature standard deviation, is the number of elevation zones, is the projected area of the jth elevation zone, is the total area of the region, is the mean elevation of all points within the jth elevation zone, is the standard deviation of the elevation within the jth elevation zone, is the maximum of all the elevation zone standard deviations for the entire region, is the zonation weight, is the terrain feature sensitivity factor, is the surface roughness.

2. The intelligent terrain feature data quality control system of claim 1, wherein, The second data processing submodule includes: Median filter unit, gradient direction constraint unit, noise discrimination unit, and data integration unit; The median filter unit is used to process the terrain feature dataset using a median filtering algorithm to obtain a preliminary data subset after median filtering. The gradient direction constraint unit is used to apply gradient direction constraints to the preliminary data subset based on elevation gradient direction information to obtain gradient direction constraint data. The noise discrimination unit is used to classify the gradient direction constraint data to determine high-sensitivity factor region data and low-sensitivity factor region data. The data integration unit is used to perform secondary filtering on the high-sensitivity factor region data and integrate the filtered data with the low-sensitivity factor region data to obtain a first preprocessed terrain feature data subset.

3. The intelligent terrain feature data quality control system of claim 1, wherein, The third data processing submodule includes: Adaptive Gaussian kernel smoother unit, feature preservation unit, accuracy verification unit; The adaptive Gaussian kernel smoother unit is used to perform adaptive Gaussian kernel smoothing on the terrain feature dataset to obtain an adaptively smoothed data subset. The feature preservation unit is used to perform feature stabilization processing on the adaptively smoothed data subset using a feature preservation algorithm to obtain a stable feature dataset. The accuracy verification unit is used to evaluate the accuracy of the stable feature dataset using historical data and remove data whose evaluation results are lower than the evaluation threshold to obtain a second preprocessed terrain feature data subset.

4. The intelligent terrain feature data quality control system of claim 1, wherein, The rule base construction module includes: Based on the feature analysis unit, rule design unit, and rule configuration unit; The data feature analysis unit is used to extract key features and anomaly patterns from the preprocessed terrain feature dataset and generate feature description information; the rule design unit designs adaptive quality inspection rules based on the extracted feature description information and combined with domain knowledge and experience; the rule configuration unit is used to parameterize the designed quality inspection rules and store them in the target rule base to build the final rule base.

5. The intelligent terrain feature data quality control system of claim 1, wherein, The quality inspection module includes: The system comprises a data identification unit, a rule execution unit, an error detection unit, and a result feedback unit. The data identification unit is used to receive the preprocessed terrain feature dataset and identify the elements that need to be inspected; the rule execution unit is used to scan and analyze the elements that need to be inspected according to the rule base to obtain preliminary inspection results; the error detection unit is used to perform error type analysis and error location analysis on the preliminary inspection results to obtain final inspection results; the result feedback unit is used to compare the elements that need to be inspected with the preprocessed terrain feature dataset to obtain a spatial heat map.

6. The intelligent terrain feature data quality control system of claim 1, wherein, The scoring management module and the report generation module interact with each other via a feedback channel.

7. An intelligent terrain feature data quality control method applied to the system of any one of claims 1 to 6, characterized in that, The method includes: Obtain the terrain feature dataset; The improved terrain feature-preserving filtering algorithm is used to filter out noise from the terrain feature dataset to obtain a preprocessed terrain feature dataset. Based on the preprocessed terrain feature dataset, quality inspection rules are designed and configured to build a rule base; the preprocessed terrain feature dataset is then scanned and identified according to the rule base to obtain an error spatial distribution heatmap. The current terrain feature data is scored based on the error spatial distribution heatmap, correction trace data, and rule base data to obtain score data; The current rating data is visualized.

Images