A river survey data preprocessing method based on artificial intelligence

Abstract

The present invention relates to the technical field of river section measurement, and in particular to a method for preprocessing river channel measurement data based on artificial intelligence. The method comprises the following steps: obtaining river channel section measurement point data, evaluating the density of measurement points based on different scales and changes in section elevation, thereby obtaining high-density measurement point data; identifying the deep-sea points and turning points of the section based on the changes in the slopes of adjacent measurement points, and performing measurement point thinning processing to retain the deep-sea points and turning points, thereby obtaining a retained measurement point data set; automatically detecting and calibrating measurement points that exceed a preset measurement point deviation distance threshold according to a preset section line and a retained measurement point data set, and reconstructing the coordinates and correcting the positions of measurement points that deviate from the preset section line, thereby obtaining calibrated measurement point data. The present invention can effectively solve the problems of data unevenness, missed measurements and deviations in traditional measurements, making river channel measurement data more accurate and complete.

Description

Technical Field

[0001] The invention relates to the technical field of river section measurement, and in particular to a river measurement data preprocessing method based on artificial intelligence. Background Art

[0002] Channel section is an important concept in river engineering and hydrology. It refers to the plane figure after vertical sectioning along a certain direction of the river. Channel sections are generally divided into two types: longitudinal section and cross section: Channel longitudinal section: a channel section vertically cut along the river deep line. The deep line refers to the line connecting the deepest points on each cross section along the river. The channel longitudinal section is composed of the water surface line and the deep riverbed line. It is a curve of the change of river bottom and water surface elevation along the course, which is often summarized by longitudinal slope or gradient. Channel cross section: a river channel section perpendicular to the mainstream line. Mountain rivers often present "V" or "J" shaped cross sections under the action of water erosion; the channel cross sections of plain rivers vary more and are related to the characteristics of the river. The cross section of the curved channel in the lower reaches of the Yellow River is an asymmetric triangle, while the cross section of the wandering channel is very irregular.

[0003] However, traditional river channel measurement data preprocessing methods often have the following problems: Due to the limitations of manual measurement or measurement equipment, river channel cross-section measurement often has sparse measurement points, especially in the deep area of ​​the river channel or the turning point of rapid changes, resulting in a lack of necessary fine data to reflect the true shape of the river channel. In some special terrains or complex waters, the measurement equipment may not be able to cover all areas, resulting in blank areas in some sections. These blank areas lack data support, affecting the comprehensive analysis of the river channel morphology. Summary of the invention

[0004] Based on this, it is necessary for the present invention to provide a river measurement data preprocessing method based on artificial intelligence to solve at least one of the above technical problems.

[0005] To achieve the above purpose, a river channel measurement data preprocessing method based on artificial intelligence includes the following steps:

[0006] Step S1: obtaining river section measuring point data, and evaluating the density of the river section measuring point data based on different scales and changes in section elevation, thereby obtaining high-density measuring point data;

[0007] Step S2: identifying the cross-section deep points and turning points based on the slope change of adjacent measuring points for the high-density measuring point data, and performing a measuring point thinning process to retain the deep points and turning points, thereby obtaining a retained measuring point data set;

[0008] Step S3: automatically detecting and calibrating the measuring points exceeding the preset measuring point deviation distance threshold according to the preset section line and the retained measuring point data set, and reconstructing the coordinates and correcting the positions of the measuring points deviating from the preset section line, thereby obtaining calibration measuring point data;

[0009] Step S4: identifying blank areas of the cross section according to the high-density measuring point data, and extracting distribution features of the cross section measuring points adjacent to the blank areas, thereby obtaining blank area measuring point data; and performing reserved point addition correction on the calibration measuring point data according to the blank area measuring point data, thereby obtaining corrected measuring point data;

[0010] Step S5: Verify the corrected measuring point data based on accuracy and consistency according to the preset engineering requirement data, so as to obtain river channel measurement data preprocessing data.

[0011] The present invention can adjust the density of measuring points appropriately for river sections in different regions by evaluating the density of measuring points, so as to ensure the accuracy of data. High-density measuring points can better reflect the complex changes of the river channel and avoid missing important terrain features, especially in areas with large terrain changes. According to different scales and elevation changes, the density of measuring points can be dynamically adjusted, measuring points can be added in more complex areas, unnecessary measuring points can be reduced, and the number of measuring points can be reduced in relatively flat areas, thereby effectively reducing data redundancy and improving measurement efficiency. By generating high-density data sets for different regions, more accurate data support can be provided for subsequent model establishment, further improving the availability and reliability of measurement data. By accurately identifying deep-sea points and turning points, key measuring point positions (such as deep-sea points and turning points) can be retained, and for those areas where the position changes are not large, thinning processing is performed to remove redundant data, thereby reducing the amount of data and improving processing efficiency. Deep-sea points and turning points are important terrain features of the river channel, which reflect the changing trend and structure of the river channel section. By retaining these key points, the validity of the data and the complete reflection of the river channel morphology can be ensured. The thinned data not only retains the necessary terrain information, but also improves the efficiency of data processing and analysis, and avoids processing delays caused by data redundancy. Automatic detection and calibration of deviated measuring points helps to find abnormal data caused by measurement errors or other factors, and timely correct the deviated measuring points to ensure the accuracy of the measurement data. Through coordinate reconstruction and position correction, the measuring points that deviate from the section line can be adjusted to the correct position to ensure that all measuring points meet the requirements of engineering design, thereby improving data consistency and meeting the needs of subsequent analysis. The accuracy of the calibrated measuring point data is improved, avoiding the risk of errors gradually amplifying in subsequent steps, thereby improving the reliability of the overall measurement data. By identifying blank areas, the vacancies in the measurement data can be found in time, which is crucial to ensure the comprehensiveness and continuity of the data. Using the distribution feature extraction method of adjacent section measuring points, the blank areas can be filled to avoid affecting the accuracy of subsequent analysis due to missing data. The generation of supplementary measuring points helps to improve the integrity of the measurement data, ensure the balanced data density of the entire river section, and avoid the problem of incomplete data caused by blank areas. By correcting the calibration point data with the blank area point data, the point data is further optimized to ensure the rationality and uniformity of the spatial distribution of the measurement data. Verifying the accuracy and consistency according to the preset engineering requirements can ensure that the corrected data meets the engineering design standards. Through this verification method, the points that still have problems can be effectively identified to further ensure the high quality of the data. Through the verification of accuracy and consistency, it is ensured that the final data can meet the requirements of subsequent engineering design, analysis and application, and avoid problems caused by inaccurate or inconsistent data.The verified data can not only ensure the feasibility of the design, but also provide reliable data support for construction, monitoring and subsequent maintenance, thereby improving the overall feasibility of the project. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

[0013] Figure 1 It is a schematic diagram of the steps of the river survey data preprocessing method based on artificial intelligence of the present invention;

[0014] Figure 2 for Figure 1 Detailed step flow diagram of step S1;

[0015] Figure 3 for Figure 1 Detailed step flow chart of step S2 in FIG. DETAILED DESCRIPTION

[0016] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0017] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0018] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0019] To achieve this, please refer to Figures 1 to 3The present invention provides a river channel measurement data preprocessing method based on artificial intelligence, the method comprising the following steps:

[0020] In the embodiment of the present invention, reference Figure 1 FIG. 1 is a schematic diagram of a process flow of a river channel measurement data preprocessing method based on artificial intelligence according to the present invention. In this example, the river channel measurement data preprocessing method based on artificial intelligence includes the following steps:

[0021] Step S1: obtaining river section measuring point data, and evaluating the density of the river section measuring point data based on different scales and changes in section elevation, thereby obtaining high-density measuring point data;

[0022] The embodiment of the present invention obtains the river section measurement point data, usually through remote sensing measurement equipment or hydrological measurement system to obtain the original river section data. After the data is obtained, the data is preprocessed using standardized methods of different scales, including projecting the measurement point data into a unified coordinate system, and calculating the distribution density of the measurement points through elevation changes. According to the change in the elevation of the section, an evaluation method based on the elevation change rate is used to evaluate the density of measurement points in different areas. For example, in areas where the depth of the river changes greatly, the measurement point density is higher, while the density is reduced in flat areas. A set of high-density measurement point data is obtained through this method, and these data will be used for subsequent identification of deep points and turning points. This method is suitable for large-scale river channel measurement, especially for rivers with drastic changes in water flow, such as rapids or bends.

[0023] Step S2: identifying the cross-section deep points and turning points based on the slope change of adjacent measuring points for the high-density measuring point data, and performing a measuring point thinning process to retain the deep points and turning points, thereby obtaining a retained measuring point data set;

[0024] The embodiment of the present invention processes high-density measuring point data and identifies deep points and turning points in the section by calculating the slope change rate between adjacent measuring points. Specifically, the three-point method is used to analyze the slope change between adjacent measuring points. If the change rate exceeds a preset threshold, the point is judged to be a deep point or a turning point. For deep points, the absolute difference method of elevation is used. If the elevation change exceeds the set threshold and presents extreme value characteristics, it is judged to be a deep point. For turning points, the curvature analysis of the slope change sequence between measuring points is used. If the curvature reaches a local extreme value at a certain point, the point is identified as a turning point. Then, the deep points and turning points are retained, and other measuring points are thinned out. The thinning method is based on the extraction of the geometric topological structure of the section to ensure that the shape and topological characteristics of the section remain consistent. Through this method, the redundancy of measuring points can be effectively removed, while retaining the measuring points that are most meaningful to the changes in river channel morphology.

[0025] Step S3: automatically detecting and calibrating the measuring points exceeding the preset measuring point deviation distance threshold according to the preset section line and the retained measuring point data set, and reconstructing the coordinates and correcting the positions of the measuring points deviating from the preset section line, thereby obtaining calibration measuring point data;

[0026] The embodiment of the present invention identifies the measuring points that deviate from the preset cross-section line by calculating the vertical distance between each measuring point in the retained measuring point data set and the preset cross-section line. At this time, the preset cross-section line can be obtained through the existing river terrain data or engineering design drawings, and the calculation method is based on the vertical projection model. For each deviated measuring point, a classification method based on the deviation direction of the measuring point is adopted to divide it into different categories, such as upward deviation or downward deviation. Then, by analyzing the error source of the deviated measuring point, considering the terrain influence and measurement error, a measuring point error feature data set is generated. According to the error feature data set, the coordinates of the deviated measuring point are reconstructed, and the reverse interpolation method is used to adjust it to the preset cross-section line position. After the measuring point coordinates are updated, the position is corrected, and finally the calibrated measuring point data is obtained. These calibrated measuring points will provide accurate benchmark data for the subsequent supplementation of blank areas.

[0027] Step S4: identifying blank areas of the cross section according to the high-density measuring point data, and extracting distribution features of the cross section measuring points adjacent to the blank areas, thereby obtaining blank area measuring point data; and performing reserved point addition correction on the calibration measuring point data according to the blank area measuring point data, thereby obtaining corrected measuring point data;

[0028] The embodiment of the present invention uses high-density measuring point data to identify blank areas of the cross section. The specific operation is to perform cluster analysis on the measuring point data by applying DBSCAN (density clustering algorithm) to identify areas with sparse measuring points as blank areas. Then, the boundary detection algorithm is used to extract the boundaries of the blank area to determine the shape and range of the blank area. The boundary shape and size are verified to ensure the accuracy of blank area identification. Next, the spatial density of the adjacent cross-section measuring points at the boundary of the blank area is analyzed to identify potential supplementary areas. Through geometric morphological change analysis, the connection points between the blank area and the adjacent measuring points are extracted to ensure the morphological continuity of the supplementary measuring points. Based on the inverse distance weighted method, supplementary measuring points are generated in the blank area, and quality assessment is performed based on the spatial distribution characteristics, elevation changes and density distribution of the measuring points to screen out high-quality supplementary measuring points. Finally, these blank area measuring point data are used to correct the calibration measuring point data to ensure the continuity and accuracy of the river data.

[0029] Step S5: Verify the corrected measuring point data based on accuracy and consistency according to the preset engineering requirement data, so as to obtain river channel measurement data preprocessing data.

[0030] The embodiment of the present invention verifies the accuracy and consistency of the corrected measuring point data according to the preset engineering requirement data. For example, the consistency of the data can be verified by comparing the historical measurement data, the existing engineering design standards or the measurement results of the adjacent areas. If there are significant errors in the measuring point data, recalibration and correction can be performed to ensure that the final data meets the engineering requirements. At the same time, the error analysis method is used in combination with the existing engineering model to perform a detailed accuracy assessment on the corrected data. This process ensures that the final output river measurement data has high reliability and meets the actual application needs, such as applications in the fields of river management, water conservancy project construction, etc.

[0031] As an embodiment of the present invention, refer to Figure 2 As shown, Figure 1 Detailed step flow diagram of step S1 in the embodiment of the present invention, step S1 includes the following steps:

[0032] Step S11: obtaining original measuring point data, removing invalid coordinates and measuring points beyond the geographical range according to the original measuring point data, and unifying the coordinate system, thereby obtaining river section measuring point data;

[0033] The embodiment of the present invention obtains the original data of the river section measuring points through remote sensing measurement equipment (such as laser radar, drone aerial photography or ground measurement equipment). The data contains the spatial coordinates of the measuring points (such as longitude and latitude, coordinate system information) and their corresponding elevation values. Then, invalid data such as missing coordinate values ​​and abnormal elevation of measuring points (such as measuring points below sea level or far away from the actual river channel range) are filtered out through an algorithm. These invalid data will be eliminated. Secondly, check whether the measuring points exceed the predetermined geographical range (such as setting a geographical boundary box to eliminate data with coordinates outside the frame). All valid measuring point data that meet the conditions will be converted into a unified coordinate system, such as a UTM coordinate system, to ensure the consistency of data from different sources in subsequent analysis. Finally, after the above cleaning steps, the obtained river section measuring point data will be used for subsequent spatial coordinate extraction and analysis.

[0034] Step S12: extracting the spatial coordinates of the measuring points on the river section, thereby obtaining the spatial coordinate data of the measuring points;

[0035] The embodiment of the present invention extracts the spatial coordinates of each measuring point from the acquired river section measuring point data, including the three-dimensional coordinate values ​​of X, Y, and Z. The measuring point data is organized into a table or array format using data processing software (such as Matlab, Pandas or NumPy library in Python) for subsequent processing. For each measuring point, its corresponding geographic coordinates (longitude, latitude) and elevation data are extracted. The purpose of this step is to convert the original measuring point data into structured data suitable for subsequent processing, ensuring that the coordinate data of each measuring point has a clear identification and is easy to calculate and analyze. This method is suitable for large-scale river section measurements, especially when spanning a large area, to ensure the spatial accuracy and consistency of the data.

[0036] Step S13: Calculate the elevation change rate according to the spatial coordinate data of the measuring points, thereby obtaining a multi-dimensional measuring point feature matrix;

[0037] The embodiment of the present invention first calculates the elevation change rate between each adjacent measuring point based on the measuring point spatial coordinate data extracted in step S12. Specifically, the elevation difference between adjacent measuring points is calculated using the two-point method, that is, the elevation change between two measuring points is calculated and divided by the distance between them. If the calculated elevation change rate exceeds the preset threshold, it is considered that the elevation change in the area is large. A set of elevation change rate data is generated by performing similar calculations on the elevation changes of all adjacent measuring points. Then, based on the elevation change rate and the spatial position of the measuring point, a multi-dimensional measuring point feature matrix is ​​constructed, which contains the elevation change rate, spatial position coordinates (X, Y, Z) and other possible features (such as river flow direction, etc.). This step can help identify areas with large changes in the river section, provide necessary information for subsequent density assessment and data processing, and is particularly suitable for areas with large changes in river morphology, such as sharp bends and rapids.

[0038] Step S14: Evaluate the measurement point density of the multi-dimensional measurement point feature matrix to obtain high-density measurement point data.

[0039] The embodiment of the present invention first performs a density assessment on the multi-dimensional measuring point feature matrix obtained in step S13. Specifically, a clustering algorithm based on Euclidean distance, such as K-means or DBSCAN, is used to group the measuring points, and the measuring points with close distances are classified into one category. The density difference between the measuring points reflects the distribution of measuring points in different areas of the river section. By calculating the distance between each measuring point and its neighboring measuring points, the spatial distribution of the measuring points on the section is evaluated. Then, based on the kernel density estimation (KDE) method, the local measuring point density is calculated to identify areas with high measuring point density. The evaluation results will help distinguish between measuring point clustering areas and sparse areas, and ultimately filter out high-density measuring point data. For example, in the bend or rapids area of ​​the river channel, the density of measuring points is higher, while in the gentle area, the density of measuring points is lower. The high-density measuring point data obtained by this method will provide high-quality input data for the subsequent identification of deep points and turning points, ensuring accurate capture of changes in river channel morphology.

[0040] The present invention eliminates invalid coordinates and measuring points beyond the geographic range, which helps to eliminate invalid data generated by measurement errors or abnormal situations (such as GPS signal loss, equipment failure, etc.). This ensures that subsequent analysis and modeling are based on valid and accurate measuring point data. By unifying the coordinate system, it is ensured that all measuring point data conform to the same geographic reference frame. This can avoid data confusion caused by different coordinate systems and improve the geographic consistency of data. Eliminating invalid or non-compliant measuring points can avoid the propagation of data errors in subsequent processing, prevent erroneous data from affecting analysis results, and ensure the quality and reliability of data. After data cleaning, the number of valid measuring points retained is reduced, the amount of data is reduced, the efficiency of data processing is improved, and computing resources and time are saved for subsequent steps. By extracting the spatial coordinate data of the measuring points, it can ensure that the spatial position of each measuring point is clearly described, supporting subsequent spatial analysis, modeling and visualization. Spatial coordinates are the basis for spatial analysis (such as cross-sectional morphology analysis, watershed analysis, etc.). Accurately extracting the spatial coordinates of the measuring points provides necessary data support for subsequent geographic information analysis, model building and numerical calculation. After extracting the spatial coordinates, all the measuring points can be converted into a standard spatial format to facilitate subsequent processing, such as geometric analysis, airspace division, and three-dimensional modeling. The elevation change rate can effectively reflect the slope and undulation changes at different locations of the river section, and provide important geographic information for the subsequent analysis of river morphology, landform changes, and water flow dynamics. Through the calculation of the elevation change rate, a single coordinate data can be converted into a data matrix containing multidimensional features such as slope and terrain change rate. These feature matrices can provide rich input data for subsequent pattern recognition, trend analysis, or modeling. The obtained multi-dimensional measuring point feature matrix can support more complex analysis. For example, machine learning modeling can be performed through the feature matrix to help identify the law of river depth changes, predict water flow trends, or evaluate the potential for river reconstruction. Through the evaluation of the density of measuring points, the density of measuring points can be determined based on features such as the elevation change rate, ensuring that there is sufficient density of measuring points in areas where the river morphology changes significantly, thereby improving the spatial resolution and accuracy of the data. Evaluating the density of measuring points helps ensure that measurements are covered at high density in key areas of the river section, while reducing unnecessary measurements in flat areas. This feature-based optimization can reduce unnecessary resource waste while ensuring data quality. By evaluating density and appropriately adjusting the distribution of measurement points, data redundancy can be reduced while ensuring sufficient coverage of key area data, thereby effectively controlling the amount of data and making subsequent analysis more efficient. By acquiring high-density measurement point data, river channel characteristics at different scales can be analyzed, which can not only evaluate the entire river channel morphology at a macro level, but also observe local changes at a micro scale, thus providing data support for multi-scale analysis.

[0041] Preferably, step S14 comprises the following steps:

[0042] Step S141: calculating the Euclidean distances between the measuring points on the multi-dimensional measuring point feature matrix, thereby obtaining a measuring point distance distribution matrix;

[0043] The embodiment of the present invention calculates the Euclidean distance between the measuring points based on the multi-dimensional measuring point feature matrix obtained in step S13. The specific operation is to use the standard Euclidean distance formula for each pair of measuring points in the matrix: ,in , , and , , are the coordinates of the two measuring points in three-dimensional space. This calculation will generate a measuring point distance matrix, and each matrix element represents the distance between the corresponding measuring point pairs. This step is suitable for processing high-precision river section data, and can accurately reflect the spatial relationship between each measuring point, especially in complex river morphology areas. The Euclidean distance matrix provides the basis for subsequent density calculation and regional division.

[0044] Step S142: performing local measurement point density calculation based on kernel density estimation on the measurement point distance distribution matrix, and performing measurement point spatial distribution uniformity evaluation, thereby obtaining measurement point density distribution data;

[0045] The embodiment of the present invention uses the kernel density estimation (KDE) method to calculate the local measuring point density based on the measuring point distance distribution matrix obtained in step S141. Specifically, the kernel density estimation method regards each measuring point as a probability density function, and calculates the measuring point density within a certain range around each measuring point according to a preset bandwidth parameter (such as the bandwidth parameter is selected as 0.1m). The KDE method obtains a continuous density estimate by smoothing the spatial distribution of the measuring points, which can accurately reflect the distribution characteristics of the measuring points on the river section. By calculating the local density of each measuring point, the measuring point clustering area and the sparse area can be identified. This step is suitable for application scenarios in which the measuring point distribution needs to be accurately identified in the river section measurement, especially when the measuring points are unevenly distributed, and the dense areas and blank areas can be effectively found.

[0046] Step S143: identifying the measurement point clustering area and the sparse area according to the measurement point density distribution data, thereby constructing a measurement point distribution heat map;

[0047] According to the measurement point density distribution data obtained in step S142, the embodiment of the present invention identifies the clustered areas and sparse areas of the measurement points through the threshold segmentation method. First, sort according to the size of the local density, and set a threshold (such as a threshold of 0.05 measurement points / m²), and define the area with a density greater than the threshold as the measurement point clustering area, and the area with a density lower than the threshold as the sparse area. Then, based on the spatial distribution of these areas, a heat map generation tool (such as Matplotlib or Seaborn library in Python) is used to construct a measurement point distribution heat map, and the depth of the color on the heat map represents the density of the measurement points. This step is suitable for evaluating whether the spatial distribution of the measurement points of the river section is uniform, especially for river sections with complex morphology, which can help identify which areas have too sparse or dense measurement points, thereby guiding subsequent measurement point screening and data supplementation.

[0048] Step S144: screening high-density measuring points based on a preset measuring point density threshold according to the measuring point distribution heat map and the multi-dimensional measuring point feature matrix, thereby obtaining high-density measuring point data.

[0049] The embodiment of the present invention uses a preset measurement point density threshold (such as 0.08 measurement points / m²) to screen high-density measurement points based on the measurement point distribution heat map and multi-dimensional measurement point feature matrix obtained in step S143. The specific operation is to obtain high-density measurement point data by screening measurement points with a density greater than the set threshold according to the clustered areas and sparse areas displayed by the heat map. At this time, for the darker areas on the heat map, it means that the measurement point density is higher, and the measurement points in these areas are selected as high-density measurement points. Then, combined with the spatial coordinates and elevation changes of each measurement point, the measurement points that meet the conditions are further screened. These screened high-density measurement points will serve as the basic data for the subsequent identification of deep points and turning points. This step is particularly suitable for areas in the river where the distribution of measurement points needs to be accurately grasped, such as rapids, river bends, etc., to ensure that there is enough measurement point data in these important areas to support subsequent analysis.

[0050] The present invention can quantitatively describe the spatial relationship between the measuring points by calculating the Euclidean distance between the measuring points. This is of great significance for evaluating the distribution density, regional morphology and location characteristics of the measuring points on the river section. The measuring point distance distribution matrix can reveal the spatial structure information of the measuring points, especially the distribution law of the measuring points in the region. Through this matrix, it can be judged that some areas have dense measuring points, while some areas are sparse, providing basic data for subsequent analysis. According to the distance between the measuring points, dense areas or areas where data redundancy may exist can be identified, thereby optimizing the subsequent measurement work, making data collection more targeted, saving time and resources. Kernel density estimation can calculate the density distribution of local areas according to the spatial position of the measuring points, and then find the concentrated areas and sparse areas of measuring points. This helps to reveal the density of measuring points on the river section and provide data support for subsequent measurement strategies. The spatial distribution uniformity evaluation of measuring points can help identify whether there is uneven density caused by regional changes or improper measurement methods during the measurement process. Through this evaluation, it can be judged which areas need more measuring points and which areas can reduce the density of measuring points, thereby optimizing the data collection plan. The accuracy of data analysis can be further improved through kernel density estimation and uniformity assessment. Identifying areas with low measurement point density can help engineers or researchers concentrate more resources on measurement and ensure sufficient collection of river section data. By constructing a heat map of measurement point distribution, the spatial distribution of measurement points on the river section can be intuitively displayed. The heat map clearly indicates which areas have dense measurement points and which areas have sparse measurement points, thereby helping users quickly identify hot spots and blank areas for data collection. The heat map can provide intuitive data support for surveyors or data analysts, enabling them to make more appropriate decisions based on visualization, such as in which areas to increase the density of measurement points and in which areas to reduce the arrangement of measurement points. Through the construction of the heat map, the spatial characteristics of the river section can be further analyzed. For example, areas with large terrain fluctuations usually require more measurement point data, while relatively flat areas can appropriately reduce the number of measurement points. The spatial distribution information provided by the heat map can better support subsequent modeling and analysis. By setting a preset measurement point density threshold, areas with high measurement point density can be accurately identified. These areas may be areas with large terrain changes and fast flow rates in the river section, or areas that need special attention. Screening high-density measurement points helps ensure that data in these key areas is fully collected. By screening out high-density measurement point data, data redundancy can be avoided and unnecessary data storage and computing burdens can be reduced. The focus on high-density measurement points makes the data more representative, improves the accuracy of the data, and provides a more reliable basis for subsequent analysis and modeling. By screening high-density measurement points, it can be ensured that data collection meets engineering needs, especially in river section areas that require detailed analysis, ensuring that the density and accuracy of the collected data meet the design requirements.The screened high-density measurement point data can be used as the key input for subsequent river modeling, flow simulation, and water flow prediction. High-density data can provide more detailed cross-sectional features and support more accurate model training and predictive analysis.

[0051] As an embodiment of the present invention, refer to Figure 3 As shown, Figure 1 Detailed step flow diagram of step S2 in the embodiment of the present invention, step S2 includes the following steps:

[0052] Step S21: Identifying invalid cross-section measuring points based on the slope change of adjacent measuring points on the high-density measuring point data, thereby obtaining preliminary over-dense measuring point data;

[0053] The embodiment of the present invention analyzes the slope change of adjacent measuring points for high-density measuring point data. The specific operation is to calculate the slope change for each pair of adjacent measuring points. If the slope change of a certain section of adjacent measuring points is too large (such as greater than the preset threshold value of 0.05), it is considered that there are invalid section measuring points in the area. Through this method, it is possible to identify the overly dense or meaningless parts between the measuring points, and then eliminate the invalid measuring points to obtain preliminary over-dense measuring point data. This step is applicable to the situation where the measuring points in the river section data are unevenly distributed and there are over-dense measuring points, and helps optimize the selection of measuring points.

[0054] Step S22: identifying the deep points and turning points of the cross section according to the high-density measurement point data, and using the deep points and turning points as the reserved cross section measurement points, thereby obtaining the reserved cross section measurement point data;

[0055] The embodiment of the present invention adopts a method for identifying deep and turning points of a section based on elevation changes according to the high-density measuring point data screened out in step S21. The specific operation is to first calculate the elevation difference between the measuring points, sort the measuring points according to the elevation change rate, and select the position with the largest elevation change as the deep and turning points, usually in the bends of the river, deep water areas and other important positions. In order to ensure accurate identification, a threshold judgment method is adopted. When the elevation change rate of the measuring point exceeds the set threshold (such as 0.1m / m), it is determined to be a turning point or a deep point. All identified deep and turning points will be used as retained measuring point data, and the remaining measuring points that do not meet the conditions will be excluded. This step is particularly suitable for complex river sections, and it is necessary to ensure that key geographical feature points are retained, so that there are important reference data in subsequent modeling and analysis.

[0056] Step S23: performing measurement point thinning processing on the high-density measurement point data according to the initial over-dense measurement point data and the reserved cross-section measurement point data, thereby obtaining the initial reserved measurement point data;

[0057] The embodiment of the present invention processes the high-density measurement point data through a measurement point thinning algorithm based on the preliminary over-dense measurement point data obtained in step S21 and the reserved section measurement point data obtained in step S22. The specific operation is to selectively remove some redundant measurement points based on the spatial distribution and elevation changes of adjacent measurement points. The specific steps are to calculate the spatial distance between each measurement point and the surrounding reserved measurement points, and set a distance threshold (such as 50m). If the distance between a certain measurement point and the nearest reserved measurement point is less than the threshold, the measurement point is retained; if the distance between the measurement point and the nearest reserved measurement point exceeds the threshold, the measurement point is removed. In this way, redundant data is removed to obtain the initial reserved measurement point data. This processing step is suitable for scenarios where high-density measurement point data needs to be streamlined, especially when resources are limited, while retaining sufficient information while reasonably controlling the number of measurement points.

[0058] Step S24: extracting the topological structure and geometric continuity of the river section from the initially retained measuring point data, thereby obtaining the geometric topological feature data of the section;

[0059] The embodiment of the present invention extracts the topological structure and geometric continuity of the river section based on the initial retained measuring point data. The specific operation is to calculate the geometric characteristics of the section line, such as the curvature and flatness of the section, by analyzing the adjacent relationship of each retained measuring point, and use an algorithm (such as Delaunay triangulation) to identify the spatial connection relationship between the measuring points. These geometric topological features reflect the morphological characteristics of the river section, including the regularity and change trend of the measuring point distribution. Through this method, the geometric continuity of the river section can be effectively captured, thereby providing support for subsequent measuring point screening and ensuring that the measuring point data can accurately represent the topographic characteristics of the river.

[0060] Step S25: evaluating information loss by calculating the similarity index of the cross-section morphology before and after thinning according to the cross-section geometric topological feature data, thereby obtaining thinning effectiveness data;

[0061] The embodiment of the present invention uses the calculation of the cross-section morphology similarity index to evaluate the information loss before and after thinning. The specific operation is to calculate the changes in the cross-section morphology of the river channel before and after thinning, and use the morphological similarity index (such as Hausdorff distance, Frechet distance, etc.) for comparative analysis. This method will evaluate whether the retained measuring points can better maintain the morphology of the original section during the thinning process. If the similarity index is lower than the set threshold (such as 0.85), it is considered that the thinning operation may cause a large information loss; if the similarity index is high, it means that the thinning operation is effective. This step is particularly suitable for evaluating the impact of measuring point thinning on the representation capability of the river channel section, to ensure that the accuracy of the data after processing is not greatly affected.

[0062] Step S26: performing secondary screening of the reserved measuring points on the high-density measuring point data and the initial reserved measuring point data according to the thinning effectiveness data, thereby obtaining a reserved measuring point data set.

[0063] The embodiment of the present invention performs secondary screening on the high-density measuring point data and the initially retained measuring point data according to the thinning effectiveness data obtained in step S25. The specific operation is to determine which measuring points have less impact on the final result by calculating the thinned section similarity index, and further remove unnecessary measuring points. The screening process sorts the contribution of each measuring point to the entire river section morphology, and selects to retain the measuring points that are most representative of the river morphology. In this way, the final retained measuring point data set is obtained, in which the retained measuring point data can effectively reduce redundant information on the basis of ensuring data integrity. This step is suitable for application scenarios that require further refinement of measuring point selection to ensure that the final measuring point data set can not only accurately reflect the river morphology, but also improve data processing efficiency.

[0064] The present invention identifies invalid cross-section measuring points by slope changes, and can eliminate those measuring points with redundant measurements. For example, when the slope change of two adjacent measuring points is very small (i.e., the change between the two points is almost zero), it means that the river morphology described by these measuring points has high repetition and no new information, so they are identified as invalid measuring points. In this way, data redundancy can be reduced and the efficiency of subsequent analysis can be improved. The elimination of valid measuring points ensures that the data only contains necessary measuring points, reduces the interference of noise, and makes subsequent analysis more accurate. Reducing invalid data can reduce the storage and computing burden, optimize the management and processing efficiency of data, especially in the case of high-density data sets. Deep-sea points and turning points are the most representative measurement positions on the river section. Retaining these measuring points can accurately describe the morphological characteristics of the river and ensure that key features are not missed during data processing. By retaining deep-sea points and turning points, the geometric structure of the river section can be better reflected, which helps in subsequent modeling and analysis, and improves the application value of data. In particular, in projects such as flow simulation and flood warning, deep-sea points and turning points play a key role in analysis. By retaining these important points, over-simplification of cross-section information is avoided, thereby ensuring the accuracy of subsequent data analysis and cross-section modeling. By thinning out the combination of the initial over-dense measurement point data and the retained cross-section measurement point data, redundant measurement points can be effectively removed while ensuring the retention of important data. Thinning not only reduces the amount of data, but also retains the key structural information of the river section. The reduction in the amount of measurement point data after thinning helps to make the subsequent data processing and calculation process more efficient. Especially in large-scale river channel measurement, reducing redundant data can significantly improve the calculation speed. By retaining representative measurement points and removing redundant measurement points, the accurate presentation of the river section morphology in data processing is ensured. By analyzing the topological structure and geometric continuity of the river section, the key features of the river section geometry can be extracted from the measurement point data. For example, the curvature, straightness, and concave-convexity of the river can be identified through the topological structure and geometric continuity, which is of great significance for river modeling and flow state analysis. By extracting geometric continuity, it can be ensured that the geometric relationship between each measurement point in the data is accurately maintained. This is crucial for building a smooth and natural river section model to avoid model errors caused by discontinuous or inconsistent data. The extracted geometric topological features can provide the necessary spatial and geometric information for subsequent river morphology analysis, flow velocity simulation, sediment transport analysis, etc., to help make more refined model construction and engineering decisions. By calculating the similarity index of the cross-sectional morphology before and after thinning, the impact of the thinning process on the cross-sectional morphology can be quantified to determine whether there is information loss. Ensuring that important river channel features are not lost during the thinning process is a key step in verifying the quality of data processing. This evaluation step helps to confirm that the thinned data can still represent the actual situation of the river section. If the similarity index is high, it means that the thinning operation is effective; if the difference is large, it means that key information may be lost during the thinning process and the thinning strategy needs to be adjusted.Through information loss assessment, the quality of the data processing process can be effectively controlled to ensure that the final retained measurement point data meets the engineering requirements in terms of accuracy. Secondary screening based on rarefaction validity data can further optimize the data set of the retained measurement points. By evaluating which measurement points are most accurate in describing the river morphology, those measurement points with less impact on the overall section can be effectively removed, and the most representative measurement point data can be retained. Through secondary screening, it is ensured that the remaining retained measurement points can accurately reflect the geometric characteristics of the river section and provide high-quality input data for subsequent modeling and analysis. This can significantly improve the accuracy and reliability of the river section model. Through further screening, the retained data set will be more streamlined, which will help improve computing efficiency and reduce storage requirements while ensuring the validity and representativeness of the data. The final retained measurement point data set has high accuracy and adaptability, and can be flexibly used according to different engineering needs, such as flow state analysis, river sediment prediction and other application scenarios.

[0065] Preferably, step S21 includes the following steps:

[0066] The slope change calculation between adjacent measuring points based on the three-point method is performed on the high-density measuring point data, and the cross-section measuring points with a slope change rate lower than a preset threshold are grouped as preliminary over-dense measuring point data, wherein the slope change calculation between adjacent measuring points based on the three-point method is specifically to calculate the initial slope between the first measuring point and the second measuring point, calculate the change slope between the second measuring point and the third measuring point, and then calculate the slope change rate between adjacent measuring points based on the initial slope and the change slope.

[0067] In the embodiment of the present invention, three adjacent measuring points are selected, which are recorded as measuring point 1, measuring point 2 and measuring point 3, and their coordinates in space are respectively , and ,in , and is the elevation value of the measuring point. Then, the three-point method is used to calculate the initial slope between the first measuring point and the second measuring point. The slope formula is: in, is the horizontal distance between measuring point 1 and measuring point 2, and the calculation formula is: Next, calculate the change slope between the second measuring point and the third measuring point. The calculation formula is: in, is the horizontal distance between measuring point 2 and measuring point 3, and the calculation formula is: Finally, according to the slope of the first segment and the second slope , calculate the slope change rate between adjacent measuring points, the change rate calculation formula is: If the calculated slope change rate is lower than the preset threshold (e.g. 0.05), it is considered that the distribution of measuring points in the area is relatively flat and does not need to be over-refined, so the measuring point is considered to be a preliminary over-dense measuring point. This method can be used to identify which areas of the measuring point in the measuring area are too dense due to the lack of obvious slope changes, thus providing a basis for subsequent thinning processing. This method is particularly suitable for situations where the density of measuring points in river section data is high and the slope change is small, ensuring the rationality of the data and reducing the processing of redundant information.

[0068] The present invention can identify those measuring points that do not significantly reflect the change of river channel morphology by calculating the slope change rate. Specifically, when the slope change rate between two adjacent measuring points is lower than the preset threshold, it means that the two measuring points contribute less to the description of the change of river channel morphology and the data is redundant. Therefore, these measuring points can be identified as "overcrowded measuring points" and eliminated or adjusted. This screening process helps to reduce the interference of redundant data, optimize the data set, and improve the quality and effectiveness of the measurement data. By evaluating the slope change between adjacent measuring points, the method can effectively identify those measuring points that provide repeated or redundant information. For example, in areas where the river section is relatively flat and the morphological changes are not large, the slope change rate between the measuring points is usually low, indicating that these measuring points contribute less to the description of the change of river channel morphology. At this time, by removing these overcrowded measuring points from the data, the measurement data in the key area can be more concentrated to avoid the interference of useless information on the model. This not only improves the data processing efficiency, but also enhances the accurate description of the river section morphology. In high-density measurement point data, by applying the three-point method to calculate the slope change, repeated information can be systematically identified and eliminated, thereby significantly reducing the amount of data. The streamlined data set is more compact, which can improve the computational efficiency in subsequent analysis, especially when processing large-scale data, it can reduce storage requirements and computing burden. Data simplification can accelerate subsequent model construction and simulation analysis, shorten analysis time, and improve efficiency. Through the careful calculation of slope changes, the key changes in river morphology can be better captured. Compared with directly relying on the number of measurement points, the analysis based on slope changes can more accurately identify important morphological features and ensure that the measurement point data retained in the end has high representativeness and accuracy. This method avoids collecting a large amount of redundant data in flat or repeated areas, so that the final data set is more in line with the actual changes in river morphology and improves the application effect of the data. The slope change rate method calculated based on the three-point method can adapt to the morphological change requirements of different types of rivers. For example, for river sections with relatively gentle cross-sectional morphological changes, the density of measurement points may be relatively high, while for areas with more drastic cross-sectional changes, more details need to be retained. In this case, by setting a reasonable slope change rate threshold, the density of measuring points can be flexibly adjusted to ensure that the data can accurately reflect the actual situation of the river in both space and form. By extracting the initial over-dense measuring point data, the basis for subsequent data processing and analysis is formed. Over-dense measuring point data will become an important source of data for screening and optimization in the subsequent thinning, modeling, data fusion and other processes. By accurately identifying redundant data, the accuracy and efficiency of subsequent analysis can be ensured, and the complexity of data processing can be reduced.

[0069] Preferably, step S22 includes the following steps:

[0070] Step S221: Calculate the absolute difference of the elevations of adjacent measuring points for the high-density measuring point data. When the elevation change rate exceeds a preset threshold and presents an extreme value feature, it is determined to be a deep point, thereby obtaining the cross-section deep point data;

[0071] The embodiment of the present invention calculates the absolute difference in elevation of adjacent measuring points for high-density measuring point data. The specific steps are: for each pair of adjacent measuring points, calculate the elevation difference ,in and They are the elevation values ​​of measuring point 1 and measuring point 2 respectively. Next, calculate the rate of change of the difference relative to the horizontal distance. The rate of change formula is: in is the horizontal distance between adjacent measuring points. If the elevation change rate of a certain measuring point exceeds the preset threshold (for example, 0.2 m / m) and presents extreme value characteristics (i.e., the elevation difference increases significantly), the measuring point is determined to be a deep fen point, indicating that the water depth at this location is deep and it is one of the key areas of the river. Finally, all the measuring points determined to be deep fen points are aggregated to form the cross-sectional deep fen point data.

[0072] Step S222: performing continuous slope change calculation between the measuring points on the high-density measuring point data, thereby establishing a slope change sequence between the measuring points;

[0073] The embodiment of the present invention calculates the continuous slope change between the measuring points to provide data support for the subsequent turning point identification. The specific method is: first, select each pair of adjacent measuring points and , calculate the slope between two points, the calculation formula is: in, and is the elevation of the measuring point, is the horizontal distance between the two measuring points. Then, the slope changes of all adjacent measuring points are calculated to obtain a continuous slope change sequence. This sequence reflects the slope changes of each measuring point in the river section and provides basic data for subsequent turning point identification.

[0074] Step S223: performing curvature change analysis on the slope change sequence, and determining that the slope change exceeds a preset threshold and exhibits an inflection point feature as a turning point, thereby obtaining cross-section turning point data, wherein the inflection point feature is specifically that the curvature reaches a local extreme value at the measuring point;

[0075] The embodiment of the present invention identifies the turning point by performing curvature change analysis on the slope change sequence. The specific operation is as follows: first, the curvature of the slope change sequence is calculated, and the curvature calculation formula is: in, and The measuring points arrive and arrive The slope of the curve is calculated by calculating the curvature value, and determining that the curvature reaches a local extreme value at certain measuring points (i.e., the local curvature changes greatly and presents an inflection point feature), so that these points are determined to be turning points. If the curvature change exceeds a preset threshold (e.g., 0.3) and presents obvious extreme value features, these measuring points are determined to be turning points, forming the cross-section turning point data.

[0076] Step S224: merging the cross-section deep point data and the cross-section turning point data into the retained cross-section measurement point data.

[0077] The embodiment of the present invention merges the cross-section deep-water point data and the cross-section turning point data to obtain the retained cross-section measuring point data. The specific steps are: merging the deep-water point data identified in step S221 with the turning point data identified in step S223, and forming a retained cross-section measuring point data set after deduplication. These retained measuring point data have important geometric and hydrological characteristics, are indispensable measuring point data in river section analysis, and play a key role in subsequent modeling, analysis and optimization. Finally, the merged retained cross-section measuring point data is the required retained measuring point data set.

[0078] The present invention can identify the parts of the cross section where the elevation changes abnormally drastically by calculating the absolute difference in elevation. These parts usually represent the deep areas of the river channel, that is, the places where the riverbed is lower. Deep points usually have an important influence on the flow characteristics and riverbed morphology of the river channel, so they are key features that cannot be ignored in the analysis of river channel cross sections. By setting a preset threshold and detecting points where the elevation change rate exceeds the threshold and presents extreme value characteristics, these deep points can be accurately found without misjudging the changes in other areas. This method makes the extraction of deep points more reliable and accurate, which is helpful for the subsequent analysis of hydraulic characteristics such as river channel depth and flow rate. Through the calculation of slope change and the analysis of curvature change, the turning point in the cross section, that is, the place where the river channel morphology changes significantly, can be accurately identified. Turning points often correspond to the turning point or drastic change of the river channel morphology, which is of great significance for subsequent river channel reconstruction, sedimentation analysis, flow rate change, etc. Curvature analysis can identify the local extreme points of slope change, which are the turning points of the cross section and the signs of drastic changes in the river channel morphology. By accurately identifying these turning points, we can better understand the geometric characteristics of the river channel and provide key information for subsequent hydrological and hydraulic models. In the process of identifying deep fens and turning points, by analyzing the elevation change rate and slope change of the measuring points, we can eliminate the false changes caused by measurement errors or data noise, ensuring that the final retained cross-section measuring point data has high accuracy and reliability. Accurate identification of deep fens and turning points can effectively filter out redundant measuring point data and ensure that subsequent analysis and modeling are based on measuring point data that truly reflect the changes in river channel morphology. Deep fens and turning points are important features in the changes in river channel morphology. Retaining these points can make the river channel cross-section measurement data more representative and cover the important changes in the river channel morphology. Deep fens reflect the low-lying areas of the riverbed and are key areas for water flow. Retaining these points is of great significance for describing water flow patterns, sediment distribution, etc. Turning points represent the turning or morphological change areas of the river channel and are key nodes that cannot be ignored in the flow state, flow velocity and hydrological model of the river channel. By retaining turning points, a more accurate data basis can be provided for flow pattern analysis and river management. By identifying and retaining deep-sea points and turning points, the originally redundant high-density measurement point data can be effectively streamlined, avoiding the interference of too many irrelevant measurement points, and reducing the complexity of subsequent analysis and modeling. Instead of retaining all measurement points, accurately identifying and retaining key feature points helps optimize the data, making the data set more compact and effective, and greatly improving the computational efficiency. Such processed data can not only reduce storage and processing overhead, but also improve the accuracy and efficiency of subsequent analysis. This data optimization is particularly important in the case of large-scale river channel measurement and long-term monitoring. The data set after screening and retaining deep-sea points and turning points can provide more reliable and accurate input for subsequent river section analysis, river dynamic simulation, hydraulic calculations, etc.These retained measuring points are often the "key nodes" of river morphology changes. They are of great reference value for river reconstruction, flow rate prediction, sedimentation analysis, etc. In addition, retaining key section measuring point data helps to better identify the evolution trend of river morphology and water flow patterns, support long-term monitoring and governance effect evaluation, and provide strong support for river management decisions. Combining the deep-sea point and turning point data can be used to construct a visual model of the river section, helping engineers and decision makers to more intuitively understand the changes in river morphology and their impacts. This kind of visualization support can improve the rationality of engineering design, governance plan formulation and resource allocation. By displaying these key measuring point data through intuitive heat maps or 3D models, river governance and management measures can be optimized to ensure focus and resource investment in different areas.

[0079] Preferably, step S3 comprises the following steps:

[0080] Step S31: calculating the vertical distance between each measuring point in the retained measuring point data set and the preset section line, thereby generating a deviation measuring point data set;

[0081] The embodiment of the present invention needs to convert the preset river section line into a mathematical expression. For example, if the section line is a straight line, the equation can be expressed as ,in , and is a known constant. Usually, two points on the cross section can be measured to obtain and , and then use these points to fit the equation of the straight line. The slope of the straight line can be calculated by and the intercept Get the equation , which is further converted to For each measuring point , calculate its vertical distance to the section line The calculation formula is: in and are the coordinates of the measuring point, , and is the coefficient of the cross-section equation. This formula is derived from the formula for the perpendicular distance from a point to a line. It represents the distance from each measuring point to the section line and is a positive value. The deviation values ​​of all measuring points can be summarized into a deviation measuring point data set for subsequent analysis.

[0082] Step S32: classifying the deviation points in the deviation measurement point data set based on the deviation direction, thereby generating classified deviation measurement point data;

[0083] For each measuring point, The vertical distance calculated from it The specific operation is as follows: First, determine the position of the measuring point. If the deviation value of the measuring point is positive, the measuring point is on one side of the section line; if If it is negative, the measuring point is located on the other side. The change of this sign indicates the deviation direction of the measuring point relative to the section line. According to the deviation direction, the deviation measuring point data set is divided into two categories: positive deviation measuring point data set and negative deviation measuring point data set. Positive deviation measuring point means that the measuring point is located on one side of the section line, and negative deviation measuring point is located on the other side. This classification helps to adopt different correction strategies for different types of errors in subsequent analysis, especially when the measurement error is large or the data is abnormal, it can effectively distinguish the error source and make corresponding corrections.

[0084] Step S33: analyzing the error sources of the classified deviation measuring point data based on the terrain influence and the measurement error, thereby obtaining a deviation measuring point error feature data set;

[0085] The embodiment of the present invention first needs to analyze the cause of the deviation in combination with the spatial distribution of the measuring points and the topographic features of the river (such as bends, slopes, riverbed changes, etc.). For example, if the deviation points appear in the bends of the river or in areas where the slope changes sharply, it may be a deviation caused by the influence of the terrain. In these areas, the deviation of the measuring points may be due to the local influence of the terrain rather than the measurement error. Secondly, it is necessary to consider the error factors in the measurement process, such as the accuracy of the measuring instrument, the operating errors of the operator, and environmental factors (such as temperature changes, equipment aging, etc.). If the deviation of the measuring points is relatively regular and exists in a large range, it may be caused by the accumulation of measurement errors. At this time, these errors can be identified and corrected by analyzing the regularity of the errors, the accuracy requirements of the measuring tools, etc. Through the above analysis, the error source of each deviation point is determined and summarized as an error feature data set. This data set contains information such as the error source, terrain influence, measurement error, etc. of each deviation measuring point, providing data support for subsequent coordinate reconstruction and measurement correction.

[0086] Step S34: reconstructing the coordinates of the deviated measuring points according to the deviation measuring point error characteristic data set, and correcting the overall section line position of the updated measuring point coordinates, thereby obtaining section line equation correction data;

[0087] The embodiment of the present invention formulates an error correction scheme for each deviated point based on the error source analysis of the previous step. If the deviation is caused by measurement error, the coordinates can be corrected by smoothing (such as neighborhood weighted averaging); if the deviation is caused by terrain factors, it may be necessary to correct the terrain model and repair the coordinates in combination with surrounding measuring points. According to the error correction method, the coordinates of the deviated measuring points are reconstructed. For example, the least squares method or other smoothing methods are used to adjust the coordinates of the measuring points so that they are more in line with the distribution trend of the surrounding points. Specifically, for measuring points with large errors, the corrected coordinates can be inferred from the adjacent valid measuring points. For all measuring points, a unified correction is performed based on the reconstructed coordinates to adjust the overall position of the measuring points to ensure that all measuring points are closer to the preset cross-section line. The purpose of this step is to improve measurement accuracy and reduce the impact of errors.

[0088] Step S35: performing a supplementary calibration of the elevation and position consistency of the densely populated measurement point area near the section line equation correction data, thereby obtaining calibration consistency measurement point data;

[0089] The embodiment of the present invention first selects a dense area of ​​measuring points near the cross-section line equation correction data for further calibration. Dense areas refer to areas where the distance between measuring points is small and the data is more concentrated. These areas are more likely to reflect systematic errors or inconsistencies in measurement. Analyze the elevation data in these dense areas to check whether there are large elevation fluctuations or deviations. If large changes are found, it means that there may be measurement errors or data inconsistencies in these areas. By analyzing the trends of neighboring measuring points, elevation calibration is performed to ensure the consistency of elevation data. In addition to elevation correction, it is also necessary to check the consistency of the spatial position of the measuring points. If there is an obvious deviation in the position of the measuring points in the dense area, the position of these measuring points can be corrected to keep them consistent with the distribution of the surrounding measuring points.

[0090] Step S36: converting the calibration consistency measurement point data into a unified format and outputting it as a standardized data file, thereby generating calibration measurement point data.

[0091] The embodiment of the present invention converts the calibration consistency point data into a standardized data format according to the requirements of the application scenario. For example, it can be converted into CSV format to facilitate subsequent data processing and analysis. Other possible standard formats include GeoJSON, Shapefile, etc. The specific choice depends on the subsequent usage requirements. Export the converted data as a file, and ensure that the data file contains the necessary fields (such as measurement point number, coordinates, elevation, etc.) for subsequent analysis, modeling or engineering design.

[0092] The present invention can accurately identify the measuring points that deviate from the cross-section line by calculating the vertical distance between each measuring point and the preset cross-section line. Due to the measurement error or terrain influence of the river section, the measuring points often have a certain deviation. This step can effectively mark these deviated points as the focus of subsequent correction. By calculating the vertical distance and generating a deviation measuring point data set, it is possible to accurately determine whether the position of each measuring point meets the requirements of the preset cross-section line, thereby providing a basis for subsequent error analysis and correction. By classifying the deviated points, it is possible to further understand the impact of different types of deviations on the measurement data. According to the different deviation directions, it may be due to measurement method problems, equipment errors or the influence of terrain undulations. Classifying the deviation points according to directions helps to identify the laws and potential causes of deviations. Through classification processing, different types of deviations can be corrected in a targeted manner, which helps to optimize the accuracy of the correction and ensure that no potential problems are missed. By analyzing the error sources of the deviated measuring points, considering the terrain influence and measurement errors, the specific factors that cause the deviation can be identified. River section measurement is often affected by multiple factors such as terrain undulations, measurement tool accuracy and operating errors. By comprehensively analyzing these sources of error, we can more clearly understand the cause of the deviation and carry out targeted processing in subsequent corrections. The error feature data set not only helps to find the specific cause of the deviation, but also provides a theoretical basis for coordinate reconstruction and cross-section line correction, improving the reliability and accuracy of the correction results. The error feature data set is used to reconstruct the coordinates of the deviated points. By updating the coordinates of the measuring points and correcting the position of the cross-section line, the influence of the deviated measuring points on the overall cross-section line can be eliminated, thereby accurately restoring the true shape of the river section. Through coordinate reconstruction and cross-section line correction, not only the influence of measurement errors is eliminated, but also the position of the cross-section line is ensured to be more consistent with the actual river morphology. This step can improve the accuracy of the river morphology model and provide more reliable data support for subsequent hydrological and hydraulic analysis. After the cross-section line is corrected, the elevation and position consistency of the nearby dense measuring point areas are supplemented and calibrated to ensure that the measuring points in these areas are highly consistent, further reducing the deviation caused by measurement errors or terrain undulations. This process ensures that the corrected measuring point data is smoother and continuous, and enhances the reliability and accuracy of the river section data. This additional calibration step is particularly important for river cross-section measurements in large-scale or complex terrain areas, and can further optimize the overall data quality. By converting the calibration consistency point data into a unified format and outputting it as a standardized data file, the corrected point data can be integrated into a format that conforms to a unified standard, which is convenient for subsequent data analysis, processing, storage and sharing. Standardized data files can be compatible with other systems or models, and are also convenient for data exchange between team members. This standardized output not only enhances data interoperability, but also facilitates long-term data archiving and analysis, ensuring the sustainable use of data in the future.Through correction and calibration, the final generated calibration point data has high accuracy and reduces the interference of invalid data. The standardized data provides a clearer and more consistent data source for subsequent analysis, improving the efficiency and accuracy of subsequent analysis and model building. Standardized and high-quality point data help ensure the efficiency and reliability of related analysis in projects such as engineering design, river management, and watershed management. The corrected and calibrated data provides a solid data foundation for subsequent river morphology analysis, flow prediction, river management, etc. Through high-precision and standardized data, the flow state changes and sediment distribution of the river can be simulated more accurately, providing strong data support for decision makers in river management, ecological protection and infrastructure construction. The high accuracy and consistency of the data ensure the reliability of subsequent research and application, making the decision more scientific and reasonable.

[0093] Preferably, step S31 includes the following steps:

[0094] Step S311: converting the preset cross-section line into a mathematical expression, and calculating the vertical projection point position of each measuring point in the retained measuring point data set to the cross-section line, thereby obtaining the projection point position data;

[0095] The embodiment of the present invention assumes that the preset river section line is a straight line. In order to convert the section line into a mathematical expression, a standard linear equation form can be used: ,in , and is a known constant, and is the coordinate of the point. The cross-section equation can be obtained based on the coordinates of two points in the actual scene or by fitting the existing measurement point data. Specifically: Assume there are two endpoints and , then the coefficients of the cross-section equation can be obtained by linear algebra. First, calculate the slope of the line segment , then according to Get the intercept , so the equation of the cross section is If the equation does not meet the standard Once we have the equation of the cross section, for each measuring point , it is necessary to calculate its vertical projection point to the section line The vertical projection can be calculated using the following formula: These two formulas are derived from the normal direction of the original measuring point to the cross section line by the least square method. By calculating these formulas, the position of the vertical projection point of each measuring point can be obtained.

[0096] Step S312: Calculate the Euclidean distance from the measuring point to its corresponding vertical projection point according to the projection point position data, thereby generating a measuring point deviation distance matrix;

[0097] The embodiment of the present invention is based on the projection point position data, and needs to calculate each measuring point to its corresponding projection point The Euclidean distance between them. The Euclidean distance is calculated as follows: For each measured point and its corresponding projection point, the Euclidean distance The calculation formula is: in, and It is the measuring point The coordinates of and is the corresponding projection point The coordinates of . This formula can be used to calculate the straight-line distance from each measuring point to its projection point. Apply the above Euclidean distance calculation to all measuring points in the retained measuring point data set to obtain a measuring point deviation distance matrix. Each item in the matrix Indicates measuring point The deviation distance from its projection point. This matrix is ​​used for subsequent deviation measurement point screening and error analysis.

[0098] Step S313: Mark the measurement points whose deviation distance exceeds the preset threshold as deviation measurement points, so as to obtain a deviation measurement point data set.

[0099] The embodiment of the present invention needs to be based on a preset deviation distance threshold , mark the measurement points whose deviation distance exceeds the threshold as deviation measurement points. Set deviation threshold: Set a reasonable threshold , the threshold can be set according to the measurement accuracy requirements in the actual scene. For example, if the threshold is assumed to be 0.5 meters, then for each measuring point, if its vertical deviation distance to the section line is greater than 0.5 meters, the measuring point is considered to be a deviation measuring point. Traverse each item in the measuring point deviation distance matrix ,if , then the measuring point Mark as a deviation point. This process can be completed by simple conditional judgment. Finally, a deviation point data set is obtained, which contains all the measurement points that deviate from the section line by more than the preset threshold. For example, in a river channel measurement project, assuming that the measurement accuracy is required to be within 1 meter, the preset deviation distance threshold It can be set to 0.5 meters, which means that if the deviation distance of a measuring point is greater than 0.5 meters, the measuring point is regarded as an abnormal point and needs further analysis and correction.

[0100] The present invention can abstract the section line in the geographic space into a mathematical object by converting the preset section line into a mathematical expression, which is convenient for subsequent calculation and processing. This step ensures that the section line has a clear definition in the numerical model, so that the spatial relationship between the measuring point and the section line can be accurately quantified. The precise mathematical expression provides a theoretical basis for the subsequent measurement point position calculation and deviation analysis, avoids the errors that may occur in manual operation, and ensures the consistency and accuracy of the analysis results. By calculating the vertical projection point position of the measuring point to the section line (step S311), it can be ensured that the spatial relationship between each measuring point and the section line is clear. The position of this vertical projection point directly reflects the distance between the measuring point and the section line, which is convenient for subsequent deviation analysis. The vertical projection point provides a clear reference position for each measuring point, which can effectively evaluate the deviation of the measuring point relative to the section line, avoids complex spatial geometric calculations, and simplifies the subsequent processing flow. By calculating the Euclidean distance from the measuring point to its corresponding vertical projection point, the degree of deviation between the measuring point and the section line can be quantified. Euclidean distance is a standard way to measure the straight-line distance between two points in space. It is applicable to the distance calculation between the measuring point and the projection point in this step, and provides an accurate deviation value. Euclidean distance provides a standardized distance measurement, making the deviation analysis scientific and consistent. By generating a measuring point deviation distance matrix, the distance distribution between all measuring points and the cross-section line can be intuitively viewed, providing a basis for subsequent screening of deviated measuring points. By setting a threshold, the measuring points whose deviation distance exceeds the threshold are marked as deviated measuring points. This step can quickly filter out abnormal measuring points and focus on processing potential measurement errors or inaccurately positioned measuring points. By marking the deviated measuring points, the subsequent steps can focus on error correction or further analysis of these measuring points. The threshold screening mechanism can effectively identify abnormal measuring points caused by errors or other factors in the measurement process. By marking the deviated measuring points, the abnormal data can be quickly distinguished from the valid data, reducing unnecessary interference and ensuring the efficiency and accuracy of subsequent data processing. By screening out the deviated measuring point data set, the measuring points that may be affected by the error can be eliminated or focused on calibration, thereby improving the accuracy and reliability of the measurement data. The identification and processing of deviated measurement points helps to optimize the subsequent data correction process and ensure that the river section data can better fit the actual terrain. By marking and subsequently processing the deviated measurement points, it is possible to ensure that the measurement data is more accurate and avoid inaccurate data caused by deviated measurement points affecting river analysis, modeling and design. The final result is more reliable cross-sectional data that supports more accurate decision-making and planning. As the basis for subsequent correction and optimization, the deviated measurement point data set can help identify measurement points that do not meet the requirements and provide targets for further data cleaning and correction. By marking these deviated measurement points, further error analysis, coordinate reconstruction, and cross-sectional line position correction can be performed on them.The generation of the deviation measurement point data set provides a clear focus for the subsequent cross-section data correction, which can effectively improve the efficiency and accuracy of subsequent corrections and ensure that the river cross-section data better meets actual needs.

[0101] Preferably, step S34 includes the following steps:

[0102] Step S341: performing vector analysis on the error direction of the deviation measuring point according to the deviation measuring point error feature data set, thereby generating measuring point vector offset data;

[0103] The embodiment of the present invention first uses the deviation measurement point error feature data set to analyze the error direction of each deviation measurement point. The specific operation is: by comparing the actual coordinates of each deviation measurement point with the vertical deviation value of the preset section line, the error direction of the measurement point is obtained. Assume that a certain measurement point The distance from the section line is , and the deviation direction of the measuring point is vertical. Using the definition of error direction, the vector offset of each measuring point can be calculated ,in It is the deviation direction angle of the measuring point. The vector offsets of all measuring points are summarized as the measuring point vector offset data. These data can clearly show the deviation direction and size of each measuring point, and become the basis for subsequent coordinate adjustment and section line correction.

[0104] Step S342: adjusting the deviated measuring point to the cross-section line position along the vertical direction based on the measuring point vector offset data, updating the spatial coordinates of the measuring point, thereby obtaining measuring point coordinate reconstruction data;

[0105] The embodiment of the present invention adjusts the deviated measuring point in the vertical direction according to the measuring point vector offset data obtained in step S341 to ensure that its position is consistent with the cross-section line. The specific operation is: , adjust the position of each measuring point. Assume that the measuring point The vector offset from the section line is ,in For the deviation distance, is the deviation angle. In the vertical direction, adjust the measuring point along the normal direction of the section line and calculate the new measuring point position as ,in (Assuming that the adjustment is mainly carried out in the plane.) In this way, the coordinates of all deviated measuring points will be adjusted along the vertical direction of the section line to obtain new coordinate reconstruction data.

[0106] Step S343: refitting the mathematical expression of the cross-section line according to the measurement point coordinate reconstruction data to generate cross-section line equation refitting data;

[0107] The embodiment of the present invention uses the measured point coordinates obtained in step S342 to reconstruct the data and refit the mathematical expression of the cross section line. The specific operation is as follows: first, use the updated measured point coordinate set , recalculate the equation of the cross section line by the least squares method or other fitting algorithms. Assuming that the fitted cross section line is a straight line, the equation is , the fitting can be achieved through the following steps: construct a least square fitting model between the measuring points and the straight line, calculate the minimum error of all reconstructed measuring point coordinates relative to the fitting straight line, and obtain the best through iterative optimization. , and Parameters. After fitting is completed, a new section line equation is generated , and the equation can better adapt to the corrected coordinates of the measuring points. The refitted data provides an accurate mathematical expression for subsequent error analysis and correction.

[0108] Step S344: Perform residual statistical analysis on the cross-section line equation refitting data, and perform screening based on the error range according to the measurement point coordinates to reconstruct the data, so as to obtain cross-section line equation correction data.

[0109] The embodiment of the present invention performs residual statistical analysis on the cross-section line equation refitting data and screens out unqualified measuring points through the error range. The specific operation is as follows: first, the residual of each measuring point in the new cross-section line equation is calculated, that is, the vertical distance between the measuring point and the fitted cross-section line. For each measuring point , according to the cross-section equation after refitting , calculate the residual error of the measured point: Then, by counting the residual data of all measuring points, the distribution characteristics of the residuals are analyzed, and the measuring points are screened according to the set error range (for example, the residual is set to be no more than 0.1 meters to be qualified). For measuring points that exceed the error range, they can be marked as abnormal points and considered for removal or re-measurement. Finally, the cross-section line equation correction data is obtained, which contains the coordinates of the measuring points that meet the error range after screening and the correction equation of the fitting cross-section line. These data will provide more accurate cross-section line parameters and measuring point locations for subsequent engineering applications.

[0110] The present invention can clearly identify the deviation direction and size of each deviated measuring point by vectorizing the error direction of the deviated measuring point. This process transforms the error characteristics of the measuring point from a simple spatial deviation into more systematic vector data, which is convenient for subsequent correction operations. By vectorizing the error direction, the deviation of each measuring point can be better quantified, providing accurate directional data for subsequent error correction. This process converts complex measurement error information into structured data, which is convenient for efficient processing and correction. According to the vector offset data of the measuring point, the deviated measuring point is adjusted to the cross-section line position in the vertical direction. In this way, the coordinates of the deviated measuring point can be directly corrected to obtain updated measuring point coordinate data. This process is essentially to eliminate the deviation and correct the error by applying a vertical adjustment to the cross-section line position. This process can directly solve the problem of measuring point deviation and ensure that the coordinates of each measuring point are consistent with the cross-section line position. Through coordinate reconstruction, the deviation error is effectively eliminated, ensuring that the measurement data is more accurate and consistent, and reducing the data unreliability caused by errors. Based on the measuring point data after coordinate reconstruction, the cross-section line equation is refitted. This means that the mathematical expression of the river section line is reconstructed through the updated measuring point coordinate data to ensure that the relationship between the shape of the section line and the data points is more accurate. By refitting the section line equation, it can be ensured that the mathematical model better conforms to the actual measurement results. The original section line equation may have an unsatisfactory fitting effect due to errors and deviations. After refitting, the fitting accuracy of the section line equation can be greatly improved, making it closer to the actual river channel shape. By performing residual statistical analysis on the refitted section line equation data, the errors in the section line fitting process can be detected and evaluated. By screening the error range of the reconstructed data of the measuring point coordinates, the section line equation is further refined and optimized, and the bad data points with large residuals are eliminated, and finally more accurate section line equation correction data are obtained. This analysis process not only ensures the high accuracy of the section line equation, but also effectively eliminates abnormal data and measuring points with large errors. Residual statistical analysis can quantify the fitting error and provide important feedback information, which helps to further optimize the fitting quality of the section line and ensure the reliability and practicality of the final data. Through the comprehensive application of the above steps, the final generated section line equation correction data can more realistically reflect the actual terrain changes. The deviation of the measuring points is eliminated through coordinate reconstruction, the cross-section line equation refitting ensures more accurate data fitting, and the residual screening removes data points with large errors. This series of refined operations significantly improves the accuracy of the cross-section data. High-precision cross-section data can provide more reliable basic data for subsequent river flow analysis, sediment distribution simulation and water conservancy project design. Accurate cross-section data is crucial for fields such as river management, flood warning and watershed management. Accurate correction of cross-section data can provide solid data support for decision-making in water conservancy projects, river management, ecological restoration and other projects.The corrected data can be used for accurate river flow prediction, flood simulation, dam design, etc., which is crucial for engineering design and risk assessment. With the support of accurate cross-sectional data, decisions in related fields can be made more scientific and reasonable. Whether it is river regulation, urban flood control, or ecological restoration, optimized decisions can be made based on more accurate data to ensure the effectiveness and safety of the project.

[0111] Preferably, step S4 comprises the following steps:

[0112] Step S41: performing DBSCAN-based measurement point clustering analysis based on the high-density measurement point data to identify blank areas of measurement point distribution, thereby obtaining blank area data;

[0113] The embodiment of the present invention uses the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm to perform cluster analysis on high-density measurement point data. The specific operation is as follows: first, collect all measurement point data and calculate the spatial position of each measurement point. Then, set two important parameters in the DBSCAN algorithm: neighborhood radius and minimum number of neighbors The neighborhood radius is defined as the range of the area of ​​interest around the measuring point, and the minimum number of neighbor points is defined as the minimum number of neighbors for a point to be considered as a core point. Through the DBSCAN algorithm, the measuring points will be divided into several clusters, each cluster represents a high-density measuring point area, and the non-clustered measuring points are regarded as noise points. In the clustering results, the areas without measuring points are identified, that is, blank areas. Based on these blank areas, blank area data is generated to provide a basis for the subsequent analysis of supplementary measuring points.

[0114] Step S42: extracting the boundary of the blank area according to the blank area data using a boundary detection algorithm, and verifying the position and range of the blank area according to the shape and size of the boundary, thereby obtaining the boundary data of the blank area;

[0115] The embodiment of the present invention first uses a boundary detection algorithm to extract the boundary of the blank area. The specific operation is: according to the blank area data obtained in step S41, a common boundary detection algorithm, such as the Canny edge detection algorithm or the Sobel operator, is used to extract the boundary line of the blank area. The boundary detection algorithm determines the outer contour of the blank area based on the spatial distribution of the measuring points. Then, the position and range of the blank area are verified based on the shape and size of the boundary. The verification process includes calculating the closure of the boundary, the regularity of the shape, and the length of the boundary line to ensure that the extracted boundary meets the actual terrain and measurement requirements. The blank area boundary data finally obtained can be used for the subsequent identification of potential supplementary areas and the generation of supplementary measuring points.

[0116] Step S43: Calculate the spatial density of adjacent section measuring points according to the blank area boundary data, identify potential measuring point supplementary areas through density change analysis, and obtain potential measuring point supplementary area data;

[0117] The embodiment of the present invention analyzes the spatial density of the measuring points in the adjacent cross-sectional area based on the boundary data of the blank area. The specific operation is: by performing spatial distribution analysis on the measuring point data in the adjacent cross-sectional area, the local density of each measuring point is calculated. For example, the K nearest neighbor algorithm (KNN) is used to calculate the number of measuring points within a certain radius around each measuring point to obtain the density value of each measuring point. Then, based on these density values, the density change trend is analyzed, and the areas with low density and close to the boundary of the blank area are identified and marked as potential supplementary areas. The purpose of this process is to find potential supplementary areas for measuring points through spatial density analysis, so as to provide a basis for the distribution of subsequent supplementary measuring points.

[0118] Step S44: Analyze the geometric changes of adjacent measuring points on the blank area boundary data, extract the morphological features of the surrounding area, identify the continuity and connection points between the blank area and the adjacent sections, and obtain geometric connection point data;

[0119] The embodiment of the present invention performs geometric morphological change analysis on the measuring points near the boundary of the blank area, and identifies the connection points between the blank area and the adjacent sections. The specific operation is as follows: first, based on the boundary data of the blank area, the geometric morphology of the measuring points around the boundary is analyzed. For example, the geometric features such as curvature, slope and continuity of the measuring points near the boundary of the blank area are calculated, and the geometric relationship between these measuring points and the adjacent sections is identified. By analyzing these geometric features, the continuity and connection points between the blank area and the adjacent sections are determined, and the positions of the measuring points that need to be supplemented between the two are found. Finally, the data of these connection points are extracted to provide important information for the generation of supplementary measuring points and accuracy optimization.

[0120] Step S45: generating supplementary measuring points in the blank area based on the inverse distance weighted method according to the potential measuring point supplementary area data and the geometric morphology connection point data, and performing quality assessment on the generated supplementary measuring points based on spatial distribution characteristics, density distribution and elevation changes, screening high-quality supplementary measuring points, and thus obtaining blank area measuring point data;

[0121] The embodiment of the present invention generates supplementary measuring points using the inverse distance weighted method (IDW) based on the potential measuring point supplementary area data and the geometric connection point data. The specific operation is as follows: first, based on the potential measuring point supplementary area data, select the location where the supplementary measuring points need to be supplemented in the blank area, and use the geometric connection point data as a reference for the generation of supplementary measuring points. Then, the inverse distance weighted method is applied to calculate the elevation and position of the supplementary measuring points based on the spatial distance between the known measuring points and the measuring points to be supplemented. The basic principle of the inverse distance weighted method is that the measuring points with a closer distance have a greater impact on the supplementary measuring points, while the measuring points with a farther distance have a smaller impact. The supplementary measuring points generated by this method will be evaluated for quality based on the spatial distribution characteristics, density distribution and elevation changes, and finally high-quality supplementary measuring point data will be screened out.

[0122] Step S46: Correcting the calibration points that are accurate or have large deviations in the calibration point data according to the blank area measurement point data, thereby obtaining calibration correction measurement point data, wherein the correction processing includes position reconstruction, elevation adjustment and spatial relationship optimization;

[0123] The embodiment of the present invention corrects the calibration point data according to the blank area point data obtained in step S45. The specific operation is: by comparing the position and elevation differences between the calibration point and the supplementary point, the points with obvious deviations in the calibration point are identified. For these points with large deviations, corrections are made by methods such as position reconstruction, elevation adjustment and spatial relationship optimization. For example, position reconstruction can correct the position of points with large deviations by interpolation methods (such as inverse distance weighted method or Kriging interpolation), and the elevation adjustment is adjusted according to the elevation trend of nearby points. The spatial relationship optimization considers the geometric relationship between the points to ensure that the spatial distribution of the points after correction is more consistent. Through these correction operations, the calibration and correction point data are finally obtained, which provides accurate data support for subsequent further analysis and model optimization.

[0124] Step S47: performing reserved point addition correction according to the calibration correction measuring point data, thereby obtaining the corrected measuring point data.

[0125] The embodiment of the present invention adds corrections to the reserved points based on the calibration and correction of the measuring point data. The specific operation is as follows: first, the difference between the reserved measuring points and the corrected calibration measuring points is analyzed to identify the positions and elevations that may need to be adjusted. Then, the reserved points are finely corrected through the position reconstruction and elevation adjustment methods to ensure that the data of the reserved points is consistent with the corrected measuring point data. Finally, based on these corrected measuring point data, the reserved point data set is updated to obtain the final corrected measuring point data. These corrected measuring points will become the basis for further river channel analysis, modeling and engineering design, ensuring high accuracy and consistency of the data.

[0126] The present invention analyzes high-density measurement point data by applying the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) clustering algorithm, and can effectively identify blank areas in the measurement point distribution. The DBSCAN algorithm can distinguish high-density areas from blank areas according to the density characteristics of the data, avoiding the problem of over-reliance on manually set thresholds that may occur in traditional methods. This process ensures the automatic identification of blank areas, reduces human errors, and provides accurate blank area data for the next step of supplementary measurement points. The boundary detection algorithm is used to extract the boundary of the blank area and verify its position and range. Boundary detection can clearly calibrate the shape and position of the blank area, thereby providing a clear spatial range for subsequent supplementary measurement points. By verifying the shape and size of the boundary, it is avoided that the supplementary measurement points are mistakenly placed in areas or outside the boundaries that do not actually need to be supplemented, thereby improving the accuracy of the supplementary data. By analyzing the spatial density changes of adjacent section measurement points through the blank area boundary data, potential measurement point supplementary areas can be identified. Density change analysis helps to find out which areas have sparse or blank measurement points and accurately determine the areas that need to be supplemented. This step can ensure that the area of ​​the supplementary measuring points is located in the blank area that really needs to be supplemented, thereby avoiding unnecessary duplication of measuring points or over-coverage and improving the coverage integrity of the data. By analyzing the geometric changes of the blank area and its adjacent areas, the continuity and connection points between the blank area and the adjacent sections can be identified, which can ensure the natural connection of the supplementary measuring points in terms of geographical location. This analysis extracts the morphological characteristics of the surrounding area so that the supplementary measuring points can be consistent with the existing section data in terms of morphology. This step ensures that the supplementary measuring point data is morphologically continuous and well connected with the surrounding area and section data, avoiding data discontinuity or confusion caused by improper arrangement of measuring points. Supplementary measuring points are generated based on the inverse distance weighted method (IDW), and the quality of the supplementary measuring points is evaluated through spatial distribution characteristics, density distribution and elevation changes to screen out high-quality supplementary measuring points. The IDW method generates supplementary measuring points based on the distance and weighting coefficient of the known measuring points to ensure the rationality and accuracy of the location of the supplementary measuring points. Through quality evaluation and screening, it is ensured that the supplementary measuring points can effectively fill the blank area and meet the predetermined quality standards. This process not only ensures the spatial consistency of the supplementary data, but also ensures the rationality of the elevation and distribution of the supplementary measuring points, avoiding the introduction of low-quality measuring points. By correcting the calibration measuring point data, reconstructing the position, adjusting the elevation and optimizing the spatial relationship of the calibration points with large deviations, the accuracy of the measuring point data can be significantly improved. This process eliminates errors caused by measurement errors or equipment deviations, ensuring the accuracy of the calibration data. The correction process can effectively eliminate measurement errors, making the data of the calibration measuring points more consistent with the actual terrain, thereby improving the accuracy and reliability of the overall measurement data.Based on the corrected calibration point data, the reserved points are added and corrected to ensure that the final data set has higher consistency and accuracy. This process further optimizes the original point data, so that each point undergoes necessary quality control and adjustment. Through this step, the final corrected point data generated is not only more accurate in space and elevation, but also more in line with the actual situation of the river section as a whole, providing accurate basic data for subsequent analysis and engineering design.

[0127] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0128] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A river channel measurement data preprocessing method based on artificial intelligence, characterized in that: The following steps are involved: Step S1: obtaining river section measuring point data, and evaluating the density of the river section measuring point data based on different scales and changes in section elevation, thereby obtaining high-density measuring point data; Step S2: identifying the cross-section deep points and turning points based on the slope change of adjacent measuring points for the high-density measuring point data, and performing a measuring point thinning process to retain the deep points and turning points, thereby obtaining a retained measuring point data set; Step S3: automatically detecting and calibrating the measuring points exceeding the preset measuring point deviation distance threshold according to the preset section line and the retained measuring point data set, and reconstructing the coordinates and correcting the positions of the measuring points deviating from the preset section line, thereby obtaining calibration measuring point data; Step S4: identifying blank areas of the cross section based on the high-density measurement point data, and extracting distribution features of cross-section measurement points adjacent to the blank areas, thereby obtaining measurement point data of the blank areas; According to the blank area measuring point data, the calibration measuring point data is corrected by adding reserved points, so as to obtain the corrected measuring point data; The method of identifying blank areas in the cross section according to the high-density measurement point data and extracting distribution features of cross section measurement points adjacent to the blank areas includes: Based on the high-density measurement point data, DBSCAN-based measurement point clustering analysis is performed to identify the blank areas of the measurement point distribution, thereby obtaining blank area data; The boundary detection algorithm is used to extract the boundary of the blank area according to the blank area data, and the position and range of the blank area are verified according to the shape and size of the boundary, so as to obtain the boundary data of the blank area; According to the boundary data of the blank area, the spatial density of the adjacent section measuring points is determined, and the potential measuring point supplementary area is identified through density change analysis, thereby obtaining the potential measuring point supplementary area data; The geometric morphological changes of adjacent measuring points are analyzed for the boundary data of the blank area, and the morphological features of the surrounding area are extracted to identify the continuity and connection points between the blank area and the adjacent sections, thereby obtaining the geometric morphological connection point data; Generate supplementary measuring points in blank areas based on the inverse distance weighted method according to the potential measuring point supplementary area data and geometric morphology connection point data, and conduct quality assessment on the generated supplementary measuring points based on spatial distribution characteristics, density distribution and elevation changes, screen high-quality supplementary measuring points, and thus obtain blank area measuring point data; Step S5: Verify the corrected measuring point data based on accuracy and consistency according to the preset engineering requirement data, so as to obtain river channel measurement data preprocessing data.

2. The artificial intelligence-based river survey data preprocessing method according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: obtaining original measuring point data, eliminating invalid coordinates and measuring points beyond the geographical range according to the original measuring point data, and unifying the coordinate system, thereby obtaining river section measuring point data; Step S12: extracting the spatial coordinates of the measuring points on the river section, thereby obtaining the spatial coordinate data of the measuring points; Step S13: Calculate the elevation change rate according to the spatial coordinate data of the measuring points, thereby obtaining a multi-dimensional measuring point feature matrix; Step S14: Evaluate the measurement point density of the multi-dimensional measurement point feature matrix to obtain high-density measurement point data.

3. The artificial intelligence-based river survey data preprocessing method according to claim 2 is characterized in that: Step S14 includes the following steps: Step S141: calculating the Euclidean distances between the measuring points on the multi-dimensional measuring point feature matrix, thereby obtaining a measuring point distance distribution matrix; Step S142: performing local measurement point density calculation based on kernel density estimation on the measurement point distance distribution matrix, and performing measurement point spatial distribution uniformity evaluation, thereby obtaining measurement point density distribution data; Step S143: Identify the measurement point clustering area and sparse area according to the measurement point density distribution data, so as to construct a measurement point distribution heat map; Step S144: screening high-density measuring points based on a preset measuring point density threshold according to the measuring point distribution heat map and the multi-dimensional measuring point feature matrix, thereby obtaining high-density measuring point data.

4. The artificial intelligence-based river survey data preprocessing method according to claim 3 is characterized in that: Step S2 includes the following steps: Step S21: Identifying invalid cross-section measuring points based on the slope change of adjacent measuring points on the high-density measuring point data, thereby obtaining preliminary over-dense measuring point data; Step S22: identifying the deep points and turning points of the cross section according to the high-density measurement point data, and using the deep points and turning points as the reserved cross section measurement points, thereby obtaining the reserved cross section measurement point data; Step S23: performing measurement point thinning processing on the high-density measurement point data according to the initial over-dense measurement point data and the reserved cross-section measurement point data, thereby obtaining the initial reserved measurement point data; Step S24: extracting the topological structure and geometric continuity of the river section from the initially retained measuring point data, thereby obtaining the geometric topological feature data of the section; Step S25: evaluating information loss by calculating the similarity index of the cross-section morphology before and after thinning according to the cross-section geometric topological feature data, thereby obtaining thinning effectiveness data; Step S26: performing secondary screening of the reserved measuring points on the high-density measuring point data and the initial reserved measuring point data according to the thinning effectiveness data, thereby obtaining a reserved measuring point data set.

5. The artificial intelligence-based river survey data preprocessing method according to claim 4 is characterized in that: Step S21 includes the following steps: The slope change calculation between adjacent measuring points based on the three-point method is performed on the high-density measuring point data, and the cross-section measuring points with a slope change rate lower than a preset threshold are grouped as preliminary over-dense measuring point data, wherein the slope change calculation between adjacent measuring points based on the three-point method is specifically to calculate the initial slope between the first measuring point and the second measuring point, calculate the change slope between the second measuring point and the third measuring point, and then calculate the slope change rate between adjacent measuring points based on the initial slope and the change slope.

6. The artificial intelligence-based river survey data preprocessing method according to claim 5, characterized in that: Step S22 includes the following steps: Step S221: Calculate the absolute difference of the elevations of adjacent measuring points for the high-density measuring point data. When the elevation change rate exceeds a preset threshold and presents an extreme value feature, it is determined to be a deep point, thereby obtaining the cross-section deep point data; Step S222: Calculate the continuous slope change between the measuring points for the high-density measuring point data, thereby establishing a slope change sequence between the measuring points; Step S223: performing curvature change analysis on the slope change sequence, and determining that the slope change exceeds a preset threshold and exhibits an inflection point feature as a turning point, thereby obtaining cross-section turning point data, wherein the inflection point feature is specifically that the curvature reaches a local extreme value at the measuring point; Step S224: merging the cross-section deep point data and the cross-section turning point data into the retained cross-section measurement point data.

7. The artificial intelligence-based river survey data preprocessing method according to claim 6 is characterized in that: Step S3 includes the following steps: Step S31: calculating the vertical distance between each measuring point in the retained measuring point data set and the preset section line, thereby generating a deviation measuring point data set; Step S32: classifying the deviation points in the deviation measurement point data set based on the deviation direction, thereby generating classified deviation measurement point data; Step S33: analyzing the error sources of the classified deviation measuring point data based on the terrain influence and the measurement error, thereby obtaining a deviation measuring point error feature data set; Step S34: reconstructing the coordinates of the deviated measuring points according to the deviation measuring point error characteristic data set, and correcting the overall section line position of the updated measuring point coordinates, thereby obtaining section line equation correction data; Step S35: performing a supplementary calibration of the elevation and position consistency of the densely populated measurement point area near the section line equation correction data, thereby obtaining calibration consistency measurement point data; Step S36: converting the calibration consistency measurement point data into a unified format and outputting it as a standardized data file, thereby generating calibration measurement point data.

8. The artificial intelligence-based river survey data preprocessing method according to claim 7, characterized in that: Step S31 includes the following steps: Step S311: converting the preset cross-section line into a mathematical expression, and calculating the vertical projection point position of each measuring point in the retained measuring point data set to the cross-section line, thereby obtaining the projection point position data; Step S312: Calculate the Euclidean distance from the measuring point to its corresponding vertical projection point according to the projection point position data, thereby generating a measuring point deviation distance matrix; Step S313: Mark the measurement points whose deviation distance exceeds the preset threshold as deviation measurement points, so as to obtain a deviation measurement point data set.

9. The artificial intelligence-based river survey data preprocessing method according to claim 8, characterized in that: Step S34 includes the following steps: Step S341: performing vector analysis on the error direction of the deviation measuring point according to the deviation measuring point error feature data set, thereby generating measuring point vector offset data; Step S342: adjusting the deviated measuring point to the cross-section line position along the vertical direction based on the measuring point vector offset data, updating the spatial coordinates of the measuring point, thereby obtaining measuring point coordinate reconstruction data; Step S343: refitting the mathematical expression of the cross-section line according to the measurement point coordinate reconstruction data to generate cross-section line equation refitting data; Step S344: Perform residual statistical analysis on the cross-section line equation refitting data, and perform screening based on the error range according to the measurement point coordinates to reconstruct the data, so as to obtain cross-section line equation correction data.

10. The artificial intelligence-based river survey data preprocessing method according to claim 9, characterized in that: The reserved point addition and correction described in step S4 includes: Correct the calibration points that are missing or have large deviations in the calibration point data according to the blank area measurement point data, so as to obtain the calibration correction measurement point data, wherein the correction processing includes position reconstruction, elevation adjustment and spatial relationship optimization; According to the calibration correction measuring point data, the reserved points are added and corrected to obtain the corrected measuring point data.

Images