An automatic drawing method for the precise thalweg of a river course

Abstract

The present invention relates to the technical field of thalweg drawing, and discloses an automatic drawing method for accurate thalweg of a river channel. The method includes river channel vector data, obtaining the required river channel vector data from a river channel management agency, scanning the underwater topography of the river channel with a multibeam sounding instrument to obtain river bottom topography data, performing cloud thinning and noise reduction on the river bottom topography data points to obtain the full river bottom data, extracting the river channel center line from the river channel vector data, converting the river channel center line into river channel turning point coordinate data, extracting and fusing the full river bottom data and the river channel turning point coordinate data, thereby determining a number of river channel cross-section line data, and connecting the minimum z coordinates in the number of river channel cross-section line data into a line to obtain the thalweg of the river channel. The present invention collects river bottom topography data through a multibeam sounding system, and uses intelligent algorithms to thin and denoise the river channel basic data and the river bottom topography data, thereby improving the data accuracy and automatically generating the thalweg.

Description

Technical Field

[0001] The present invention relates to the field of thalweg drawing, and specifically to an automatic drawing method for accurately drawing the thalweg of a river channel. Background Art

[0002] The thalweg, also known as the stream line, is the connection line of the deepest water points along the direction of the river. The cross-section along this line is the longitudinal section of the river. In engineering hydrology, the thalweg is often misinterpreted as the median line, but in fact, there are obvious differences between the two. The median line refers to the connection line of the maximum flow velocity points on the surface of each cross-section of the river, while the thalweg is the connection line of the maximum water depth points of each cross-section.

[0003] Existing methods for drawing the thalweg of a river channel all use single-beam sounding technology to obtain river channel cross-sections with a spacing of more than 20 meters. For the drawing work of the thalweg, deep thalweg points are mostly manually extracted and then drawn using mapping software, or custom development based on geographic information systems is adopted to fuse and fit the lowest points of discrete cross-sections to generate the thalweg. Due to the fusion of multiple data sources and errors between individual data, there are large deviations between the locally generated thalweg and the actual thalweg, and the drawing efficiency is low. Summary of the Invention

[0004] The present invention provides an automatic drawing method for accurately drawing the thalweg of a river channel, which solves the problems mentioned in the above background art, namely, the existing thalweg drawing methods involve the fusion of multiple data sources, there are errors between individual data, there are large deviations between the locally generated thalweg and the actual thalweg, and the drawing efficiency is low.

[0005] The present invention provides the following technical solution: An automatic drawing method for accurately drawing the thalweg of a river channel, including the following steps:

[0006] River channel vector data, obtaining the required river channel vector data from the river channel management agency;

[0007] S1. A multi-beam sounder scans the underwater terrain of the river channel to obtain river bottom terrain data points, and the river bottom terrain data points include structured point X n , structured point Y n and structured point Z n ;

[0008] S2. Performing cloud thinning and noise reduction on the river bottom terrain data points obtained in S1 to obtain the full river bottom data;

[0009] S3. Extracting the river channel center line from the river channel vector data, and converting the river channel center line into river channel turning point coordinate data;

[0010] S4. Extract and fuse the full riverbed data obtained in S2 and the river channel turning point coordinate data obtained in S3 to determine a number of river channel cross-section line data points, where the river channel cross-section line data points include coordinate X, coordinate Y, and coordinate Z;

[0011] S5. Connect the minimum values of coordinate Z in a number of the river channel cross-section line data to form a line to obtain the thalweg line of the river channel.

[0012] As an alternative solution of the automatic drawing method for the accurate thalweg line of the river channel in the present invention, among them: a number of structured points X n , structured points Y n and structured points Z n are used to construct a first KD tree;

[0013] Variance operations are respectively performed on the structured points X n , structured points Y n and structured points Z n to select a dimension accordingly, and the data is divided according to the selected dimension;

[0014] Variance operations are performed on the selected divided data to obtain a number of first point cloud data, and the number of first point cloud data is divided into a first point cloud data set.

[0015] The variance calculation formula is ,

[0016] where N represents the number of data sets, Di represents the value of the points in the data set, and μ represents the average value of the data set;

[0017] A first distance threshold is set, the point cloud data greater than the set first distance threshold is removed, and the point cloud data less than the set first distance threshold is retained to complete the thinning of the data point cloud, and the retained point cloud data forms a second KD tree structure.

[0018] As an alternative solution of the automatic drawing method for the accurate thalweg line of the river channel in the present invention, among them: the Euclidean operation formula, the spacing of the retained points in the point cloud data is obtained through the Euclidean operation formula, and the Euclidean operation formula is ;

[0019] When the spacing of the retained points is less than the distance threshold, redundant points are determined and the retained points are thinned;

[0020] When the spacing of the retained points is greater than the distance threshold, it is determined as a key node and the retained point is retained;

[0021] The key nodes are merged to obtain a point cloud data set, and the cloud data set is converted into output structured points X m , output structured points Y m and output structured points Z m .

[0022] As an alternative solution of the automatic drawing method for the accurate thalweg of the river in the present invention, wherein: by calculating the output structured point X m , the output structured point Y m and the output structured point Z m a distance data set C is obtained;

[0023] The average distance D in the distance data set is obtained by operating on the distance data set C;

[0024] The median meanD of the average distance is obtained by operating on the average distance D in the distance data set;

[0025] The variance S is obtained by operating on the average distance D in the distance data set;

[0026] By comparing the average distance D with MaxD, the calculation formula of MaxD is:

[0027] MaxD = meanD + k * S;

[0028] meanD represents the median of the data set, k represents the standard deviation multiple, and S represents the standard deviation of the data set;

[0029] When the average distance D is greater than the value of MaxD, the output structured point is considered a noise point and the output structured point is removed;

[0030] When the average distance D is less than the value of MaxD, the output structured point is considered a pointing point and the output structured point is retained.

[0031] As an alternative solution of the automatic drawing method for the accurate thalweg of the river in the present invention, wherein: the output structured point X m , the output structured point Y m and the output structured point Z m are used to divide the data according to the selected dimension to obtain a number of second point cloud data, and the number of second point cloud data is divided into a second point cloud data set;

[0032] The second point cloud data set includes (X1, Y1, Z1), (X2, Y2, Z2), …, (X 10 , Y 10 , Z 10 ) base points;

[0033] Calculate the distance C from the adjacent points in the base point data set to the calculation base point, and the calculation formula is, ;

[0034] Distance data sets C1, C2, C3…C9 are obtained.

[0035] Calculate the average distance D in the distance dataset. The calculation formula is ,

[0036] to obtain the average distance datasets D1, D2, D3…D9;

[0037] In the formula, N represents the number of the dataset.

[0038] As an alternative solution of the automatic drawing method for the precise thalweg line of the river channel described in the present invention, wherein: calculate the median meanD of the average distance. The calculation formula is

[0039] meanD = median(D1, D2,…, Dn)

[0040] Calculate the variance S of the average distance. The calculation formula is ,

[0041] In the formula, N represents the number of the dataset, and μ represents the average value of the dataset;

[0042] The calculation formula of MaxD is

[0043] MaxD = meanD + k*S;

[0044] In the formula, meanD represents the median of the dataset, k represents the standard deviation multiple, and S represents the standard deviation of the dataset.

[0045] As an alternative solution of the automatic drawing method for the precise thalweg line of the river channel described in the present invention, wherein: convert the turning point coordinate data of the river channel into a number of serialized cross-sections;

[0046] The pointing points are distributed on a number of serialized cross-sections, and the pointing points not distributed on the serialized cross-sections are removed.

[0047] As an alternative solution of the automatic drawing method for the precise thalweg line of the river channel described in the present invention, wherein: set a range value F, and the set range value F is used to determine the pointing points on the serialized cross-sections;

[0048] Automatically screen out the appropriate pointing points ie, xe, ye, and ze.

[0049] As an alternative solution of the automatic drawing method for the precise thalweg line of the river channel described in the present invention, wherein: use JavaScript to identify the pointing points ie, xe, ye, and ze and perform segmentation processing;

[0050] The array format of the segmentation processing is xs, ys, and zs;

[0051] Merge xs, ys, and zs. Take the first element as the ID and the remaining elements as coordinate values, and combine them under the same ID. For the coordinate value array under each ID, filter out the third value, and take the smallest value as the final result. The smallest value is the thalweg point data.

[0052] As an alternative solution of the automatic drawing method for the precise thalweg line of the river channel described in the present invention, wherein: automatically draw the longitudinal profile change diagram of the thalweg line and the plane distribution diagram of the thalweg line through the thalweg point data.

[0053] The present invention has the following beneficial effects:

[0054] 1. For the automatic drawing method of the precise thalweg line of the river channel, collect the river bottom terrain data through a multi-beam sounding system, and use intelligent algorithms to thin and denoise the river channel basic data and the river bottom terrain data, so as to improve the data accuracy and automatically generate the thalweg line.

[0055] 2. For the automatic drawing method of the precise thalweg line of the river channel, serialize and extract and fuse the river channel basic data and the river bottom terrain data to further improve the data accuracy and greatly remove noise points. Brief Description of the Drawings

[0056] Figure 1 It is the method step diagram of the present invention.

[0057] Figure 2 It is the three-dimensional underwater terrain diagram of the present invention.

[0058] Figure 3 It is the schematic diagram of the turning point in the center of the river channel of the present invention.

[0059] Figure 4 It is the schematic diagram of the serialized cross-section line of the river channel of the present invention.

[0060] Figure 5 It is the schematic diagram of the extraction of river channel cross-section points of the present invention.

[0061] Figure 6 It is the longitudinal profile data change diagram of the thalweg line of the present invention.

[0062] Figure 7 It is the plane distribution diagram of the thalweg line of the present invention. Detailed Embodiments

[0063] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0064] Embodiment 1

[0065] Please refer to Figure 1-2 , a method for automatically drawing the precise thalweg of a river channel includes the following steps:

[0066] River channel vector data, obtain the required river channel vector data from the river channel management agency;

[0067] S1. Use a multibeam echosounder to scan the underwater terrain of the river channel to obtain river bottom terrain data points, and the river bottom terrain data points include structured points X n , structured points Y n and structured points Z n ;

[0068] S2. Perform cloud thinning and noise reduction on the river bottom terrain data points obtained in S1 to obtain the full river bottom data;

[0069] S3. Extract the river channel center line from the river channel vector data, and convert the river channel center line into river channel turning point coordinate data;

[0070] S4. Extract and fuse the full river bottom data obtained in S2 with the river channel turning point coordinate data obtained in S3 to determine several river channel cross-section line data points, and the river channel cross-section line data points include coordinate X, coordinate Y, and coordinate Z;

[0071] S5. Connect the minimum values of coordinate Z in several river channel cross-section line data to form a line to obtain the thalweg of the river channel.

[0072] Use an unmanned boat equipped with a multibeam echosounder to preliminarily scan the underwater terrain of the river channel to obtain the preliminary scanned three-dimensional river bottom terrain data,

[0073] Construct the first KD tree for several structured points X n , structured points Y n and structured points Z n ;

[0074] Perform variance operations on structured points X n , structured points Y n and structured points Z n respectively to select the dimension, and divide the data according to the selected dimension;

[0075] Perform variance operations on the selected divided data to obtain several first point cloud data, and divide the several first point cloud data into a first point cloud data set. This data set consists of structured point clouds X n , Y n and Z n constitute, the point cloud spacing is 0.25m, the total number of point clouds is 526352, the coordinate system uses the CGCS2000 plane coordinate, the elevation uses the geodetic elevation system, and obtain the required river channel vector data from the river channel management agency;

[0076] It should be noted that the river channel vector data format is shp, which is convenient for extracting the river channel center line;

[0077] Use the KD tree and Euclidean algorithm to thin the point cloud of the three-dimensional riverbed terrain data obtained in step 1. Thinning can reduce the data volume and thus improve the processing efficiency of subsequent steps;

[0078] Since the point cloud data collected by multibeam has the characteristic of high density, to reduce the amount of calculation, the first KD tree can be constructed according to the structured point X n 、structured point Y n and structured point Z n for the three-dimensional point cloud data set;

[0079] When k = 3, the structured point X n 、structured point Y n and structured point Z n respectively represent the coordinate values of the points in the three-dimensional space. X represents the first dimension, Y represents the second dimension, and Z represents the third dimension;

[0080] Select the dimension to start dividing the data according to the variance size in each dimension. The larger the variance, the more dispersed the data is distributed in that dimension. First, select the dimension with a large variance for division;

[0081] The variance calculation formula is, ,

[0082] where N represents the number of data sets, Di represents the value of the points in the data set, and μ represents the average value of the data set;

[0083] After calculation, the variance of the first dimension x is 2951.86, the variance of the second dimension y is 6121.16, and the variance of the third dimension z is 9.90. That is, select the second dimension y to divide the data.

[0084] For the selected division dimension y, obtain the median m = 3432185.12 of the data set in this dimension (when the number of data points is even, the median takes the average of the two middle numbers), and then divide the data set into two subsets:

[0085] Pl = {pi|pi ≤ m}, that is, the subset composed of points whose coordinate values in the second dimension y are less than or equal to the median m;

[0086] Pt = {pt|pt > m}, that is, the subset composed of points whose coordinate values in the second dimension y are greater than the median m;

[0087] Finally, recursively construct the second KD tree on Pl and Pt according to the above steps until the number of data points in the subset is less than the predetermined threshold. The threshold is set to 30 this time, and finally a complete KD tree structure is formed;

[0088] It should be noted that for point cloud data containing millions of points, the threshold can be set to 10 - 100;

[0089] Set the distance threshold B = 0.5m. This threshold determines the minimum distance between point clouds during the thinning process. The minimum three-dimensional spatial distance between any two points in the thinned point cloud will not be less than this value. The larger the set value, the fewer points are retained. For each node in the KD-tree structure, calculate the distance between the points (30) within the area represented by this node and the selected retained points. For two points p(X a ,Y a ,Z a ) and q(X b ,Y b ,Z b );

[0090] Using the Euclidean operation formula, the distance between the retained points in the point cloud data is obtained through the Euclidean operation formula. The Euclidean operation formula is ;

[0091] If the distance d from a certain point to the retained point is less than or equal to the distance threshold B, then this point is considered a redundant point and can be thinned out. If the distance d is greater than the threshold B, then this point is used as a key node and is retained. Merge all the retained point cloud data to obtain the thinned point cloud data set and output the result. Output the structured point cloud {(X m ,Y m ,Z m ). The point cloud spacing is 0.5m, and the total number of points in the point cloud is 112,171;

[0092] Due to various factors such as signals, hull measurement environments, and sound speeds during the acquisition of multi-beam point cloud data, there will be noise points to varying degrees. Therefore, it is necessary to denoise to improve the quality and usability of the data;

[0093] When the spacing between the retained points is less than the distance threshold, determine the redundant points and thin out the retained points; H

[0094] When the spacing between the retained points is greater than the distance threshold, determine it as a key node and retain the retained point;

[0095] Merge the key nodes to obtain the point cloud data set, and convert the cloud data set into the output structured points X m , the output structured point Y m and the output structured point Z m ;

[0096] {(X m ,Y m ,Z m}Divide the data according to the second dimension of Y, set the threshold to 10, and reform the complete KD-tree structure.

[0097] Take the subset {(X1, Y1, Z1), (X2, Y2, Z2), …, (X 10 , Y 10 , Z 10 )} as an example. Each point in the subset {e.g., (X1, Y1, Z1)} is used as the calculation base point in turn, and the distance C from the neighboring points in the dataset to the calculation base point is calculated;

[0098] Calculate the distance C from the neighboring points in the dataset of the calculation base point to the calculation base point. The calculation formula is ;

[0099] Obtain the distance datasets C1, C2, C3…C9;

[0100] Operate on the distance dataset C to obtain the average distance D in the distance dataset;

[0101] Calculate the average distance D in the distance dataset. The calculation formula is ,

[0102] Obtain the average distance datasets D1, D2, D3…D9;

[0103] Obtain the median meanD of the average distance by operating on the average distance D in the distance dataset;

[0104] The calculation formula for MaxD is

[0105] MaxD = meanD + k * S;

[0106] In the formula, meanD represents the median of the dataset, k represents the standard deviation multiple, and S represents the standard deviation of the dataset;

[0107] Obtain the variance S by operating on the average distance D in the distance dataset;

[0108] By comparing the average distance D with MaxD, the calculation formula for MaxD is:

[0109] MaxD = meanD + k * S;

[0110] meanD represents the median of the dataset, k represents the standard deviation multiple, and S represents the standard deviation of the dataset;

[0111] When the average distance D is greater than the value of MaxD, the output structured point is considered a noise point and the output structured point is removed;

[0112] When the average distance D is less than the value of MaxD, the output structured point is considered a pointing point and the output structured point is retained;

[0113] Output the structured point X m , output the structured point Y m and output the structured point Z m Divide the data according to the selected dimension to obtain a number of second point cloud data, and divide the number of second point cloud data into second point cloud data sets;

[0114] The second point cloud data set includes (X1, Y1, Z1), (X2, Y2, Z2), …, (X 10 , Y 10 , Z 10 ) base points;

[0115] In the formula, N represents the number of data sets.

[0116] Calculate the median meanD of the average distance. The calculation formula is

[0117] meanD = median(D1, D2, …, Dn)

[0118] Calculate the variance S of the average distance. The calculation formula is ,

[0119] In the formula, N represents the number of data sets, and μ represents the average value of the data set;

[0120] Use the "rcScan" tool in the ArcGIS geographic information system to extract the center line of the river channel from the river channel vector data. The obtained center line of the river channel is connected by turning points. The greater the river channel bending coefficient, the more turning points. After extracting the center line of the river channel, obtain the turning point coordinate data using the CGCS2000 plane coordinate;

[0121] As Figure 3 shown, the obtained turning point coordinate data is as follows:

[0122] X1, Y1;

[0123] X2, Y2;

[0124] X3, Y3;

[0125] ……;

[0126] X P , Y P ;

[0127] After obtaining the turning point coordinate data of the center line of the river channel, the straight line L1 formed by two adjacent points (X1, Y1) (X2, Y2)

[0128] The serialization cross-sectional line formula of the straight line L1 is expressed as: ,

[0129] Among them, ,

[0130] In the formula, k represents the slope of the straight line L1; Q represents the spacing between the layout line cross-sections. The larger the value, the fewer the generated serialized cross-sections. In this case, to achieve an accuracy of 1 m, Q is set to 1 m. Q0 represents that the vertical distance of the cross-section line remains Q, i represents the cross-section line number. When starting to draw the cross-section line from the turning point (X1, Y1), i = 0, and B represents the y-axis intercept of the layout line.

[0131] Among them, the cross-section number i conforms to the following formula: ,

[0132] After calculation, the serialized cross-sections of the straight line formed by the turning point (X1, Y1) and the turning point (X2, Y2) are as Figure 4 shown;

[0133] Convert the river channel turning point coordinate data into several serialized cross-sections;

[0134] The pointing points are distributed on several serialized cross-sections, and the pointing points that are not distributed on the serialized cross-sections are removed;

[0135] After forming the serialized cross-section line, due to the dispersive characteristics of the point cloud, it is impossible to ensure that all the point clouds are on the cross-section line. Set the range value F, and the range value F is used to determine the pointing points on the serialized cross-sections;

[0136] Set the range value F = 0.25:

[0137] The calculation process is as Figure 5 shown. The set range value F is used to include the points near the cross-section line, and its formula can be expressed based on step 3 as: ,

[0138] Among them, the range value F conforms to the following formula: ,

[0139] Automatically screen out the pointing points ie, xe, ye, and ze that meet this formula, and the output format is cross-section line number i, coordinate x, coordinate y, coordinate z;

[0140] Automatically screen out the appropriate pointing points ie, xe, ye, and ze;

[0141] Use JavaScript to identify the pointing points ie, xe, ye, and ze and perform segmentation processing;

[0142] The segmentation processing array format is xs, ys, and zs;

[0143] Merge xs, ys, and zs, with the first element as the ID and the remaining elements as coordinate values, under the same ID. For the coordinate value arrays under each ID, filter out the third value, and take the smallest value as the final result. The smallest value is the thalweg point data;

[0144] Automatically draw the longitudinal profile change diagram of the thalweg line and the plan distribution diagram of the thalweg line through the thalweg point data. The longitudinal profile change diagram of the thalweg line can visually display the elevation information of the thalweg line, and the plan distribution diagram of the thalweg line can visually display the plan information of the thalweg line;

[0145] Draw the longitudinal profile change diagram of the thalweg line for the data xs, ys, zmin output in step 4 in Echarts. First, specify the chart type as 'line' in Echarts;

[0146] As Figure 6 shown, specify the data source of the line chart as xs, ys, zmin. Each data point corresponds to a point on the line chart. Set the line of the line chart as a smooth curve, connect the data points through the smooth curve, set the size of the inflection point to 10 pixels, define the style of the coordinate axis line, set the color of the coordinate axis line to black, set the width of the coordinate axis line to 1 pixel, and set the transparency of the coordinate axis line to 1 (completely opaque);

[0147] The X-axis represents the straight-line distance between thalweg points, and the y-axis represents zmin in is: [xs, ys, zmin].

[0148] As Figure 7 shown, convert xs and ys in xs, ys, zmin to longitude and latitude format by calling the Proj4js library to obtain new data Bj and Lj. Use the Cesium.Entity class of Cesium to create the thalweg line. Set the polyline representation of the thalweg line through the polygon property, specify the vertex positions (xj, yj) of the polyline using the positions property, set the width property of the polyline to control the line width to 5.0 pixels, and add the thalweg line object to the Gaode imagery scene by calling the add method of the entities property of the CesiumViewer instance. You can use the view control functions of Cesium, such as zooming, panning, and rotating, to observe the three-dimensional shape and positional relationship of the thalweg line.

[0149] It should be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprising", "including" or any other variant thereof are intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements not only includes those elements but also includes other elements not expressly listed, or elements inherent to such process, method, article or device.

[0150] The above are only the preferred embodiments of the present invention. It should be pointed out that for those of ordinary skill in the art, without departing from the technical principle of the present invention, several improvements and modifications can be made, and these improvements and modifications should also be regarded as the protection scope of the present invention.

Claims

1. An automatic drawing method for the precise thalweg of a river channel, characterized in that, It includes the following steps: River channel vector data, and obtain the required river channel vector data from the river channel management agency; S1. The multi-beam sounding instrument scans the underwater terrain of the river channel to obtain river bottom terrain data points, and the river bottom terrain data points include structured points structured points and structured points ; S2. Perform cloud thinning and noise reduction on the river bottom terrain data points obtained in S1 to obtain the full river bottom data, and output the structured point cloud; S3. Extract the center line of the river channel from the river channel vector data. The obtained center line of the river channel is connected by turning points. After extracting the center line of the river channel, obtain the turning point coordinate data using the CGCS2000 plane coordinate; S4. Extract and fuse the full river bottom data obtained in S2 and the river channel turning point coordinate data obtained in S3. Take each point in the point cloud dataset as the calculation base point in turn to obtain multiple sets of distance datasets C, and calculate the average distance D and the median meanD of the average distance through the distance dataset C; Obtain the median meanD of the average distance by calculating the average distance D in the distance dataset; Obtain the variance S by calculating the average distance D in the distance dataset; By comparing the average distance D with MaxD, the calculation formula of MaxD is: MaxD = meanD + k * S; k represents the standard deviation multiple; When the average distance D is greater than the value of MaxD, it is considered that the output structured point is a noise point, and the output structured point is removed; When the average distance D is less than the value of MaxD, it is considered that the output structured point is a pointing point, and the output structured point is retained; Connect every two adjacent points of the determined several river channel turning point coordinates to form a straight line L1. Obtain the serialized cross-section through the serialized cross-section line formula of the straight line L1. Set the range value F. The range value F is used to determine the pointing points distributed on the serialized cross-section, so as to determine several river channel cross-section line data points, and remove the pointing points not distributed on the serialized cross-section. The river channel cross-section line data points include coordinate x, coordinate y, and coordinate z; S5. Connect the minimum values of the coordinate z in several river channel cross-section lines to form a line to obtain the thalweg line of the river channel.

2. The automatic drawing method for the precise thalweg of a river channel according to claim 1, characterized in that: Construct a first KD tree from a number of structured points , structured points and structured points ; Structured points and structured points and structured points Perform variance calculations respectively to select dimensions, and divide the data according to the selected dimensions. Structured points Structured points and structured points respectively represent the coordinate values of points in three-dimensional space, where x represents the first dimension, y represents the second dimension, and z represents the third dimension; Perform variance operation on the selected divided data to obtain several first point cloud data, and divide the several first point cloud data into the first point cloud dataset, The variance calculation formula is, ； where N represents the number of data sets, represents the value of the points in the data set, represents the average value of the data set; Set the first distance threshold, remove the point cloud data greater than the set first distance threshold, and retain the point cloud data less than the set first distance threshold to complete the thinning of the data point cloud. The retained point cloud data constitutes the second KD tree structure.

3. The method for automatically drawing the accurate thalweg line of the river channel according to claim 2, characterized in that: The Euclidean operation formula, the point cloud data obtains the spacing of the retained points in the point cloud data through the Euclidean operation formula. The Euclidean operation formula is, ； When the spacing of the retained points is less than the distance threshold, determine the redundant points and thin the retained points; When the spacing of the retained points is greater than the distance threshold, determine it as a key node and retain the retained point; Merge the key nodes to obtain a point cloud data set, and convert the cloud data set into an output structured point and output structured point and output structured point .

4. The method for automatically drawing the precise thalweg of a river channel according to claim 3, characterized in that: By operating on the output structured points , the output structured points and the output structured points , multiple groups of distance data sets C are obtained by using each point in the subset as a calculation base point in sequence; Calculate the dataset C of the adjacent points in the base point dataset to each calculation base point. The calculation formula is, ； ； … ； Obtain the distance datasets C1, C2, C3... C9; Operate on the distance dataset C to obtain the average distance D in the distance dataset; Calculate the average distance D in the distance dataset. The calculation formula is, ； Obtain the average distance datasets D1, D2, D3... D9.

5. The method for automatically drawing the precise thalweg of a river channel according to claim 4, wherein: Structured output points Structured output points Structured output points Divide the data according to the selected dimension to obtain a number of second point cloud data, and divide the number of second point cloud data into a second point cloud data set.

6. The method for automatically drawing the precise thalweg of a river channel according to claim 5, characterized in that: Calculate the median meanD of the average distance. The calculation formula is meanD = median(D1, D2, …, Dn); Calculate the variance S of the average distance. The calculation formula is ； In the formula, N represents the number of data sets, represents the average value of the data set.

7. The method for automatically drawing the precise thalweg of a river channel according to claim 6, characterized in that: The output format is the cross-section line number i, coordinate x, coordinate y, coordinate z, and automatically filter out the appropriate pointing points ie, xe, ye, and ze.

8. The method for automatically drawing the precise thalweg of a river channel according to claim 7, characterized in that: Use JavaScript to identify the pointing points ie, xe, ye, and ze and perform segmentation processing; The segmentation processing array format is xs, ys, and zs; Take the first element of xs, ys, and zs as the ID, and the remaining elements as the coordinate values, and merge them under the same ID. For the coordinate value array under each ID, filter out the third value, and the smallest value among them is used as the final result. The said minimum value is the thalweg point data.

9. The method for automatically drawing the precise thalweg of a river channel according to claim 8, wherein: Automatically draw the longitudinal variation diagram of the thalweg line and the plane distribution diagram of the thalweg line through the said thalweg point data.

Images