Hydropower station live-action three-dimensional model construction method based on texture mapping technology

Abstract

The invention relates to the technical field of three-dimensional model construction, in particular to a hydropower station live-action three-dimensional model construction method based on a texture mapping technology, and the construction method comprises the steps: obtaining hydropower station point cloud data through employing a wireless technology; constructing a three-dimensional point cloud model of the hydropower station based on the collected point cloud data; constructing a mapping relation between the texture image and the three-dimensional point cloud model; based on the mapping relationship between the texture image and the three-dimensional point cloud model, completing the construction of a three-dimensional model of the hydropower station; the unmanned aerial vehicle carries the camera equipment and the laser scanner to obtain image data of the hydropower station and the surrounding area of the hydropower station, high-precision positioning and attitude determination of the hydropower station are achieved, and therefore the construction precision of the three-dimensional model is improved; data preprocessing, data segmentation and classification methods are adopted to process the collected point cloud data, so that the authenticity and accuracy of the data are improved; by setting a judgment rule, automatic noise point elimination is realized, and the automation degree of data processing is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of three-dimensional model construction, and specifically to a method for constructing a real-scene three-dimensional model of a hydropower station based on texture mapping technology. Background Technique

[0002] The three-dimensional model of a hydropower station can clearly show the overall layout of the hydropower station. It is particularly important to construct a three-dimensional model of the hydropower station during the operation and management of the power station. Although the three-dimensional point cloud model obtained by means such as laser scanning can reflect the spatial characteristics of the target object, the surface rendering effect is poor and the authenticity is lacking.

[0003] Texture mapping is the mapping from a two-dimensional image to the surface of a three-dimensional model, which can improve the surface rendering effect of the three-dimensional model and increase the authenticity. However, at present, the texture mapping of three-dimensional models mainly relies on manual construction, with low efficiency and easy errors.

[0004] The Chinese invention application "A Method and System for Texture Mapping of Point Cloud Models" with the publication number CN108280870A discloses establishing a mapping relationship between a texture picture and a point cloud model through a virtual sphere. First, establish the corresponding relationship between the texture picture and the virtual sphere, and then establish the corresponding relationship between the virtual sphere and the point cloud model to obtain the mapping relationship between the texture picture and the point cloud model, realizing the automatic mapping of the texture picture and the point cloud model. However, the construction of the virtual sphere greatly increases the computational complexity of image processing and reduces the efficiency of constructing the three-dimensional model. Summary of the Invention

[0005] In view of the above existing problems, the present invention is proposed.

[0006] Therefore, the present invention provides a method for constructing a real-scene three-dimensional model of a hydropower station based on texture mapping technology. Through the mapping relationship between the texture image and the three-dimensional point cloud model, the automatic mapping and fusion of the texture image and the three-dimensional point cloud model are realized, and a real-scene three-dimensional model of the hydropower station at the macro and meso levels is obtained, solving the problem that the current three-dimensional point cloud model lacks surface texture details and has a poor model rendering effect.

[0007] To solve the above technical problems, the present invention provides the following technical solution: A method for constructing a real-scene three-dimensional model of a hydropower station based on texture mapping technology, including the following steps,

[0008] Obtain hydropower station point cloud data using wireless technology;

[0009] Construct a three-dimensional point cloud model of the hydropower station based on the collected point cloud data;

[0010] Construct the mapping relationship between the texture image and the three-dimensional point cloud model;

[0011] Based on the mapping relationship between the texture image and the three-dimensional point cloud model, complete the construction of the three-dimensional model of the hydropower station.

[0012] As a preferred solution of the method for constructing a real - scene three - dimensional model of a hydropower station based on texture mapping technology described in the present invention, wherein: the acquisition of point cloud data of the hydropower station using wireless technology is to use a drone equipped with a camera device, and obtain image data of the hydropower station and its surrounding area through aerial photography, so as to obtain the positioning and orientation data of the hydropower station, and then use the GPS - RTK technology to collect control point data. The acquisition of image data of the hydropower station and its surrounding area through aerial photography is to obtain image data of the hydropower station and its surrounding area by using a drone equipped with a line - array camera and a laser scanner. The acquisition of the positioning and orientation data of the hydropower station is to match the control points in the image data with the control points in the real - world geographical coordinates, and calculate the positioning and orientation parameters of the image according to the geographical coordinates of the known control points and the corresponding pixel coordinates in the image. The specific calculation is as follows:

[0013] It is known that the position coordinates of the line - array camera carried by the drone are C(X, Y, Z) and the attitude coordinates are And the position coordinates and attitude coordinates of the camera are extracted according to the rotation matrix R and the translation matrix T, and then the acquisition of the positioning and orientation data of the hydropower station is completed. The specific calculation formula is as follows:

[0014] P geo = RP img + T

[0015] Wherein, R represents the rotation matrix, T represents the translation matrix, P img is the pixel point coordinate in the image data, and P geo represents the positioning and orientation parameters calculated according to the image data.

[0016] As a preferred solution of the method for constructing a real - scene three - dimensional model of a hydropower station based on texture mapping technology described in the present invention, wherein: the construction of the three - dimensional point cloud model of the hydropower station based on the collected point cloud data is to construct the three - dimensional point cloud model of the hydropower station based on the point cloud data. Before constructing the three - dimensional point cloud model of the hydropower station, the collected point cloud data is processed, and according to the data processing results, the three - dimensional point cloud model of the hydropower station is constructed. The processing of the collected point cloud data includes data pre - processing and data segmentation and classification. The data pre - processing is to pre - process the collected point cloud data to ensure the authenticity and accuracy of the data, including statistical filtering and radius filtering. The statistical filtering is to process based on the statistical characteristics of the neighborhood around the point cloud data points. For each data point, calculate the statistical information of the surrounding points and compare the calculation results with the set threshold, and then complete the removal of noise points. The specific implementation formula is as follows:

[0017]

[0018]

[0019] α = μ + 2σ

[0020] Among them, μ represents the average value of the points in the neighborhood, σ represents the standard deviation of the points in the neighborhood, x i represents the value of each point in the neighborhood, and α represents the set threshold for judging and removing noise points.

[0021] As a preferred solution of the method for constructing a three-dimensional real scene model of a hydropower station based on texture mapping technology according to the present invention, wherein: the judgment for removing noise points is automatically judged according to the set judgment rules, and the specific judgment rules are as follows:

[0022] When the comparison between the value of each point in the neighborhood and the set threshold satisfies the formula α - 1 ≤ x i ≤ α + 1, it means that the current point in the neighborhood is not a noise point, and the current data point is retained;

[0023] When the comparison between the value of each point in the neighborhood and the set threshold satisfies the formula α - 1 > x i it means that the current point in the neighborhood is a noise point, and the current data point is removed;

[0024] When the comparison between the value of each point in the neighborhood and the set threshold satisfies the formula x i > α + 1, it means that the current point in the neighborhood is a noise point, and the current data point is removed;

[0025] The radius filtering uses the radius r as the set threshold to complete the judgment of noise points, and the specific judgment rules are as follows:

[0026] When the comparison result between the distance between the data point p j and the center point p i and the radius r satisfies the formula |P j - P i | > R, it means that the current data point is a noise point, and the current data point is removed;

[0027] When the comparison result between the distance between the data point p j and the center point p i and the radius r satisfies the formula |P j - P i | ≤ R, it means that the current data point is not a noise point, and the current data point is retained.

[0028] As a preferred solution of the method for constructing a three-dimensional real-scene model of a hydropower station based on texture mapping technology described in the present invention, wherein: the data segmentation and classification is to segment the collected hydropower station point cloud data into different parts and classify each part. The specific calculation formula is as follows:

[0029]

[0030]

[0031] Among them, f(x) represents the objective function, which represents the sum of the distances between the sample points in all categories and their respective cluster centers. k represents the total number of clusters preset, C i represents all the sample points in the i-th cluster, and μ i represents the center point of the i-th cluster;

[0032] The classification of each part is carried out according to the calculation result of the objective function, specifically as follows:

[0033] For each sample point x i , by calculating its distance from each cluster center μ i , and based on the calculated distance, each sample point is assigned to the category corresponding to the nearest cluster center. For each category, their cluster center μ j is recalculated, and the process of sample point assignment and cluster center update is repeated until all sample points are assigned, thereby completing the data segmentation and classification.

[0034] As a preferred solution of the method for constructing a three-dimensional real-scene model of a hydropower station based on texture mapping technology described in the present invention, wherein: the construction of the mapping relationship between the texture image and the three-dimensional point cloud model is to project the collected three-dimensional point cloud data onto a planar image and construct the mapping relationship of the three-dimensional point cloud model according to the pixel coordinates in the planar image. The specific implementation is as follows:

[0035]

[0036] Among them, (u, v) represents the pixel coordinates projected onto the plane, and the color information corresponding to the pixel points in each texture image can be accessed according to the coordinates. (X c , Y c , Z c ) represents the transformed camera coordinates, f x , f y represents the focal length of the camera, and (c x , c y ) represents the principal point coordinates.

[0037] As a preferred solution of the method for constructing a real - scene 3D model of a hydropower station based on texture mapping technology described in the present invention, wherein: the mapping relationship for constructing the 3D point cloud model is established by taking the point feature as the homologous feature of the texture image and the point cloud model, and the mapping relationship between the texture image and the 3D point cloud model is established. Taking the point feature as the homologous feature of the texture image and the point cloud model uses the nearest - neighbor search method to find the 3D points corresponding to the feature points in the point cloud model. The specific implementation is as follows:

[0038] When the pixel coordinates in the texture image are (u, v), and the 3D point coordinates in the point cloud model are , the corresponding 3D points are found by calculating the Euclidean distance. The specific calculation formula is as follows:

[0039]

[0040] where (u, v) represents the pixel coordinates in the texture image, and (u i , v i ) represents the pixel coordinates of the 3D points in the point cloud model projected onto the image plane. By calculating the Euclidean distance of all points in the current area until the Euclidean distance satisfies the formula D ≤D i-1 ≤D i ≤D i+1 , it means that the 3D points in the current point cloud model are the points corresponding to the feature points, and thus the establishment of the mapping relationship between the pixel coordinates in the texture image and the 3D point coordinates in the point cloud model is completed.

[0041] As a preferred solution of the method for constructing a real - scene 3D model of a hydropower station based on texture mapping technology described in the present invention, wherein: the construction of the 3D model of the hydropower station is based on the established mapping relationship to perform the fusion of the texture image and the 3D point cloud model, and thus the construction of the 3D model of the hydropower station is completed; the fusion of the texture image and the 3D point cloud model is to calculate the point cloud fitting error corresponding to the mapping relationship according to the established mapping relationship until the point cloud fitting error is the same as the Euclidean distance formula corresponding to the mapping relationship. The point cloud fitting error is the distance from the feature point to the fitting equation. The specific calculation formula is as follows:

[0042]

[0043] where N represents the total number of points in the point cloud data, (x i , y i , z i ) represents the pixel coordinates of the i - th point in the point cloud dataset, (x c , y c , z c ) represents the center coordinates of the fitted sphere, r represents the radius of the fitted sphere, and D′ represents the calculated fitting error;

[0044] The fusion of the texture image and the three-dimensional point cloud model is based on the calculated fitting error, and the specific fusion rules are as follows:

[0045] When the calculated fitting error D′ and the Euclidean distance D satisfy the formula D - D′ ≤ 0, it indicates that the current feature point meets the fusion requirements, and the current texture image and the three-dimensional point cloud model are automatically fused;

[0046] When the calculated fitting error D′ and the Euclidean distance D satisfy the formula D - D′ > 0, it indicates that the current feature point does not meet the fusion requirements, and the establishment of the mapping relationship between the pixel coordinates in the texture image and the three-dimensional point coordinates in the point cloud model continues until the calculated fitting error D′ and the Euclidean distance D satisfy the formula D - D′ ≤ 0.

[0047] A computer device, including a memory and a processor, the memory stores a computer program, characterized in that when the processor executes the computer program, it realizes the steps of the method for constructing a real-scene three-dimensional model of a hydropower station based on texture mapping technology.

[0048] A computer-readable storage medium, on which a computer program is stored, characterized in that when the computer program is executed by a processor, it realizes the steps of the method for constructing a real-scene three-dimensional model of a hydropower station based on texture mapping technology.

[0049] Advantages of the present invention: The present invention uses a drone to carry a camera device and a laser scanner to obtain image data of a hydropower station and its surrounding area, and combines GPS-RTK technology to collect control point data, realizing high-precision positioning and orientation of the hydropower station, thereby improving the construction accuracy of the three-dimensional model; and uses data preprocessing, data segmentation and classification methods to process the collected point cloud data, effectively removing noise points, improving the authenticity and accuracy of the data, while reducing the complexity of data processing and improving the processing efficiency; by setting judgment rules, automatic elimination of noise points is realized, reducing the need for manual intervention and improving the automation degree of data processing; also projects the three-dimensional point cloud data onto a plane image, constructs the mapping relationship of the three-dimensional point cloud model according to the pixel coordinates in the plane image, realizes high-precision texture mapping, and improves the visual effect and authenticity of the model; at the same time, the constructed three-dimensional model of the hydropower station has high precision and realism, can be used in multiple fields such as the design, planning, and operation of hydropower stations, and improves the practicality of the three-dimensional model of the hydropower station. Description of the Drawings

[0050] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. Among them:

[0051] Figure 1 It is a schematic diagram of the overall method step structure of a method for constructing a three-dimensional real-scene model of a hydropower station based on texture mapping technology of the present invention.

[0052] Figure 2 It is a schematic diagram of the logical judgment for removing noise points in a method for constructing a three-dimensional real-scene model of a hydropower station based on texture mapping technology of the present invention. Specific Embodiments

[0053] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will make a detailed description of the specific embodiments of the present invention in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all embodiments. Based on the embodiments in 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.

[0054] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0055] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that can be included in at least one implementation manner of the present invention. The "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that mutually excludes other embodiments.

[0056] The present invention is described in detail in conjunction with the schematic diagrams. When detailing the embodiments of the present invention, for the convenience of explanation, the cross-sectional views showing the device structure will be enlarged locally not in accordance with the general scale, and the schematic diagrams are only examples and should not limit the protection scope of the present invention herein. In addition, in actual production, three-dimensional spatial dimensions of length, width, and depth should be included.

[0057] Meanwhile, in the description of the present invention, it should be noted that the orientation or positional relationship indicated by terms such as "upper, lower, inner, and outer" is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the present invention. In addition, the terms "first, second, or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0058] Unless otherwise clearly specified and defined in the present invention, the terms "installation, connection, and coupling" shall be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can also be a mechanical connection, an electrical connection, or a direct connection, and can also be indirectly connected through an intermediate medium, or it can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0059] Embodiment 1

[0060] Referring to Figures 1 to 2 , as the first embodiment of the present invention, a method for constructing a real - scene three - dimensional model of a hydropower station based on texture mapping technology is provided, including the following steps.

[0061] S1: Obtain the point cloud data of the hydropower station using wireless technology.

[0062] Specifically, the step of obtaining the point cloud data of the hydropower station using wireless technology is to use a drone equipped with a camera device, and obtain the image data of the hydropower station and its surrounding area through aerial photography, so as to obtain the positioning and pose data of the hydropower station, and then use the GPS - RTK technology to collect the control point data.

[0063] Further, the step of obtaining the image data of the hydropower station and its surrounding area through aerial photography is to obtain the image data of the hydropower station and its surrounding area by using a drone equipped with a line - array camera and a laser scanner. The step of obtaining the positioning and pose data of the hydropower station is to match the control points in the image data with the control points in the real - world geographical coordinates, and calculate the positioning and pose parameters of the image according to the geographical coordinates of the known control points and the corresponding pixel coordinates in the image. The specific calculation is as follows:

[0064] It is known that the position coordinates of the line - array camera carried by the drone are C(X,Y,Z) and the attitude coordinates are And the position coordinates and attitude coordinates of the camera are extracted according to the rotation matrix R and the translation matrix T, and then the acquisition of the positioning and pose data of the hydropower station is completed. The specific calculation formula is as follows:

[0065] P geo = RP img + T

[0066] Among them, R represents the rotation matrix, T represents the translation matrix, and P img is the pixel point coordinate in the image data, and P geo represents the positioning and pose parameters calculated according to the image data.

[0067] It should be noted that the acquisition of control point data using the GPS-RTK technology is achieved by establishing a base station with a known position at the control points in the real geographical coordinates, deploying a mobile signal receiver in the base station to correct the acquired data, and using the differential positioning technology to improve the accuracy of position measurement. The specific implementation formula is as follows:

[0068] φ′ = φ obs -Δφ + Nλ + ε

[0069] Among them, Δφ represents the difference in the GPS signal carrier phase received between the signal receiver and the base station, φ obs represents the carrier phase observation value observed by the receiver, φ′ represents the corrected carrier phase observation value, Nλ represents the integer ambiguity degree, which is the offset of the signal with respect to an integer number of wavelengths during propagation, and ε represents the error term, which is the allowable error range of the receiver during actual measurement.

[0070] S2: Construct a three-dimensional point cloud model of the hydropower station based on the collected point cloud data.

[0071] Specifically, the construction of the three-dimensional point cloud model of the hydropower station based on the collected point cloud data is to construct a three-dimensional point cloud model of the hydropower station based on the point cloud data. Before constructing the three-dimensional point cloud model of the hydropower station, the collected point cloud data is processed, and a three-dimensional point cloud model of the hydropower station is constructed according to the data processing results.

[0072] It should be noted that the data processing of the collected point cloud data includes data preprocessing and data segmentation and classification. The data preprocessing is to preprocess the collected point cloud data to ensure the authenticity and accuracy of the data, including statistical filtering and radius filtering. The statistical filtering is processed based on the statistical characteristics of the neighborhood around the point cloud data points. For each data point, the statistical information of the points around it is calculated, and the calculation result is compared with the set threshold, and then the noise points are removed. The specific implementation formula is as follows:

[0073]

[0074]

[0075] α = μ + 2σ

[0076] Among them, μ represents the average value of the points in the neighborhood, σ represents the standard deviation of the points in the neighborhood, and xi represents the value of each point in the neighborhood, and α represents the set threshold for determining and removing noise points;

[0077] The determination and removal of the noise points are automatically judged according to the set judgment rules. The specific judgment rules are as follows:

[0078] When the value of each point in the neighborhood is compared with the set threshold and satisfies the formula α - 1 ≤ x i ≤ α + 1, it means that the current point in the neighborhood is not a noise point, and the current data point is retained;

[0079] When the value of each point in the neighborhood is compared with the set threshold and satisfies the formula α - 1 > x i it means that the current point in the neighborhood is a noise point, and the current data point is removed;

[0080] When the value of each point in the neighborhood is compared with the set threshold and satisfies the formula x i > α + 1, it means that the current point in the neighborhood is a noise point, and the current data point is removed;

[0081] The radius filtering takes the radius r as the set threshold to complete the judgment of noise points. The specific judgment rules are as follows:

[0082] When the comparison result of the distance between the data point p j and the center point p i with the radius r satisfies the formula |P j - P i | > R, it means that the current data point is a noise point, and the current data point is removed;

[0083] When the comparison result of the distance between the data point p j and the center point p i with the radius r satisfies the formula |P j - P i | ≤ R, it means that the current data point is not a noise point, and the current data point is retained.

[0084] Furthermore, the data segmentation and classification is to segment the collected hydropower station point cloud data into different parts and classify each part. The specific calculation formula is as follows:

[0085]

[0086]

[0087] Among them, f(x) represents the objective function, which represents the sum of the distances between the sample points in all categories and their respective cluster centers. k represents the total number of clusters preset, C idenotes all sample points in the \(i\)-th cluster, \(\mu\) i denotes the center point of the \(i\)-th cluster;

[0088] It should be noted that the classification of each part is based on the calculation result of the objective function, specifically as follows:

[0089] For each sample point \(x\) i , by calculating its distance from each cluster center \(\mu\) i , and based on the calculated distance, each sample point is assigned to the category corresponding to the nearest cluster center. For each category, their cluster center \(\mu\) j is recalculated, and the process of sample point assignment and cluster center update is repeated until all sample points are assigned, thus completing the segmentation and classification of the data.

[0090] Furthermore, the construction of the three-dimensional point cloud model of the hydropower station is to perform epipolar rectification on RGB color images and use the epipolar image pair technology for stereo matching, thereby completing the construction of the point cloud model. The epipolar rectification is to detect feature points in two RGB color images, perform feature matching on the feature points in the two images, establish corresponding point pairs at the same time, and then remap the corresponding points in the two images to the same horizontal direction to achieve image correction and alignment for subsequent construction of the three-dimensional point cloud model. The specific implementation of the epipolar rectification is as follows:

[0091] For the feature points in the first RGB image, the epipolar line of the current feature point in the second RGB image is calculated through the fundamental matrix \(F\) and the coordinates of the feature points in the current image. For each feature point, the pixel points on the currently calculated epipolar line are compared with the feature points in the first image through the epipolar line corresponding to the feature point calculated on the second image, and then the optimal matching points are compared, and the compared matching points are moved to the same horizontal direction as their corresponding feature points, thereby completing the epipolar rectification. At the same time, the pixel points in the second image are remapped according to the epipolar rectification result to obtain the corrected image data.

[0092] Even further, the construction of the three-dimensional point cloud model is to perform stereo matching on the image data corrected by the epipolar image pair to obtain a disparity map, and convert the disparity map into point cloud data, thereby completing the construction of the three-dimensional point cloud data model. The specific implementation is as follows:

[0093] For the pixel coordinates \((x, y)\) in the known disparity map, the depth image corresponding to the disparity map is calculated according to the baseline length \(B\) and focal length \(f\) of the camera. The specific formula is as follows:

[0094]

[0095] Among them, D(x, y) represents the time difference value in the disparity map, Z(x, y) represents the calculated image depth, B represents the baseline length of the camera, and f represents the focal length of the camera;

[0096] And use the reference matrix K in the camera to convert the pixel coordinates into camera coordinates, specifically as follows:

[0097]

[0098] Among them, (X c , Y c , Z c ) represents the converted camera coordinates, (x, y) represents the pixel coordinates in the disparity map, and K represents the reference matrix in the camera;

[0099] Based on the rotation matrix R and translation matrix T in the camera, convert the camera coordinates into actual coordinates, specifically as follows:

[0100]

[0101] Among them, represents the converted three-dimensional point cloud data, and the set of all three-dimensional point cloud data is then constructed into a three-dimensional point cloud model. R represents the rotation matrix of the camera, T represents the translation matrix of the camera, and (X c , Y c , Z c ) represents the converted camera coordinates.

[0102] S3: Construct the mapping relationship between the texture image and the three-dimensional point cloud model.

[0103] Specifically, the construction of the mapping relationship between the texture image and the three-dimensional point cloud model is to project the collected three-dimensional point cloud data onto a planar image, and construct the mapping relationship of the three-dimensional point cloud model according to the pixel coordinates in the planar image. The specific implementation is as follows:

[0104]

[0105] Among them, (u, v) represents the pixel coordinates projected onto the plane, and the color information corresponding to the pixel points in each texture image can be accessed according to the coordinates. (X c , Y c , Z c ) represents the converted camera coordinates, f x , f y represents the focal length of the camera, and (c x , c y ) represents the principal point coordinates.

[0106] It should be noted that the mapping relationship for constructing the three-dimensional point cloud model is established by taking the point features as the homologous features of the texture image and the point cloud model, and the mapping relationship between the texture image and the three-dimensional point cloud model is established. Taking the point features as the homologous features of the texture image and the point cloud model uses the nearest neighbor search method to find the three-dimensional points corresponding to the feature points in the point cloud model. The specific implementation is as follows:

[0107] When the pixel coordinates in the texture image are (u, v), and the three-dimensional point coordinates in the point cloud model are , the corresponding three-dimensional point is found by calculating the Euclidean distance. The specific calculation formula is as follows:

[0108]

[0109] where (u, v) represents the pixel coordinates in the texture image, and (u i , v i ) represents the pixel coordinates of the three-dimensional point in the point cloud model projected onto the image plane. By calculating the Euclidean distance of all points in the current area until the Euclidean distance satisfies the formula D ≤D i-1 ≤D i ≤D i+1 , it means that the three-dimensional point in the current point cloud model is the point corresponding to the feature point, and thus the establishment of the mapping relationship between the pixel coordinates in the texture image and the three-dimensional point coordinates in the point cloud model is completed.

[0110] S4: Based on the mapping relationship between the texture image and the three-dimensional point cloud model, complete the construction of the three-dimensional model of the hydropower station.

[0111] Specifically, the construction of the three-dimensional model of the hydropower station is based on the established mapping relationship to perform the fusion of the texture image and the three-dimensional point cloud model, and then complete the construction of the three-dimensional model of the hydropower station. The fusion of the texture image and the three-dimensional point cloud model calculates the point cloud fitting error corresponding to the mapping relationship according to the established mapping relationship until the point cloud fitting error is the same as the Euclidean distance formula corresponding to the mapping relationship. The point cloud fitting error is the distance from the feature point to the fitting equation. The specific calculation formula is as follows:

[0112]

[0113] where N represents the total number of points in the point cloud data, (x i , y i , z i ) represents the pixel coordinates of the i-th point in the point cloud dataset, (x c , y c , z c ) represents the center coordinates of the fitted sphere, r represents the radius of the fitted sphere, and D′ represents the calculated fitting error;

[0114] Furthermore, the fusion of the texture image and the three-dimensional point cloud model is based on the calculated fitting error, and the specific fusion rules are as follows:

[0115] When the calculated fitting error D′ and the Euclidean distance D satisfy the formula D - D′ ≤ 0, it indicates that the current feature point meets the fusion requirements, and the current texture image and the three-dimensional point cloud model are automatically fused;

[0116] When the calculated fitting error D′ and the Euclidean distance D satisfy the formula D - D′ > 0, it indicates that the current feature point does not meet the fusion requirements, and the establishment of the mapping relationship between the pixel coordinates in the texture image and the three-dimensional point coordinates in the point cloud model continues until the calculated fitting error D′ and the Euclidean distance D satisfy the formula D - D′ ≤ 0.

[0117] It should be noted that after the fusion of the texture image and the three-dimensional point cloud model is completed, the coordinates of each corresponding feature point are remapped, and the finally mapped set is exported for the real scene, thus completing the establishment of the macro and meso real scene three-dimensional model of the hydropower station.

[0118] Embodiment 2

[0119] The second embodiment of the present invention provides a hydropower station real scene three-dimensional model construction system based on texture mapping technology, including a data acquisition module, a point cloud processing module, a three-dimensional model construction module, a texture mapping module, and a hydropower station three-dimensional model output module;

[0120] Specifically, the data acquisition module uses a drone to carry camera equipment to obtain image data of the hydropower station and its surrounding areas through aerial photography. At the same time, combined with the GPS-RTK technology, control point data is accurately collected. The data acquisition module ensures the acquisition of high-quality point cloud data, providing a basis for subsequent model construction;

[0121] The point cloud processing module is responsible for preprocessing the collected point cloud data, including statistical filtering and radius filtering, to eliminate noise points and ensure the authenticity and accuracy of the data. In addition, the module also includes data segmentation and classification functions, which segment the point cloud data into different parts and classify each part to facilitate subsequent three-dimensional model construction;

[0122] The three-dimensional model construction module constructs a three-dimensional point cloud model of the hydropower station based on the processed point cloud data, and uses the coordinate information in the point cloud data to generate the structure and form of the three-dimensional model through algorithms. At the same time, the three-dimensional model construction module also has a model optimization function, which can perform detail adjustment and optimization on the constructed three-dimensional model;

[0123] The texture mapping module is responsible for mapping the collected texture images onto the three-dimensional point cloud model. By calculating the mapping relationship between the texture images and the three-dimensional model, the texture information is accurately applied to the model surface, making the model have a more realistic appearance.

[0124] The three-dimensional model output module of the hydropower station outputs the completed three-dimensional real-scene model of the hydropower station in an appropriate format. Users can select the output format according to their needs, such as STL, OBJ, etc., for subsequent model viewing, analysis, and application.

[0125] It should be noted that the purpose of this system is to use texture mapping technology to construct a three-dimensional real-scene model of a hydropower station. The system obtains the point cloud data of the hydropower station through an unmanned aerial vehicle, combines texture mapping technology, and generates a three-dimensional model with a sense of reality.

[0126] The entire system is divided into the following core modules: data acquisition module, point cloud processing module, three-dimensional model construction module, texture mapping module, and model output module; each module is closely related to each other, and data flows and transforms between each module.

[0127] The data acquisition module provides the original data for the point cloud processing module. The data processed by the point cloud processing module is used as the input of the three-dimensional model construction module, and the model generated by the three-dimensional model construction module is used as the basis for the texture mapping module. Finally, through the model output module, users can obtain a complete three-dimensional real-scene model of the hydropower station.

[0128] This system realizes a method for constructing a three-dimensional real-scene model of a hydropower station based on texture mapping technology by integrating core modules such as data acquisition, point cloud processing, three-dimensional model construction, texture mapping, and model output. Each module of the system is closely related and jointly completes the construction work of the three-dimensional real-scene model of the hydropower station, providing an effective tool for the three-dimensional visualization of the hydropower station.

[0129] Furthermore, if the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. And the aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0130] The logic and / or steps represented in the flowchart or otherwise described herein can, for example, be considered a definitional sequence of executable instructions for implementing logical functions, which can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.

[0131] More specific examples (a non-exhaustive list) of the computer-readable medium include the following: an electrical connection (electronic device) having one or more wirings, a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable medium on which the program can be printed, as the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpretation, or otherwise processing as appropriate, and then storing it in a computer memory.

[0132] Furthermore, to provide a concise description of the exemplary embodiments, all features of the actual embodiments may not be described (i.e., those features that are not relevant to the currently contemplated best mode of carrying out the invention or those features that are not relevant to the implementation of the invention).

[0133] It should be understood that in the development of any actual implementation, as in any engineering or design project, numerous specific implementation decisions may be made. Such development efforts may be complex and time-consuming, but for those of ordinary skill in the art who benefit from this disclosure, without undue experimentation, such development efforts will be a routine task of design, fabrication, and production.

[0134] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A method for constructing a three-dimensional real-scene model of a hydropower station based on texture mapping technology, characterized in that: Including the following steps, Using wireless technology to obtain cloud data of hydropower stations; Constructing a three-dimensional point cloud model of a hydropower station based on the collected point cloud data; Constructing the mapping relationship between the texture image and the three-dimensional point cloud model; Based on the mapping relationship between the texture image and the three-dimensional point cloud model, completing the construction of the three-dimensional model of the hydropower station.

2. The method for constructing a three-dimensional real-scene model of a hydropower station based on texture mapping technology according to claim 1, characterized in that: The use of wireless technology to obtain cloud data of hydropower stations is to use a drone equipped with a camera device, and obtain image data of the hydropower station and its surrounding area through aerial photography, so as to obtain the positioning and attitude data of the hydropower station. Furthermore, the GPS-RTK technology is used to collect control point data. The obtaining of image data of the hydropower station and its surrounding area through aerial photography is to obtain image data of the hydropower station and its surrounding area by using a drone equipped with a line array camera and a laser scanner. The obtaining of the positioning and attitude data of the hydropower station is to match the control points in the image data with the control points in the real geographical coordinates, and calculate the positioning and attitude parameters of the image according to the geographical coordinates of the known control points and the corresponding pixel coordinates in the image. The specific calculation is as follows: The known position coordinates of the line array camera carried by the UAV are C(X, Y, Z), and the attitude coordinates are Moreover, the position coordinates and attitude coordinates of the camera are extracted according to the rotation matrix R and the translation matrix T, and then the acquisition of the positioning and attitude data of the hydropower station is completed. The specific calculation formula is as follows: P geo = RP img + T Where, R represents a rotation matrix, T represents a translation matrix, and P img is the pixel point coordinate in the image data, and P geo represents the positioning and pose parameters calculated based on the image data.

3. The method for constructing a three-dimensional real scene model of a hydropower station based on texture mapping technology according to claim 2, wherein: The construction of a three-dimensional point cloud model of a hydropower station based on the collected point cloud data is to construct a three-dimensional point cloud model of a hydropower station based on the point cloud data. Before constructing the three-dimensional point cloud model of the hydropower station, data processing is performed on the collected point cloud data, and a three-dimensional point cloud model of the hydropower station is constructed according to the data processing results. The data processing of the collected point cloud data includes data preprocessing and data segmentation and classification. The data preprocessing is to perform data preprocessing on the collected point cloud data to ensure the authenticity and accuracy of the data, including statistical filtering and radius filtering. The statistical filtering is to process based on the statistical characteristics of the neighborhood around the point cloud data points. For each data point, calculate the statistical information of the surrounding points, and compare the calculation results with the set threshold, and then complete the elimination of noise points. The specific implementation formula is as follows: α = μ + 2σ Among them, μ represents the average value of the points in the neighborhood, σ represents the standard deviation of the points in the neighborhood, x i represents the value of each point in the neighborhood, and α represents a set threshold for determining the elimination of noise points.

4. The method for constructing a three-dimensional real scene model of a hydropower station based on texture mapping technology according to claim 3, wherein: The judgment of the elimination of noise points is automatically judged according to the set judgment rules. The specific judgment rules are as follows: When the value of each point in the neighborhood satisfies the formula α - 1 ≤ x i ≤ α + 1 when compared with the set threshold, it indicates that the current point in the neighborhood is not a noise point, and the current data point is retained; When the value of each point in the neighborhood is compared with the set threshold and satisfies the formula α - 1 > x i it indicates that the current point in the neighborhood is a noise point, and the current data point is removed; When the value of each point in the neighborhood is compared with the set threshold and satisfies the formula x i > α + 1, it indicates that the current point in the neighborhood is a noise point, and the current data point is removed; The radius filtering is to use the radius r as the set threshold, and then complete the judgment of noise points. The specific judgment rules are as follows: When the data point p j and the center point p i The comparison result between the distance and the radius r satisfies the formula |P j -P i | > R, it means that the current data point is a noise point, and the current data point is removed; When the data point p j and the center point p i The comparison result of the distance between and the radius r satisfies the formula |P j -P i | ≤ R, it means that the current data point is not a noise point, and the current data point is retained.

5. The method for constructing a three-dimensional real-scene model of a hydropower station based on texture mapping technology according to claim 4, wherein: The data segmentation and classification is to segment the collected cloud data of the hydropower station into different parts and classify each part. The specific calculation formula is as follows: Among them, f(x) represents the objective function, which represents the sum of the distances between the sample points in all categories and their respective cluster centers. k represents the total number of clusters preset in advance, and C i represents all the sample points in the i-th cluster, and μ i represents the center point of the i-th cluster; The classification of each part is based on the calculation results of the objective function. Specifically, it is as follows: For each sample point x i , by calculating its distance from each cluster center μ i , and based on the calculated distances, each sample point is assigned to the category corresponding to the nearest cluster center. For each category, their cluster center μ j is recalculated, and the process of sample point assignment and cluster center update is repeated until all sample points are assigned, thus completing the segmentation and classification of the data.

6. The method for constructing a three-dimensional real-scene model of a hydropower station based on texture mapping technology according to claim 5, wherein: The construction of the mapping relationship between the texture image and the three-dimensional point cloud model is to project the collected three-dimensional point cloud data onto a plane image, and construct the mapping relationship of the three-dimensional point cloud model according to the pixel coordinates in the plane image. The specific implementation is as follows: Among them, (u, v) represents the pixel coordinates projected onto the plane, and the color information corresponding to the pixel points in each texture image can be accessed according to the coordinates. (X c , Y c , Z c ) represents the transformed camera coordinates, f x , f y represents the focal lengths of the camera, and (c x , c y ) represents the principal point coordinates.

7. The method for constructing a three-dimensional real-scene model of a hydropower station based on texture mapping technology according to claim 6, wherein: The construction of the mapping relationship of the three-dimensional point cloud model is to establish the mapping relationship between the texture image and the three-dimensional point cloud model by using the point feature as the homologous feature of the texture image and the point cloud model. The use of the point feature as the homologous feature of the texture image and the point cloud model is to use the nearest neighbor search method to find the three-dimensional points corresponding to the feature points in the point cloud model. The specific implementation is as follows: When the pixel coordinates in the texture image are (u, v) and the three-dimensional point coordinates in the point cloud model are At this time, the corresponding three-dimensional point is found by calculating the Euclidean distance. The specific calculation formula is as follows: Among them, (u, v) represents the pixel coordinates in the texture image, and (u i , v i ) represents the three-dimensional points in the point cloud model projected onto the pixel coordinates in the image plane. By calculating the Euclidean distance of all points within the current region until the Euclidean distance satisfies the formula D i-1 ≤ D i ≤ D i+1 , it indicates that the three-dimensional point in the current point cloud model is the point corresponding to the feature point, thereby completing the establishment of the mapping relationship between the pixel coordinates in the texture image and the three-dimensional point coordinates in the point cloud model.

8. A method for constructing a three-dimensional real-scene model of a hydropower station using the texture mapping technology as described in any one of claims 7, characterized in that: The construction of the 3D model of the hydropower station is based on the established mapping relationship to fuse the texture image and the 3D point cloud model, and then complete the construction of the 3D model of the hydropower station; the fusion of the texture image and the 3D point cloud model is to calculate the point cloud fitting error corresponding to the mapping relationship according to the established mapping relationship until the point cloud fitting error is the same as the Euclidean distance formula corresponding to the mapping relationship. The point cloud fitting error is to calculate the distance from the feature point to the fitting equation, and the specific calculation formula is as follows: Among them, N represents the total number of points in the point cloud data, (x i , y i , z i ) represents the pixel coordinates of the i-th point in the point cloud dataset, (x c , y c , z c ) represents the coordinates of the center of the sphere obtained by fitting, r represents the radius of the sphere obtained by fitting, and D′ represents the calculated fitting error; The fusion of the texture image and the 3D point cloud model is based on the calculated fitting error, and the specific fusion rules are as follows: When the calculated fitting error D′ and the Euclidean distance D satisfy the formula D - D′ ≤ 0, it means that the current feature point meets the fusion requirements, and the current texture image and the 3D point cloud model are automatically fused; When the calculated fitting error D′ and the Euclidean distance D satisfy the formula D - D′ > 0, it means that the current feature point does not meet the fusion requirements, and continue to execute the establishment of the mapping relationship between the pixel coordinates in the texture image and the 3D point coordinates in the point cloud model until the calculated fitting error D′ and the Euclidean distance D satisfy the formula D - D′ ≤ 0.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it realizes the steps of the method described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it realizes the steps of the method described in any one of claims 1 to 7.

Images