High-precision transformer substation equipment three-dimensional modeling method and platform

Abstract

The invention discloses a high-precision transformer substation equipment three-dimensional modeling method and platform, and relates to the technical field of data processing, and the method comprises the steps: obtaining an equipment image collection result and an equipment point cloud collection result of transformer substation equipment; obtaining an equipment image data stream and an equipment point cloud data stream; calculating first Gaussian weight distribution according to the Gaussian weight analysis model; obtaining a first model of the substation equipment; obtaining an equipment feature guide vector set; obtaining a second model of the substation equipment; and performing UV expansion correction on the second transformer substation equipment model according to the equipment image data stream and the equipment point cloud data stream to generate a third transformer substation equipment model. The technical problems that in the prior art, the point cloud data and image data fusion precision is low, the modeling process is complex, and consequently the equipment three-dimensional modeling precision is insufficient, and the modeling efficiency is poor are solved, and the technical effect of improving the substation equipment three-dimensional modeling precision and modeling efficiency is achieved.

Description

Technical Field

[0001] This application relates to the technical field of data processing, and particularly to a high-precision three-dimensional modeling method and platform for substation equipment. Background Art

[0002] The three-dimensional modeling of substation equipment refers to obtaining the spatial geometric information of substation equipment through high-precision measurement means and generating its digital three-dimensional model in a computer. These three-dimensional models can not only accurately represent the geometric shape of the equipment but also assist in realizing the full life cycle management of the equipment, including equipment installation, operation, maintenance, and overhaul, etc. Common three-dimensional modeling methods include lidar measurement, photogrammetry, etc. Among them, lidar (LiDAR) technology can obtain high-precision three-dimensional point cloud data in space by emitting laser and measuring the time difference of the laser reflected back; camera imaging technology can provide rich visual information by acquiring high-definition images of the equipment. However, in practical applications, there are still many challenges in how to efficiently and accurately combine lidar point cloud data with camera image data to generate high-precision three-dimensional models of substation equipment, including how to perform efficient filtering and smoothing processing on point cloud data to reduce noise and errors, and how to effectively register image data with point cloud data to enhance the detailed texture of the equipment model.

[0003] Therefore, in the current related technologies, there are technical problems such as low fusion accuracy of point cloud data and image data and complex modeling processes, which in turn lead to insufficient accuracy and poor modeling efficiency of equipment three-dimensional modeling. Summary of the Invention

[0004] This application provides a high-precision three-dimensional modeling method and platform for substation equipment, solves the technical problems in the prior art such as low fusion accuracy of point cloud data and image data and complex modeling processes, which in turn lead to insufficient accuracy and poor modeling efficiency of equipment three-dimensional modeling, and achieves the technical effect of improving the accuracy and modeling efficiency of substation equipment three-dimensional modeling.

[0005] The present application provides a three-dimensional modeling method for substation equipment with high precision. The method includes: obtaining the equipment image acquisition result and the equipment point cloud acquisition result of the substation equipment according to a camera and a lidar device; inputting the equipment image acquisition result into an image filter to obtain an equipment image data stream, and inputting the equipment point cloud acquisition result into a point cloud filter to obtain an equipment point cloud data stream; based on the equipment point cloud data stream, calculating a first Gaussian weight distribution according to a Gaussian weight analysis model; performing smooth modeling on the equipment point cloud data stream according to the first Gaussian weight distribution to obtain a first model of the substation equipment; performing SIFT feature registration according to the equipment image data stream and the equipment point cloud data stream to obtain a set of equipment feature guiding vectors; based on the set of equipment feature guiding vectors, guiding the equipment image data stream to perform detail texture enhancement on the first model of the substation equipment to obtain a second model of the substation equipment; and performing UV unfolding correction on the second model of the substation equipment according to the equipment image data stream and the equipment point cloud data stream to generate a third model of the substation equipment.

[0006] In a possible implementation manner, the three-dimensional modeling method for substation equipment with high precision further performs the following processing: dividing the equipment point cloud data stream into neighborhoods of each point according to a predetermined neighborhood radius to obtain neighborhoods of each point of the equipment; calculating the centroid of each neighborhood according to the neighborhoods of each point of the equipment to obtain the centroid of each neighborhood; calculating the covariance matrix of each point according to the neighborhoods of each point of the equipment to obtain the covariance matrix of each point; inputting the centroid of each neighborhood and the covariance matrix of each point into the Gaussian weight analysis model to output a multi-point Gaussian weight; and outputting the multi-point Gaussian weight as the first Gaussian weight distribution.

[0007] In a possible implementation manner, the three-dimensional modeling method for substation equipment with high precision further performs the following processing: the Gaussian weight analysis model includes a Gaussian weight analysis function, and the Gaussian weight analysis function is: ; wherein, represents the Gaussian weight of the i-th point, i represents the i-th point in the equipment point cloud data stream, and the i-th point is any point in the equipment point cloud data stream, represents the covariance matrix of the i-th point, represents the determinant of the covariance matrix of the i-th point, represents the inverse matrix of the covariance matrix of the i-th point, represents the exponential function with the natural constant e as the base, represents the centroid of the neighborhood of the i-th point, represents the deviation vector between the i-th point and the centroid of the neighborhood of the i-th point, represents the transposed vector of the deviation vector between the i-th point and the centroid of the neighborhood of the i-th point.

[0008] In a possible implementation, the high-precision three-dimensional modeling method for substation equipment further performs the following processing: perform normalization calculation according to the first Gaussian weight distribution to obtain a second Gaussian weight distribution; perform Gaussian weighted smoothing optimization on the equipment point cloud data stream according to the second Gaussian weight distribution to obtain a set of smoothed equipment point clouds; perform three-dimensional modeling according to the set of smoothed equipment point clouds to obtain the first model of the substation equipment.

[0009] In a possible implementation, the high-precision three-dimensional modeling method for substation equipment further performs the following processing: perform SIFT feature detection on the equipment image data stream to obtain a set of equipment image feature vectors; perform SIFT feature detection on the equipment point cloud data stream to obtain a set of equipment point cloud feature vectors; perform matching cost recognition on the set of equipment point cloud feature vectors respectively according to each equipment image feature vector in the set of equipment image feature vectors to obtain multiple sets of matching cost coefficients; perform minimum matching cost screening according to the multiple sets of matching cost coefficients to determine an optimal matching cost distribution; perform registration association on the set of equipment image feature vectors and the set of equipment point cloud feature vectors according to the optimal matching cost distribution to generate a set of equipment feature guiding vectors.

[0010] In a possible implementation, the high-precision three-dimensional modeling method for substation equipment further performs the following processing: project the equipment image data stream onto the first model of the substation equipment based on the set of equipment feature guiding vectors to obtain a set of image projection points; perform detailed texture optimization on the first model of the substation equipment according to the set of image projection points to obtain an optimized model of the substation equipment; perform an evaluation of the smoothness of the detailed texture transition on the optimized model of the substation equipment to obtain a model transition smoothness coefficient; determine whether the model transition smoothness coefficient is less than a predetermined transition smoothness coefficient; if the model transition smoothness coefficient is less than the predetermined transition smoothness coefficient, perform transition smoothness optimization on the optimized model of the substation equipment according to the predetermined transition smoothness coefficient to generate the second model of the substation equipment.

[0011] In a possible implementation, the high-precision three-dimensional modeling method for substation equipment further performs the following processing: perform UV unwrapping on the second model of the substation equipment to obtain an unfolded diagram of the equipment model; construct a detailed texture map of the equipment according to the equipment image data stream; draw a geometric feature map of the equipment according to the equipment point cloud data stream; perform deviation detection on the unfolded diagram of the equipment model according to the detailed texture map of the equipment and the geometric feature map of the equipment to obtain a deviation detection result of the equipment model; correct the second model of the substation equipment according to the deviation detection result of the equipment model to obtain the third model of the substation equipment.

[0012] The present application also provides a high-precision three-dimensional modeling platform for substation equipment, including: a collection result obtaining module, configured to obtain the equipment image collection result and the equipment point cloud collection result of the substation equipment according to cameras and lidar devices; a data stream obtaining module, configured to input the equipment image collection result into an image filter to obtain an equipment image data stream, and input the equipment point cloud collection result into a point cloud filter to obtain an equipment point cloud data stream; a first Gaussian weight distribution calculation module, configured to calculate a first Gaussian weight distribution based on the equipment point cloud data stream according to a Gaussian weight analysis model; a first substation equipment model obtaining module, configured to perform smooth modeling on the equipment point cloud data stream according to the first Gaussian weight distribution to obtain a first substation equipment model; an equipment feature guiding vector set obtaining module, configured to perform SIFT feature registration according to the equipment image data stream and the equipment point cloud data stream to obtain an equipment feature guiding vector set; a second substation equipment model obtaining module, configured to guide the equipment image data stream to perform detail texture enhancement on the first substation equipment model based on the equipment feature guiding vector set to obtain a second substation equipment model; and a third substation equipment model generating module, configured to perform UV unwrapping correction on the second substation equipment model according to the equipment image data stream and the equipment point cloud data stream to generate a third substation equipment model.

[0013] It is intended to obtain the equipment image collection result and the equipment point cloud collection result of the substation equipment through a high-precision three-dimensional modeling method and platform proposed in the present application; obtain the equipment image data stream and the equipment point cloud data stream; calculate a first Gaussian weight distribution according to a Gaussian weight analysis model; obtain a first substation equipment model; obtain an equipment feature guiding vector set; obtain a second substation equipment model; and generate a third substation equipment model. This solves the technical problems in the prior art, such as the low fusion accuracy of point cloud data and image data and the complex modeling process, which lead to insufficient accuracy and poor modeling efficiency of equipment three-dimensional modeling, and achieves the technical effect of improving the accuracy and modeling efficiency of substation equipment three-dimensional modeling. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments of the present disclosure will be briefly introduced below. Flowcharts are used in the present application to illustrate the operations performed by the platform according to the embodiments of the present application. It should be understood that the operations before or below do not necessarily need to be executed precisely in sequence. On the contrary, according to needs, they can be executed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or several operations can be removed from these processes.

[0015] Figure 1 It is a schematic flowchart of a high-precision three-dimensional modeling method for substation equipment provided by an embodiment of the present application.

[0016] Figure 2 Schematic diagram of the structure of a high-precision 3D modeling platform for substation equipment provided by an embodiment of the present application.

[0017] Explanation of reference numerals in the drawings: Acquisition result obtaining module 10, data stream obtaining module 20, first Gaussian weight distribution calculation module 30, first substation equipment model obtaining module 40, equipment feature guiding vector set obtaining module 50, second substation equipment model obtaining module 60, third substation equipment model generation module 70. Detailed implementation manners

[0018] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the following specifically gives the detailed implementation manners of the present application.

[0019] In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings. The described embodiments should not be regarded as limitations of the present application. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present application.

[0020] In the following description, "some embodiments" are involved, which describe a subset of all possible embodiments. However, it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict. The terms "first\second" involved are only used to distinguish similar objects and do not represent a specific order for the objects. The terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, platform, product or server comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or modules not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0021] An embodiment of the present application provides a high-precision 3D modeling method for substation equipment, as Figure 1 shown, the method includes: Step S100, obtaining the equipment image acquisition result and the equipment point cloud acquisition result of the substation equipment according to the camera and the lidar device.

[0022] Preferably, image data of substation equipment is acquired through camera devices (such as high-definition cameras, binocular high-definition cameras, etc.) as the equipment image acquisition result of substation equipment, which usually includes information such as the appearance, surface features, color, and texture of the equipment. The camera can capture two-dimensional image data of the equipment according to different shooting angles, focal lengths, and lighting conditions for subsequent processing such as detailed texture enhancement, image filtering, and feature matching. Three-dimensional point cloud data of substation equipment is acquired through a Light Detection and Ranging (LiDAR) device as the equipment point cloud acquisition result of substation equipment. Specifically, the LiDAR emits laser beams and accurately measures the distance to each point based on the time when the laser is reflected from the surface of the object, and constructs a three-dimensional data set composed of a large number of spatial points. The point cloud data represents the three-dimensional geometric shape of the equipment, such as position, shape, surface contour, etc. The point cloud data acquired through the LiDAR usually has high precision and can accurately represent the three-dimensional structure of the equipment. The equipment image provides rich visual information, which is helpful for subsequent texture enhancement and detail processing, while the point cloud data collected by the LiDAR provides precise spatial geometric structure information for the model. Through the fusion of these two types of data, a more accurate and delicate three-dimensional model of substation equipment can be generated.

[0023] Step S200: Input the equipment image acquisition result into an image filter to obtain an equipment image data stream, and input the equipment point cloud acquisition result into a point cloud filter to obtain an equipment point cloud data stream.

[0024] Preferably, input the equipment image acquisition result into an image filter for processing, that is, use the image filter to perform denoising, contrast enhancement, brightness adjustment, smoothing, edge sharpening, etc. on the image data to improve the image quality and remove noise. After being processed by the image filter, the equipment image acquisition result outputs an optimized, denoised, and smoothed image data stream, that is, an equipment image data stream, which can be smoothly transmitted to subsequent processing modules (such as feature extraction, texture enhancement, etc.). Input the equipment point cloud acquisition result into a point cloud filter for processing. The point cloud filter is mainly used to remove noise points, fill in missing data, reduce outliers, etc., making the point cloud data more accurate and complete. The point cloud filter can also perform other operations, such as sparsification (reducing the point cloud density) or resampling. Specifically, after being processed by the point cloud filter, the equipment point cloud acquisition result outputs an optimized and denoised point cloud data stream, that is, an equipment point cloud data stream. The equipment point cloud data stream refers to the processed point cloud data. After being filtered, these data can continue to be transmitted to subsequent processing stages (such as modeling, feature extraction, etc.) and ensure the accuracy of the generation of the three-dimensional model of substation equipment.

[0025] Step S300: Based on the equipment point cloud data stream, calculate the first Gaussian weight distribution according to the Gaussian weight analysis model.

[0026] Preferably, the device point cloud data stream includes a three-dimensional point cloud data stream of the spatial information of substation devices. The position of each point in space is represented by its coordinates (such as X, Y, Z), and may also include other information, such as intensity values, color information, etc., to represent the geometric shape and spatial structure of the devices. The first Gaussian weight distribution is calculated using the device point cloud data and the Gaussian weight analysis model. Among them, the Gaussian weight analysis model is a mathematical model used to assign a weight to each point in the point cloud data based on the Gaussian distribution, representing the importance of the point in space or its similarity to surrounding points. The Gaussian distribution is used to describe the distribution of data points under certain conditions. The Gaussian weight usually depends on the distance between points in space and other relevant factors, such as the spatial distance between points, the density of points, and the local shape of points. By applying the Gaussian weight analysis model to the point cloud data, a weighted value can be calculated for each point in the point cloud. Specifically, each point cloud data point calculates a weight value (usually the weight of the central point is large, and the weight of the points far from the central point is small) according to its spatial position (such as the distance from neighboring points) and other characteristics (such as the density of points), and is optimized according to the geometric shape and structure of the device. For example, if some points are close to the object surface or key feature areas, they are given greater weights, and then the first Gaussian weight distribution is generated through a mathematical formula (Gaussian function) to help eliminate noise and better reflect the true shape of the object surface, so that the point cloud data is more accurate, and the model generated after smooth modeling can better reflect the actual shape and details of substation devices.

[0027] Further, step S300 further includes step S310 of dividing the neighborhood of each point of the device point cloud data stream according to a predetermined neighborhood radius to obtain the neighborhood of each point of the device; step S320 of calculating the neighborhood centroid according to the neighborhood of each point of the device to obtain the neighborhood centroid of each point; step S330 of calculating the covariance matrix according to the neighborhood of each point of the device to obtain the covariance matrix of each point; step S340 of inputting the neighborhood centroid of each point and the covariance matrix of each point into the Gaussian weight analysis model to output the multi-point Gaussian weight; step S350 of outputting the multi-point Gaussian weight as the first Gaussian weight distribution.

[0028] Preferably, the neighborhood radius is a predefined parameter representing the set of points within a certain range around each point. In point cloud processing, a radius range is usually selected to determine the neighborhood of each point. The neighborhood radius can be adjusted based on factors such as the scale of the device and the resolution of the point cloud. Specifically, according to the predefined neighborhood radius, the point cloud data is partitioned into neighborhoods, that is, all other points within the radius around each point in the point cloud data are identified to form the neighborhood of that point (i.e., the spatially adjacent points of that point, which constitute the local area of that point); then the centroid of each point's neighborhood on the device is calculated, that is, the coordinates of all points within the neighborhood are averaged to calculate the geometric center (centroid) of the neighborhood, and thus the centroid of each point's neighborhood is obtained, representing the spatial center of the neighborhood.

[0029] Preferably, the covariance matrix of each point's neighborhood on the device is calculated. Among them, the covariance matrix is a mathematical tool for describing the spatial relationship between data points, used to represent the spatial distribution and its directionality of all points within a certain neighborhood in the point cloud. Specifically, for any point in the point cloud data and all points within its neighborhood, the difference between each point and the centroid is calculated based on the centroid of the neighborhood of that point, that is, the deviation of each point from the centroid on each coordinate axis is calculated. Using these differences, the elements of the covariance matrix are calculated, and finally the covariance matrix is obtained, including variance (diagonal elements) and covariance (non-diagonal elements). Among them, the variance represents the degree of expansion or distribution width of the neighborhood points in the x, y, and z directions. If the variance in a certain direction is large, it means that the points are more dispersed in that direction; if the variance is small, it means that the points are more concentrated in that direction; the covariance represents the correlation between different directions. If the covariance is large, it means that the points tend to be consistent in these two directions; if the covariance is negative, it means that they show an opposite trend in these two directions.

[0030] Preferably, the centroid and covariance matrix of each point are used as inputs and fed into the Gaussian weight analysis model. The Gaussian weight analysis model calculates the Gaussian weight of that point based on the neighborhood centroid and covariance matrix through the Gaussian distribution function, that is, by considering the geometric center and spatial distribution of the neighborhood points, the weight of each point is calculated. These weights represent the importance or reliability of the points. The Gaussian weight usually assigns a higher weight to the points close to the centroid and a lower weight to the points far from the centroid; then the Gaussian weights of all points are integrated to form a Gaussian weight distribution, that is, the first Gaussian weight distribution. The Gaussian weight distribution is a weighted representation of the entire point cloud data, where the weight of each point reflects the importance of that point within the local area. In this way, the more important points in the point cloud data (such as the edges of the device and key information points) will obtain larger weights, while irrelevant or noisy points will be assigned smaller weights. By applying the Gaussian weights, the model can better reflect the geometric shape and structure of the device, reduce noise and errors, and thus improve the accuracy of 3D modeling.

[0031] Further, step S300 further includes that the Gaussian weight analysis model includes a Gaussian weight analysis function, and the Gaussian weight analysis function is: ; where represents the Gaussian weight at point i, i represents the i-th point in the device point cloud data stream, and the i-th point is any point in the device point cloud data stream, represents the covariance matrix at point i, represents the determinant of the covariance matrix at point i, represents the inverse matrix of the covariance matrix at point i, represents the exponential function with the natural constant e as the base, represents the centroid of the neighborhood of point i, represents the deviation vector between point i and the centroid of its neighborhood, represents the transposed vector of the deviation vector between point i and the centroid of its neighborhood.

[0032] Preferably, the Gaussian weight analysis function , is a mathematical function used in point cloud data processing to calculate the weight of each point through Gaussian distribution, that is, according to the relationship between a point and its neighborhood, corresponding weights are assigned to the points. Among them, the covariance matrix describes the spatial distribution of the point cloud in the neighborhood, not only representing the variance of the points on each coordinate axis, but also including the correlation between points (i.e., the covariance of the points); the determinant of the covariance matrix provides the "volume" information of the spatial distribution of the neighborhood points. A larger determinant indicates that the points are more widely distributed, while a smaller determinant indicates that the points are more concentrated; the inverse matrix of the covariance matrix is used to adjust the relative position of each point in the point cloud data in space, so that the points far from the centroid in the point cloud will be given lower weights; is used to define the attenuation characteristics of the Gaussian distribution. The shape of the Gaussian distribution is bell-shaped, with higher weights for points closer to the centroid and lower weights for points farther from the centroid; the transposed vector of the deviation vector , the transpose operation enables the deviation vector to perform matrix multiplication with the inverse matrix of the covariance matrix.

[0033] Step S400, perform smoothing modeling on the device point cloud data stream according to the first Gaussian weight distribution to obtain the first model of the substation equipment.

[0034] Step S400 further includes step S410 of performing normalization calculation according to the first Gaussian weight distribution to obtain a second Gaussian weight distribution; step S420 of performing Gaussian weighted smoothing optimization on the device point cloud data stream according to the second Gaussian weight distribution to obtain a set of device smoothed point clouds; and step S430 of performing three-dimensional modeling according to the set of device smoothed point clouds to obtain the first substation device model.

[0035] Preferably, smoothing modeling is performed on the device point cloud data stream according to the first Gaussian weight distribution (the weight distribution of each point calculated by the Gaussian weight analysis model), that is, a smoothing algorithm is applied to process the device point cloud data to reduce noise and errors in the point cloud, making the device model smoother and more continuous. Usually, it involves performing weighted smoothing processing on each point in the point cloud, so that the coordinates of each point are affected by the coordinates of other points in the neighborhood during calculation, thereby reducing the abnormal influence of a single point. Specifically, normalization calculation is performed according to the first Gaussian weight distribution, that is, the first Gaussian weight distribution is adjusted to a standard range (usually between 0 and 1) to obtain a second Gaussian weight distribution, ensuring that the weights of all points add up to 1, avoiding deviations in the modeling results caused by some points having too large or too small weights, and making the weights of all points in the point cloud balanced; then, the second Gaussian weight distribution is used to perform weighted smoothing processing on the device point cloud data, that is, the weight of each point is adjusted according to the points in its neighborhood and the covariance matrix, further optimizing the smoothing effect of the point cloud data, and then reducing the noise and errors in the point cloud data to generate a set of device smoothed point clouds; finally, three-dimensional modeling is performed according to the set of device smoothed point clouds, that is, by processing the set of device smoothed point clouds, an accurate three-dimensional model is generated as the first substation device model, which can truly reproduce the shape and details of the substation device, represents the geometric shape of the substation device, ensures the accuracy of the model, and at the same time improves the authenticity and visual effect of the model.

[0036] Step S500: Perform SIFT feature registration according to the device image data stream and the device point cloud data stream to obtain a set of device feature guiding vectors.

[0037] Preferably, SIFT feature registration (Scale-Invariant Feature Transform) is a commonly used feature matching method in computer vision, mainly used to extract and match key feature points in images. The purpose is to identify and match local features in images. Even in the case of scale, rotation, perspective, and illumination changes, the features can still remain consistent. SIFT feature registration is performed according to the device image data stream and the device point cloud data stream. Specifically, SIFT is used to extract features from the device image data stream, including detecting key points with high contrast from the image, usually in areas with complex edges or textures in the image, and generating a multi-dimensional vector for each detected key point to represent the local image structure around the key point; extracting features from the device point cloud data stream, including determining key points by detecting sparse or significantly changing areas, such as the corners, edges of the device, or other areas with large geometric changes, and generating descriptors for each key point based on the local geometric shape of the point cloud (such as normal direction, curvature, etc.) to represent the geometric features of each key point in the point cloud; SIFT feature registration is to register the feature points extracted from the device image data stream and the device point cloud data stream, that is, to find the matching relationship between the image features and the point cloud features. That is, using the nearest neighbor matching method, calculate the similarity or distance between each feature point in the two types of data streams, match the key points in the device image with the key points in the device point cloud, find the corresponding relationship between the two, and then perform geometric transformations (such as rotation, translation, etc.) on the matching points to align the key points in the image with the key points in the point cloud, obtaining the matching point pairs between the image and the point cloud, and then generating the device feature guidance vector set, which contains the corresponding relationship between the image features and the point cloud features, usually represented as vectors or feature pairs.

[0038] Furthermore, step S500 further includes step S510, performing SIFT feature detection according to the device image data stream to obtain the device image feature vector set; including step S520, performing SIFT feature detection according to the device point cloud data stream to obtain the device point cloud feature vector set; including step S530, respectively performing matching cost identification on the device point cloud feature vector set according to each device image feature vector in the device image feature vector set to obtain multiple matching cost coefficient sets; including step S540, performing matching cost minimization screening according to the multiple matching cost coefficient sets to determine the optimal matching cost distribution; including step S550, performing registration association on the device image feature vector set and the device point cloud feature vector set according to the optimal matching cost distribution to generate the device feature guidance vector set.

[0039] Preferably, perform SIFT feature detection on the device image data stream to extract feature points in each device image (usually located in areas with obvious edges, corners or textures in the image). Each feature point will have a corresponding descriptor (feature vector) that captures the geometric or texture information of the local area around the point. All feature vectors constitute the device image feature vector set. Perform SIFT feature detection on the device point cloud data stream, that is, by analyzing the key points and local geometric forms in the point cloud data, extract key points from the device point cloud data stream, and generate a feature vector for each point, reflecting the local geometric form and spatial position of each point. All feature vectors constitute the device point cloud feature vector set.

[0040] Preferably, match each image feature in the device image feature vector set with the features in the device point cloud feature vector set, that is, calculate the distance (such as Euclidean distance) between the two to measure their similarity. The smaller the matching cost, the more likely it is that these two feature points are corresponding points at the same position. Each device image feature vector will be matched with multiple feature vectors in the device point cloud feature vector set, and the matching cost (i.e., similarity) will be calculated, and then multiple sets of matching cost coefficients are obtained, indicating the matching degree between the image feature and the point cloud feature. Then use the minimization strategy to analyze all the matching costs of multiple sets of matching cost coefficients, screen the matching pairs with the smallest matching cost, that is, select those feature pairs with the smallest distance or the strongest similarity, determine which image features and point cloud features are the best matches, thus constituting the optimal matching cost distribution. Finally, according to the optimal matching cost distribution, register and associate the features in the device image feature vector set with the features in the device point cloud feature vector set, that is, align the two-dimensional image features and three-dimensional point cloud features to ensure that they correspond to the same position or the same physical structure, and then obtain the device feature guiding vector set, which can effectively associate the local features in the image and point cloud data and provide accurate input for subsequent processing such as 3D modeling and texture mapping.

[0041] Step S600: Based on the device feature guiding vector set, guide the device image data stream to perform detailed texture enhancement on the first substation device model to obtain the second substation device model.

[0042] Preferably, the detailed texture enhancement refers to refining and optimizing the texture of an existing three-dimensional point cloud model through image data. Among them, for the first model of substation equipment initially generated, there may be some problems such as unclear texture, insufficient details, or low surface quality. Specifically, by using the device feature guidance vector set, the texture information (such as color, texture, surface details, etc.) in the device image data stream is mapped into the first model of substation equipment. This can not only improve the presentation of surface details but also ensure the accurate spatial alignment of the image and the three-dimensional model according to the feature vector set, making the texture more conform to the geometric shape of the device. That is, through texture mapping technology, the details in the image data (such as the color, brightness, reflection characteristics, etc. of the object surface) are transferred to the surface of the three-dimensional model, thereby improving the realism and fineness of the model, and then generating the second model of substation equipment, which has richer surface texture and higher visual effects, can more realistically reproduce the appearance and details of substation equipment, and thus improve the quality and effect of modeling.

[0043] Further, step S600 further includes step S610 of projecting the device image data stream onto the first model of substation equipment based on the device feature guidance vector set to obtain an image projection point set; step S620 of optimizing the detailed texture of the first model of substation equipment according to the image projection point set to obtain an optimized model of substation equipment; step S630 of evaluating the smoothness of the detailed texture transition of the optimized model of substation equipment to obtain a model transition smoothness coefficient; step S640 of determining whether the model transition smoothness coefficient is less than a predetermined transition smoothness coefficient; step S650 of, if the model transition smoothness coefficient is less than the predetermined transition smoothness coefficient, performing transition smoothness optimization on the optimized model of substation equipment according to the predetermined transition smoothness coefficient to generate the second model of substation equipment.

[0044] Preferably, the texture and image information in the device image data stream are mapped into the three-dimensional model of substation equipment. Specifically, the image data is "projected" through the device feature guidance vector set, that is, the pixel positions of the image are corresponded to the surface of the three-dimensional model, and each pixel in the image corresponds to a point on the surface of the three-dimensional model, forming an image projection point set that contains all the texture mapping positions; then, the image projection point set is used to optimize the texture of the first model of substation equipment, that is, to determine which places of the device model need to be enhanced in details, such as texture clarity, contrast, color, etc., to make the surface details of the model more real, so that the model is more accurate and delicate visually, and an optimized model of substation equipment is obtained.

[0045] Preferably, a detailed texture transition smoothing evaluation is performed on the optimized model of the substation equipment. Specifically, after texture optimization, the model may have unnatural texture changes in some areas, especially the texture transition between different areas may be uneven. Perform a transition smoothing process on the model to make the transition between different texture areas more natural and smooth, and calculate a coefficient (model transition smoothing coefficient) to measure the texture transition smoothness. If the transition smoothing coefficient is high, it indicates that the texture transition is relatively natural; if the coefficient is low, there may be obvious uneven transition areas. Then, determine whether the model transition smoothing coefficient is less than a predetermined transition smoothing coefficient. Here, the predetermined transition smoothing coefficient is a set transition smoothing standard value. If the model transition smoothing coefficient is greater than or equal to the predetermined transition smoothing coefficient, the model does not need further optimization. On the contrary, if the transition smoothing coefficient is less than the predetermined transition smoothing coefficient, use the predetermined transition smoothing coefficient to perform transition smoothing optimization on the optimized model of the substation equipment, that is, adjust the model through interpolation, gradual change, etc., to make the texture transition between different areas smoother, and generate the final second model of the substation equipment, which can more vividly represent the appearance and details of the substation equipment and has higher quality and credibility in practical applications (such as simulation, analysis, display, etc.).

[0046] Step S700, perform UV unwrapping correction on the second model of the substation equipment according to the device image data stream and the device point cloud data stream to generate a third model of the substation equipment.

[0047] Preferably, UV unwrapping correction unfolds the surface of a three-dimensional object into a two-dimensional plane and maps the texture, thereby ensuring that the texture on the surface of the three-dimensional model can naturally fit the geometric shape of the model. Among them, UV unwrapping is to map the surface of a three-dimensional object onto a two-dimensional plane, and each vertex of the three-dimensional model has a corresponding two-dimensional coordinate (U, V), representing the position of each point on the object surface in the two-dimensional image. Through UV unwrapping, the surface of the three-dimensional model can be "flattened" so that textures (such as texture maps) can be pasted on it. However, this may cause stretching, distortion or unnatural deformation of the texture. The goal of UV unwrapping correction is to correct this irregular texture mapping so that the texture covers the model surface more evenly and naturally. Specifically, unfold the three-dimensional surface of the substation equipment into a two-dimensional plane, and assign a UV coordinate system to each face or local area. By adjusting the UV coordinates, the texture can be correctly and evenly mapped onto the three-dimensional model surface, avoiding obvious stretching or distortion. If there is texture stretching or distortion in some areas of the model, correct it by adjusting the UV coordinates to ensure the smooth transition and consistency of the texture map, ensuring the accuracy and authenticity of the texture, and thus obtaining the final third model of the substation equipment, which has high-quality surface texture mapping and can better display the details and appearance of the substation equipment.

[0048] Furthermore, step S700 further includes step S710 of performing UV unfolding on the second substation equipment model to obtain an unfolded diagram of the equipment model; step S720 of constructing a detailed texture map of the equipment according to the equipment image data stream; step S730 of drawing a geometric feature map of the equipment according to the equipment point cloud data stream; step S740 of performing deviation detection on the unfolded diagram of the equipment model according to the detailed texture map of the equipment and the geometric feature map of the equipment to obtain a deviation detection result of the equipment model; and step S750 of correcting the second substation equipment model according to the deviation detection result of the equipment model to obtain the third substation equipment model.

[0049] Preferably, performing UV unfolding on the second substation equipment model means flattening the surface of the three-dimensional model onto a two-dimensional plane (unfolded diagram) to provide an accurate coordinate system for texture mapping, ensuring that each surface area can correctly represent its position in three-dimensional space in the two-dimensional plane, and obtaining an unfolded diagram of the equipment model, which contains the UV coordinate mapping information of all surface areas; extracting, optimizing, and synthesizing the texture based on the equipment image data stream to obtain a detailed texture map of the equipment, which contains the detailed information of the equipment surface, such as material, color, gloss, etc.; extracting the geometric features of the equipment according to the equipment point cloud data stream, such as the contour, angle, curvature, etc. of the equipment surface, to form a geometric feature map of the equipment; then performing deviation detection on the unfolded diagram of the equipment model, that is, comparing whether the texture mapping and geometric features on the unfolded diagram of the equipment model match the detailed texture map and geometric feature map of the equipment to check whether the surface texture of the equipment model is correctly attached to the geometric shape of the equipment, and further obtaining a deviation detection result of the equipment model, which reflects the differences or mismatches between the equipment surface texture and geometric structure during the texture mapping process. For example, the texture may be stretched, distorted, or misaligned in some areas. The purpose of deviation detection is to locate and quantify these differences to ensure the accuracy of the final model; finally, according to the deviation detection result, correcting the second substation equipment model, including adjusting UV unfolding, realigning the texture, repairing parts with geometric inconsistencies, etc., to ensure that the texture and geometric shape can be seamlessly docked, obtaining the third substation equipment model, which has a significant improvement in visual effects and geometric accuracy, can ensure that the texture and geometric structure are completely matched, there is no texture distortion or geometric shape inconsistency, and more realistically reflects the actual appearance and details of the substation equipment.

[0050] In the foregoing, a high-precision three-dimensional modeling method for substation equipment according to an embodiment of the present invention is described in detail. Next, a high-precision three-dimensional modeling platform for substation equipment according to an embodiment of the present invention will be described with reference to Figure 1 In the foregoing, a high-precision three-dimensional modeling method for substation equipment according to an embodiment of the present invention is described in detail. Next, a high-precision three-dimensional modeling platform for substation equipment according to an embodiment of the present invention will be described with reference to Figure 2 a high-precision three-dimensional modeling platform for substation equipment according to an embodiment of the present invention will be described.

[0051] A three-dimensional modeling platform for substation equipment with high precision according to an embodiment of the present invention is used to solve the technical problems existing in the prior art, such as low fusion accuracy of point cloud data and image data, and complex modeling process, which result in insufficient accuracy and poor modeling efficiency of three-dimensional modeling of equipment. It achieves the technical effect of improving the accuracy and modeling efficiency of three-dimensional modeling of substation equipment. As Figure 2 shown, a three-dimensional modeling platform for substation equipment with high precision includes: an acquisition result obtaining module 10, a data stream obtaining module 20, a first Gaussian weight distribution calculation module 30, a first substation equipment model obtaining module 40, an equipment feature guiding vector set obtaining module 50, a second substation equipment model obtaining module 60, and a third substation equipment model generating module 70.

[0052] The acquisition result obtaining module 10 is used to obtain the equipment image acquisition result and the equipment point cloud acquisition result of the substation equipment according to the camera and lidar equipment; the data stream obtaining module 20 is used to input the equipment image acquisition result into an image filter to obtain an equipment image data stream, and input the equipment point cloud acquisition result into a point cloud filter to obtain an equipment point cloud data stream; the first Gaussian weight distribution calculation module 30 is used to calculate the first Gaussian weight distribution based on the equipment point cloud data stream according to the Gaussian weight analysis model; the first substation equipment model obtaining module 40 is used to perform smooth modeling on the equipment point cloud data stream according to the first Gaussian weight distribution to obtain the first substation equipment model; the equipment feature guiding vector set obtaining module 50 is used to perform SIFT feature registration according to the equipment image data stream and the equipment point cloud data stream to obtain an equipment feature guiding vector set; the second substation equipment model obtaining module 60 is used to guide the equipment image data stream to perform detail texture enhancement on the first substation equipment model based on the equipment feature guiding vector set to obtain the second substation equipment model; the third substation equipment model generating module 70 is used to perform UV unfolding correction on the second substation equipment model according to the equipment image data stream and the equipment point cloud data stream to generate the third substation equipment model.

[0053] Next, the specific configuration of the first Gaussian weight distribution calculation module 30 will be described in detail. The first Gaussian weight distribution calculation module 30 further includes: dividing the equipment point cloud data stream into neighborhoods of each point according to a predetermined neighborhood radius to obtain neighborhoods of each point of the equipment; calculating the neighborhood centroid according to the neighborhoods of each point of the equipment to obtain the neighborhood centroid of each point; calculating the covariance matrix of each point according to the neighborhoods of each point of the equipment to obtain the covariance matrix of each point; inputting the neighborhood centroid of each point and the covariance matrix of each point into the Gaussian weight analysis model to output the multi-point Gaussian weight; and outputting the multi-point Gaussian weight as the first Gaussian weight distribution.

[0054] Next, the specific configuration of the first Gaussian weight distribution calculation module 30 will be further described in detail. The first Gaussian weight distribution calculation module 30 further includes: The Gaussian weight analysis model includes a Gaussian weight analysis function, and the Gaussian weight analysis function is: ; Wherein, represents the Gaussian weight of the i-th point, i represents the i-th point in the device point cloud data stream, and the i-th point is any point in the device point cloud data stream, represents the covariance matrix of the i-th point, represents the determinant of the covariance matrix of the i-th point, represents the inverse matrix of the covariance matrix of the i-th point, represents the exponential function with the natural constant e as the base, represents the centroid of the i-th point neighborhood, represents the deviation vector between the i-th point and the centroid of the i-th point neighborhood, represents the transposed vector of the deviation vector between the i-th point and the centroid of the i-th point neighborhood.

[0055] Next, the specific configuration of the first substation equipment model acquisition module 40 will be described in detail. The first substation equipment model acquisition module 40 further includes: performing normalization calculation according to the first Gaussian weight distribution to obtain a second Gaussian weight distribution; performing Gaussian weighted smoothing optimization on the device point cloud data stream according to the second Gaussian weight distribution to obtain a device smoothed point cloud set; performing three-dimensional modeling according to the device smoothed point cloud set to obtain the first substation equipment model.

[0056] Next, the specific configuration of the device feature guiding vector set acquisition module 50 will be described in detail. The device feature guiding vector set acquisition module 50 further includes: performing SIFT feature detection on the device image data stream to obtain a device image feature vector set; performing SIFT feature detection on the device point cloud data stream to obtain a device point cloud feature vector set; respectively performing matching cost recognition on the device point cloud feature vector set according to each device image feature vector in the device image feature vector set to obtain a plurality of matching cost coefficient sets; performing matching cost minimization screening according to the plurality of matching cost coefficient sets to determine an optimal matching cost distribution; performing registration association on the device image feature vector set and the device point cloud feature vector set according to the optimal matching cost distribution to generate the device feature guiding vector set.

[0057] Next, the specific configuration of the second substation equipment model obtaining module 60 will be described in detail. The second substation equipment model obtaining module 60 further includes: projecting the equipment image data stream onto the first substation equipment model based on the set of equipment feature guiding vectors to obtain a set of image projection points; optimizing the detailed texture of the first substation equipment model according to the set of image projection points to obtain an optimized substation equipment model; evaluating the smoothness of the detailed texture transition of the optimized substation equipment model to obtain a model transition smoothness coefficient; determining whether the model transition smoothness coefficient is less than a predetermined transition smoothness coefficient; if the model transition smoothness coefficient is less than the predetermined transition smoothness coefficient, performing transition smoothness optimization on the optimized substation equipment model according to the predetermined transition smoothness coefficient to generate the second substation equipment model.

[0058] Next, the specific configuration of the third substation equipment model generating module 70 will be described in detail. The third substation equipment model generating module 70 further includes: performing UV unwrapping on the second substation equipment model to obtain an unfolded view of the equipment model; constructing a detailed equipment texture map according to the equipment image data stream; drawing an equipment geometric feature map according to the equipment point cloud data stream; performing deviation detection on the unfolded view of the equipment model according to the detailed equipment texture map and the equipment geometric feature map to obtain an equipment model deviation detection result; correcting the second substation equipment model according to the equipment model deviation detection result to obtain the third substation equipment model.

[0059] The high-precision three-dimensional modeling platform for substation equipment provided by the embodiments of the present invention can execute the high-precision three-dimensional modeling method for substation equipment provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method.

[0060] Although various references are made in this application to certain modules in the platform according to the embodiments of this application, however, any number of different modules can be used and run on the user terminal and / or the server. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of mutual distinction and do not limit the protection scope of the present invention.

[0061] The above specific embodiments do not constitute a limitation on the protection scope of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of this application shall be included within the protection scope of this application.

Claims

1. A three-dimensional modeling method for substation equipment with high precision, characterized in that, The method includes: Obtaining the device image acquisition result and the device point cloud acquisition result of the substation device according to a camera and a lidar device; Inputting the device image acquisition result into an image filter to obtain a device image data stream, and inputting the device point cloud acquisition result into a point cloud filter to obtain a device point cloud data stream; Based on the device point cloud data stream, calculating a first Gaussian weight distribution according to a Gaussian weight analysis model; Performing smoothing modeling on the device point cloud data stream according to the first Gaussian weight distribution to obtain a first model of the substation device; Performing SIFT feature registration on the device image data stream and the device point cloud data stream to obtain a device feature guidance vector set; Based on the device feature guidance vector set, guiding the device image data stream to perform detail texture enhancement on the first model of the substation device to obtain a second model of the substation device; Performing UV unwrapping correction on the second model of the substation device according to the device image data stream and the device point cloud data stream to generate a third model of the substation device.

2. The three-dimensional modeling method for substation equipment with high precision according to claim 1, wherein, Based on the device point cloud data stream, calculating a first Gaussian weight distribution according to a Gaussian weight analysis model, including: Performing neighborhood division on each point of the device point cloud data stream according to a predetermined neighborhood radius to obtain neighborhoods of each point of the device; Calculating the neighborhood centroid according to the neighborhoods of each point of the device to obtain the centroid of each point neighborhood; Calculating the covariance matrix according to the neighborhoods of each point of the device to obtain the covariance matrix of each point; Inputting the centroid of each point neighborhood and the covariance matrix of each point into the Gaussian weight analysis model to output a multi-point Gaussian weight; Outputting the multi-point Gaussian weight as the first Gaussian weight distribution.

3. A three-dimensional modeling method for substation equipment with high precision as described in claim 1, characterized in that, The Gaussian weight analysis model includes a Gaussian weight analysis function, and the Gaussian weight analysis function is: ； Among them, represents the Gaussian weight of the i-th point, where i represents the i-th point in the device point cloud data stream, and the i-th point is any point in the device point cloud data stream, represents the covariance matrix of the i-th point, represents the determinant of the covariance matrix of the i-th point, represents the inverse matrix of the covariance matrix of the i-th point, represents the exponential function with the natural constant e as the base, represents the centroid of the i-th point neighborhood, represents the deviation vector between the i-th point and the centroid of the i-th point neighborhood, represents the transposed vector of the deviation vector between the i-th point and the centroid of the i-th point neighborhood.

4. A three-dimensional modeling method for substation equipment with high precision, characterized in that, Performing smoothing modeling on the device point cloud data stream according to the first Gaussian weight distribution to obtain a first model of the substation device, including: Performing normalization calculation according to the first Gaussian weight distribution to obtain a second Gaussian weight distribution; Performing Gaussian weighted smoothing optimization on the device point cloud data stream according to the second Gaussian weight distribution to obtain a device smooth point cloud set; Performing three-dimensional modeling according to the device smooth point cloud set to obtain the first model of the substation device.

5. A three-dimensional modeling method for substation equipment with high precision, characterized in that, Performing SIFT feature registration on the device image data stream and the device point cloud data stream to obtain a device feature guidance vector set, including: Performing SIFT feature detection on the device image data stream to obtain a device image feature vector set; Performing SIFT feature detection on the device point cloud data stream to obtain a device point cloud feature vector set; For each device image feature vector in the device image feature vector set, respectively performing matching cost identification on the device point cloud feature vector set to obtain multiple matching cost coefficient sets; Performing minimum matching cost screening according to the multiple matching cost coefficient sets to determine an optimal matching cost distribution; Performing registration association on the device image feature vector set and the device point cloud feature vector set according to the optimal matching cost distribution to generate the device feature guidance vector set.

6. A three-dimensional modeling method for substation equipment with high precision as described in claim 1, characterized in that, Based on the device feature guiding vector set, guiding the device image data stream to perform detail texture enhancement on the first substation device model to obtain the second substation device model, including: Based on the device feature guiding vector set, projecting the device image data stream onto the first substation device model to obtain an image projection point set; Performing detail texture optimization on the first substation device model according to the image projection point set to obtain an optimized substation device model; Performing a smoothness evaluation of detail texture transition on the optimized substation device model according to the optimized substation device model to obtain a model transition smoothness coefficient; Judging whether the model transition smoothness coefficient is less than a predetermined transition smoothness coefficient; If the model transition smoothness coefficient is less than the predetermined transition smoothness coefficient, performing a transition smoothness optimization on the optimized substation device model according to the predetermined transition smoothness coefficient to generate the second substation device model.

7. The three-dimensional modeling method for substation equipment with high precision according to claim 1, characterized in that, Performing a UV unwrapping correction on the second substation device model according to the device image data stream and the device point cloud data stream to generate the third substation device model, including: Performing a UV unwrapping on the second substation device model to obtain an unfolded device model diagram; Constructing a device detail texture diagram according to the device image data stream; Drawing a device geometric feature diagram according to the device point cloud data stream; Performing a deviation detection on the unfolded device model diagram according to the device detail texture diagram and the device geometric feature diagram to obtain a device model deviation detection result; Correcting the second substation device model according to the device model deviation detection result to obtain the third substation device model.

8. A three-dimensional modeling platform for substation equipment with high precision, characterized in that, The platform is used to implement a high-precision three-dimensional modeling method for substation devices according to any one of claims 1 to 7, and the platform includes: An acquisition result obtaining module, configured to obtain a device image acquisition result and a device point cloud acquisition result of the substation device according to a camera and a lidar device; A data stream obtaining module, configured to input the device image acquisition result into an image filter to obtain a device image data stream, and input the device point cloud acquisition result into a point cloud filter to obtain a device point cloud data stream; A first Gaussian weight distribution calculation module, configured to calculate a first Gaussian weight distribution based on the device point cloud data stream according to a Gaussian weight analysis model; A first substation device model obtaining module, configured to perform a smooth modeling on the device point cloud data stream according to the first Gaussian weight distribution to obtain a first substation device model; A device feature guiding vector set obtaining module, configured to perform a SIFT feature registration according to the device image data stream and the device point cloud data stream to obtain a device feature guiding vector set; A second substation device model obtaining module, configured to guide the device image data stream to perform detail texture enhancement on the first substation device model based on the device feature guiding vector set to obtain a second substation device model; A third substation device model generating module, configured to perform a UV unwrapping correction on the second substation device model according to the device image data stream and the device point cloud data stream to generate a third substation device model.

Images