Three-dimensional reconstruction method, equipment and medium for substation facilities based on point cloud segmentation
Through a three-dimensional reconstruction of the substation facilities based on point cloud segmentation, the problems of low efficiency and accuracy in the existing technology are solved, and efficient and accurate three-dimensional model creation is realized, suitable for substation facilities with complex structures.
Patent Information
- Application Number
- CN202410750164.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-12
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-06-12
AI Technical Summary
The efficiency and accuracy of the three-dimensional model reconstruction of substation facilities in the prior art is low, especially due to the structural complexity of the substation and the numerous equipment components, it is difficult for geometric reconstruction methods to quickly and accurately create high-quality three-dimensional models.
Using a point cloud segmentation method, the cloud data of the substation scene is segmented, and the three-dimensional reconstruction of the bracket and the wire are carried out separately. The power equipment model is adjusted through the equipment wiring data to connect it to the bracket and the wire model to form an accurate three-dimensional model. .
The efficiency and accuracy of the three-dimensional reconstruction of substation facilities is improved, the inefficiency and inaccuracy of traditional geometric reconstruction is avoided, and the representation needs from point cloud to three-dimensional model are met, taking into account adaptability and fidelity.
Smart Images

Figure CN118334274B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of three-dimensional reconstruction of substations, and in particular to a three-dimensional reconstruction method, device and medium for substation facilities based on point cloud segmentation. Background Art
[0002] In the context of smart grid construction, data visualization, integration, intelligence, and auxiliary analysis capabilities represent the important characteristics and development direction of the new management system. Virtual reality technology has been deeply applied to many aspects of power grid operation. Virtual reality technology has visibility, intuitiveness, authenticity, and real-time interactivity, and can provide intuitive reference for design, training, operation and maintenance, and decision-making departments.
[0003] However, there are still some obstacles in virtual reality technology, one of the key ones is the rapid creation of 3D models. Substations have the characteristics of compact structure, intricate composition and numerous equipment components. In the existing technology, the 3D model reconstruction of substations and various equipment components is usually carried out by geometric reconstruction, which has the problems of low modeling efficiency and modeling accuracy. Therefore, a more efficient 3D reconstruction method for substation facilities is needed. Summary of the invention
[0004] The present invention aims to solve the technical problems in the related art at least to a certain extent. To achieve the above-mentioned purpose, a method, device and medium for three-dimensional reconstruction of substation facilities based on point cloud segmentation are provided.
[0005] In a first aspect, the present invention provides a method for 3D reconstruction of substation facilities based on point cloud segmentation, comprising:
[0006] Segment the point cloud data of the substation scene to obtain the point cloud segmentation result;
[0007] Reconstructing three-dimensionally at least one bracket point cloud and multiple wire point clouds in the point cloud segmentation results to obtain corresponding three-dimensional models of each bracket and each wire;
[0008] Determine a set of wiring points corresponding to at least one power equipment point cloud in the point cloud segmentation result according to preset equipment wiring data and the plurality of three-dimensional wire models;
[0009] Search for individual equipment models corresponding to each of the power equipment point clouds from a preset model library, and adjust at least one of the individual equipment models found into a corresponding three-dimensional model of the power equipment according to the wiring point set, wherein the three-dimensional model of the power equipment is suitable for being connected to at least one of the corresponding three-dimensional model of the bracket and the three-dimensional model of the wire.
[0010] It can be understood that, compared with the prior art, the method provided by the present invention performs instance semantic segmentation on the substation scene point cloud through a point cloud segmentation algorithm to obtain a point cloud segmentation result. After point cloud segmentation, the distribution conditions of several point clouds and each other in the point cloud segmentation result are made clearer and more concrete, wherein several point clouds respectively represent substation facilities such as at least one power equipment, at least one bracket and multiple wires. According to the distribution conditions of the point cloud, three-dimensional modeling is first performed for each bracket and each wire respectively, and then, taking into account the wiring relationship between at least one power equipment and multiple wires, a three-dimensional model of the power equipment connected to the bracket three-dimensional model and / or the wire three-dimensional model is formed through a monomer model adjustment method, which not only meets the demand for representing the substation scene from point cloud to three-dimensional model, but also avoids the use of traditional three-dimensional reconstruction technology to reconstruct the geometry and texture of the power equipment in the substation scene, which helps to take into account the adaptability, accuracy and efficiency of three-dimensional reconstruction for the substation scene.
[0011] Further, the three-dimensional reconstruction of at least one bracket point cloud and multiple wire point clouds in the point cloud segmentation result is performed respectively to obtain corresponding three-dimensional models of each bracket and each wire, specifically including:
[0012] Performing surface mesh modeling on any of the bracket point clouds to form a corresponding three-dimensional model of the bracket, wherein the three-dimensional model of the bracket is suitable for replacing the corresponding bracket point cloud;
[0013] Segmenting any of the wire point clouds, performing cylinder fitting on each of the divided wire point cloud segments, and splicing the cylinder axes that match each of the wire point cloud segments;
[0014] According to the cylinder radius adapted to any of the wire point cloud segments, three-dimensional expansion is performed toward the periphery of the overall axis obtained by splicing, so that a corresponding three-dimensional model of the wire is formed, and the three-dimensional model of the wire is suitable for replacing the corresponding wire point cloud.
[0015] It is understandable that the method provided by the present invention reconstructs the bracket point cloud and the wire point cloud in three dimensions respectively. Accurately dividing a wire point cloud into multiple wire point cloud segments is conducive to reducing the difficulty of subsequent processing. Cylindrical fitting is performed on each wire point cloud segment to achieve parametric representation of the wire, which is convenient for subsequent data processing and analysis. The sub-central axes of each cylindrical model are spliced to form a total central axis, and three-dimensional expansion is performed to the periphery of the overall axis obtained by splicing to ensure that the wire bending position is smooth and natural, and to improve the accuracy and fidelity of the wire three-dimensional model.
[0016] Furthermore, the splicing of the cylindrical axes adapted to the respective wire point cloud segments specifically includes:
[0017] Performing spatial relationship detection on two cylindrical axis sections that are matched one-to-one with any two adjacent wire point cloud sections;
[0018] If the two cylindrical axes are separated, two first target positions that are distributed on the two cylindrical axes and are closest to each other are connected by a straight line segment;
[0019] Performing spatial relationship detection on each of the first target positions and each of the line end positions that are suitable for being located on the same cylindrical axis as the first target position;
[0020] If any of the first target positions is far away from the corresponding line end positions, the line end position closest to the target position will be used as the corresponding second target position, and the axis part from the second target position to the first target position on the corresponding cylindrical axis will be deleted; otherwise, the corresponding cylindrical axis will be kept unchanged.
[0021] Furthermore, the step of determining a set of connection points corresponding to at least one power equipment point cloud in the point cloud segmentation result according to the preset equipment connection data and the plurality of three-dimensional wire models specifically includes:
[0022] Treating at least one electric power equipment point cloud in the point cloud segmentation result as a virtual connection object with input and output sides, and selecting a plurality of three-dimensional wire models suitable for matching the same virtual connection object according to the equipment wiring data;
[0023] The center points of the ends of the wire models distributed on the selected three-dimensional wire models and suitable for connecting with the virtual connection object are respectively set, so that the corresponding at least one input side wiring point and at least one output side wiring point are located in the wiring point set.
[0024] It can be understood that with the help of equipment wiring data, it is possible to accurately select multi-conductor three-dimensional models suitable for distribution on the input and output sides of the same power equipment point cloud in accordance with circuit rules, and then set the input side wiring points and output side wiring points of each conductor three-dimensional model to the wiring point concentration, so that the complex equipment and conductor relationship is structured and simplified, which helps to ensure calculation efficiency and facilitate subsequent use.
[0025] Further, adjusting the at least one individualized model of the equipment found according to the wiring point set into a corresponding three-dimensional model of the power equipment specifically includes:
[0026] The three-dimensional coordinate system where the point cloud segmentation result is located is used as the global coordinate system, and the three-dimensional coordinate system on the currently found monomer model of the device is used as the local coordinate system;
[0027] Selecting an input-side wiring point and an output-side wiring point corresponding to the same device monomer model from the wiring point set;
[0028] The selected input-side wiring point and the selected output-side wiring point are respectively regarded as sample points, and two three-dimensional coordinates distributed in the global coordinate system and the local coordinate system are obtained for the sample points, so that a corresponding three-dimensional coordinate set can be set;
[0029] Performing multiple parameter solutions according to the multiple three-dimensional coordinate sets corresponding to the same device monomer model to obtain a corresponding device model parameter set;
[0030] According to the equipment model parameter set, the corresponding equipment monomer model is adjusted to the electric equipment three-dimensional model suitable for replacing the corresponding electric equipment point cloud.
[0031] Further, the device model parameter set includes a relative position, a size scaling and a rotation angle between the local coordinate system and the global coordinate system;
[0032] The adjusting, according to the equipment model parameter set, the equipment monomer model to the electric equipment three-dimensional model suitable for replacing the corresponding electric equipment point cloud specifically includes:
[0033] Adjusting the position and size of the device monomer model according to the relative position, size scaling and rotation angle between the local coordinate system and the global coordinate system;
[0034] The adjusted individual equipment model is used as the three-dimensional model of the electric power equipment to replace the corresponding point cloud of the electric power equipment.
[0035] It can be understood that the method provided by the present invention quickly determines the input-side wiring points and output-side wiring points corresponding to each device based on the three-dimensional model of the wire, and obtains the device model parameters corresponding to the device monomer model, that is, the relative position, size scaling and rotation angle of the local coordinate system and the global coordinate system, and can adjust the posture and size of the device monomer model so that the device monomer model can be accurately represented in the three-dimensional space of the power equipment point cloud.
[0036] Furthermore, the present invention also includes: performing quality detection on the connection status between the three-dimensional model of the power equipment, the three-dimensional model of the bracket and the three-dimensional model of the wire based on the multi-view image data corresponding to the substation, and correcting at least one of the three-dimensional model of the power equipment, the three-dimensional model of the bracket and the three-dimensional model of the wire with abnormal connection based on the corresponding quality detection results.
[0037] It can be understood that after establishing the substation scene model, the method provided by the present invention detects the connection quality between each conductor, bracket and power equipment according to the multi-view image data, which can optimize the substation scene model and improve the accuracy of the model.
[0038] In a second aspect, the present invention provides an electronic device comprising a processor, a memory, and a computer program stored in the memory and suitable for being executed by the processor, wherein when the processor executes the computer program, the method for three-dimensional reconstruction of substation facilities based on point cloud segmentation as described in any one of the above items is implemented.
[0039] In a third aspect, the present invention provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the method for three-dimensional reconstruction of substation facilities based on point cloud segmentation as described in any one of the above items is implemented. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] Figure 1 A flow chart of a method for three-dimensional reconstruction of substation facilities based on point cloud segmentation provided by an embodiment of the present invention;
[0041] Figure 2 A schematic diagram of a parameterized cylindrical model provided by an embodiment of the present invention;
[0042] Figure 3 A schematic diagram of a plurality of cylindrical models connected in sequence provided by an embodiment of the present invention;
[0043] Figure 4 For the corresponding Figure 3 Schematic diagram of the four cylindrical axis segments during the splicing process;
[0044] Figure 5 A schematic diagram of another two cylindrical axis sections in the process of splicing provided by an embodiment of the present invention;
[0045] Figure 6 A schematic diagram of a single wire three-dimensional model provided in an embodiment of the present invention;
[0046] Figure 7 A schematic diagram of a topological structure of a substation facility provided by an embodiment of the present invention;
[0047] Figure 8 A schematic diagram of adjusting parameters of a device monomer model provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0049] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.
[0050] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.
[0051] Existing substation scene models usually use geometric reconstruction methods, which often ignore the power line point cloud, resulting in inaccurate substation scene models. In addition, geometric modeling of each device in each substation is inefficient and requires a lot of computing resources. A method is needed to improve the accuracy of three-dimensional reconstruction of substation facilities.
[0052] Please refer to Figure 1 , an embodiment of the present invention provides a three-dimensional reconstruction method of a substation facility based on point cloud segmentation, which includes S101-S104.
[0053] S101: Segment the substation scene point cloud data to obtain a point cloud segmentation result.
[0054] Optionally, before S101, a variety of preparatory data corresponding to the substation is loaded, and the plurality of preparatory data includes equipment wiring data, multi-view image data and laser point cloud data. According to the multi-view image data, sparse three-dimensional point cloud data is obtained by motion structure recovery, and then, the sparse three-dimensional point cloud data is reconstructed by a multi-view stereo matching (MVS) algorithm to generate dense three-dimensional point cloud data that can be used as substation scene point cloud data. The dense three-dimensional point cloud data can be registered with the laser point cloud data by at least one of an ICP algorithm, an NDT algorithm and a random sampling consistency algorithm, so as to obtain the substation scene point cloud data.
[0055] In this embodiment, the point cloud data of the substation scene is segmented by an artificial intelligence algorithm to obtain a point cloud segmentation result.
[0056] It should be noted that the substation scene point cloud data can already clearly define the distribution of the three point clouds, namely, the bracket point cloud, the conductor point cloud and the equipment point cloud, in three-dimensional space. The point cloud segmentation algorithm assigns instantiation attributes and / or semantic category attributes to the three point clouds so that the bracket point cloud, the conductor point cloud and the equipment point cloud can be distinguished, but does not change the distribution of the three point clouds in three-dimensional space.
[0057] Exemplarily, the artificial intelligence algorithm may adopt an end-to-end point cloud instance segmentation algorithm, or may adopt a fusion algorithm similar to point cloud semantic segmentation followed by point cloud instance segmentation based on spatial distance.
[0058] Optionally, check the correspondence between the substation cloud segmentation results and the equipment wiring data, and adjust the incorrectly classified point cloud segmentation results. For example, if a switch in the equipment wiring data is upstream of the transformer, but in the segmented substation cloud, the upstream of the transformer is not a switch, it means that the classification is wrong. In this case, the upstream of the substation should be changed to the switch category.
[0059] S102: Reconstruct three-dimensionally the at least one bracket point cloud and the multiple wire point clouds in the point cloud segmentation result to obtain corresponding three-dimensional models of each bracket and three-dimensional models of each wire.
[0060] In this embodiment, since the structure of the bracket is relatively simple, surface meshing modeling is performed on any of the bracket point clouds to form a corresponding three-dimensional model of the bracket. The three-dimensional model of the bracket is suitable for replacing the corresponding bracket point cloud. The surface meshing modeling method includes but is not limited to any one of Poisson reconstruction and Delaunay triangulation.
[0061] In this embodiment, due to the differences in length, material, connection method and appearance of different wires, the number of wire point clouds is relatively small compared to brackets and power equipment, and it is difficult to directly model them geometrically, so the wire point clouds are parametrically modeled. Any of the wire point clouds is segmented, and cylinder fitting is performed on each of the divided wire point cloud segments, and the cylindrical axis adapted to each of the wire point cloud segments is spliced; according to the cylinder radius adapted to any of the wire point cloud segments, three-dimensional expansion is performed to the periphery of the overall axis formed by splicing, so that the corresponding wire three-dimensional model is formed, and the wire three-dimensional model is suitable for replacing the corresponding wire point cloud.
[0062] For example, each wire point cloud is segmented by supervoxel clustering based on the distance, so that a wire point cloud is divided into multiple wire point cloud segments; a suitable cylindrical model is fitted for each wire point cloud segment by the principal component analysis method and the linear least squares method; see Figure 2 Any of the cylindrical models can be represented parameterized, and multiple cylindrical parameters include the cylinder radius r0, the cylinder axis vector (a, b, c) and any point P (x, y, z) located on the cylinder axis, among which the cylinder radius r0 can be used as a common attribute between multiple cylindrical models.
[0063] In this embodiment, a spatial relationship detection is performed on two cylindrical axes that are matched one by one with any two adjacent wire point cloud segments. If the two cylindrical axes are separated, two first target positions that are distributed on the two cylindrical axes and are closest to each other are connected by a straight line segment; a spatial relationship detection is performed on each of the first target positions and each line end position that is suitable for being on the same cylindrical axis; if any of the first target positions is separated from the corresponding line end positions, the line end position that is closest to the target position is used as the corresponding second target position, and on the corresponding cylindrical axis, the axis portion from the second target position to the first target position is deleted, otherwise, the corresponding cylindrical axis is kept unchanged.
[0064] For example, see Figure 3 and Figure 4 , the four cylindrical models connected in sequence can roughly represent a wire, the axis located in each cylindrical model is the corresponding cylindrical axis, the four cylindrical axes can be marked as La, Lb, Lc and Ld from left to right, the two cylindrical axes La and Lb are separated from each other, one end of the cylindrical axis Lb is connected to one end of the cylindrical axis Lc, and the two cylindrical axes Lc and Ld are also separated from each other, wherein for any cylindrical axis, any end located thereon (which can be called the line end position) can be set as a hollow circle and set as a solid line between the two line end positions.
[0065] It should be pointed out that the straight line segment connecting the two cylindrical axis segments is also the shortest connecting line. In order to better distinguish, the two ends of the shortest connecting line can be set as solid points and the two ends can be set as a dotted line. The deleted axis part can be set as a solid line, where this part belongs to the redundant wire part caused by the point cloud fitting error.
[0066] Exemplarily, for two cylindrical axes La and Lb, one first target position is at one end of the cylindrical axis La and the other first target position is at one end of the cylindrical axis Lb. In this way, there is no second target position, and the cylindrical axes La and Lb at both ends should be kept unchanged so that the two cylindrical axes La and Lb can be spliced; for two cylindrical axes Lc and Ld, one first target position is at one end of the cylindrical axis Lc and the other first target position is separated from both ends of the cylindrical axis Ld. There is only one second target position, which is a line end position on the cylindrical axis Ld. The cylindrical axis Lc should be kept unchanged, and the solid line portion with the second target position on the cylindrical axis Ld should be deleted so that the two cylindrical axes Lc and Ld can be spliced.
[0067] For example, Figure 5 As shown, for the other two cylindrical axis segments, neither of the two first target positions belongs to any line end position on the two cylindrical axis segments. Thus, the two second target positions are distributed at the two line end positions closest to each other on the two cylindrical axis segments, and each cylindrical axis segment has a solid line portion that needs to be deleted.
[0068] S103: Determine a set of connection points corresponding to at least one power equipment point cloud in the point cloud segmentation result according to preset equipment connection data and the plurality of three-dimensional wire models.
[0069] In this embodiment, the instance relationship between the power equipment, the bracket and the wire in the point cloud segmentation result can be represented in the form of a graph structure to form a corresponding instance structure diagram; by comparing the instance structure diagram with the equipment wiring data in the form of a circuit diagram, each power equipment point cloud is regarded as a virtual connection object with an input and output side, and multiple wire three-dimensional models suitable for matching with the same virtual connection object are selected, wherein a wire three-dimensional model can be as follows: Figure 6 Shown.
[0070] It should be noted that the wiring requirements can be described as being suitable for distributing at least one three-dimensional wire model on the input side and the output side. Therefore, the at least two three-dimensional wire models include an input side wire three-dimensional model suitable for distribution on the input side of a found power equipment point cloud and an output side wire three-dimensional model suitable for distribution on the output side of the same power equipment point cloud.
[0071] In this embodiment, center points are set for ends of each wire model distributed on each selected wire 3D model and suitable for connecting with a virtual connection object, so that the corresponding at least one input side wiring point and at least one output side wiring point are located in the wiring point set.
[0072] Exemplarily, an end of a wire model of any of the above-mentioned input side wire three-dimensional models relative to the power equipment point cloud is used as the input side wiring terminal, and the midpoint of the input side wiring terminal is used as the input side wiring point; an end of any of the above-mentioned output side wire three-dimensional models relative to the power equipment point cloud is used as the output side wiring terminal, and the center of the output side wiring terminal is used as the output side wiring point; at least one input side wiring point and at least one output side wiring point suitable for matching the same power equipment point cloud are associated as a group of wiring points and stored in a preset data set, so that a wiring point set is finally formed.
[0073] For different power equipment in the substation, the distinction between input-side wiring points and output-side wiring points is also different. For grounded power equipment such as lightning arresters, the output-side wiring points are directly connected to the bracket / ground, and the output-side wiring points can be the spatial position of the vertical projection of the input-side wiring points on the plane where the bracket / ground is located; for power equipment that processes three-phase electricity at the same time, such as transformers, there are multiple wiring points on the input and output sides, and the multiple wiring points are connected to different wires.
[0074] It would be difficult to avoid the three-dimensional model of the wires and directly look for multiple scattered wiring points on the three-dimensional model representing the power equipment. However, if a multi-wire three-dimensional model is already available, the input and output sides of the three-dimensional model representing the power equipment can be easily known by following the circuit rules. Therefore, with the help of the multi-wire three-dimensional model, it is easier and faster to find multiple wiring points suitable for representing the connection of the same power equipment with multiple wires.
[0075] It should be pointed out that for various facilities in the substation, such as power equipment, brackets (columns or poles) and conductors, at least a topological diagram (such as Figure 7 ) indicates that the equipment wiring data can reflect the connection relationship between the conductors and power equipment in the substation.
[0076] S104: Searching for individual equipment models corresponding to each of the power equipment point clouds from a preset model library, and adjusting at least one of the individual equipment models found into corresponding three-dimensional models of power equipment according to the wiring point set, wherein the three-dimensional model of power equipment is suitable for being connected to at least one of the corresponding three-dimensional model of the bracket and the three-dimensional model of the wire.
[0077] It should be noted that for any device model in the preset model library, due to different production methods, its absolute size is not fixed or accurate, but the ratio of the length, width and height of the model is fixed and unchanged, and it can be regarded as a rigid body. This rigid body is more conducive to parametric representation, that is, a single device model belongs to a parametric model. The parameters of the parametric model can be adjusted (see Figure 8 ), adapting it to the size and posture of the corresponding power equipment point cloud in the point cloud space, saving geometric modeling time.
[0078] In this embodiment, the three-dimensional coordinate system where the point cloud segmentation result is located is used as the global coordinate system, and the three-dimensional coordinate system on the currently found device monomer model is used as the local coordinate system; the input side wiring point and the output side wiring point corresponding to the same device monomer model are selected from the wiring point set; the selected input side wiring point and the output side wiring point are respectively used as sample points, and two three-dimensional coordinates distributed in the global coordinate system and the local coordinate system are obtained for the sample points, so that the corresponding three-dimensional coordinate set can be set; multiple parameters are solved according to the multiple three-dimensional coordinate sets corresponding to the same device monomer model to obtain the corresponding device model parameter set; according to the device model parameter set, the corresponding device monomer model is adjusted to the three-dimensional model of the power equipment suitable for replacing the corresponding power equipment point cloud.
[0079] Assume that a sample point corresponds to the three-dimensional coordinates (x 1 ,y 1 ,z 1 ), the above sample points correspond to the three-dimensional coordinates (x 1 ,y 1 ,z 1 ), the relative position of the origin (0,0,0) in the local coordinate system corresponding to the global coordinate system is expressed as (x 0 ,y 0 ,z 0 ), the angle of rotation around the Z axis is represented by ψ, and the size scaling is represented by s, then the following matrix equation is used to represent it.
[0080]
[0081] It should be noted that, for any power equipment in the substation, the bottom of the equipment is parallel to the ground, which can be regarded as: the parameters of rotation around the X and Y axes parallel to the ground are known, and the parameters of rotation around the Z axis perpendicular to the ground are unknown. Therefore, the angle ψ of rotation around the Z axis is set as a parameter to be solved in the above matrix equation.
[0082] For example, when a certain output side wiring point is used as sample point a, sample point a corresponds to two three-dimensional coordinates in the global coordinate system and the local coordinate system, which can form a three-dimensional coordinate set [(xa, ya, za), (xa ,ya ,za )], and (xa,ya,za) and (xa ,ya ,za ) into the above matrix equation.
[0083] For example, when a certain output side wiring point is used as a sample point b, the sample point b corresponds to two three-dimensional coordinates in the global coordinate system and the local coordinate system, which can form a three-dimensional coordinate set [(xb, yb, zb), (xb ,yb ,zb )], (xb,yb,zb) and (xb ,yb ,zb ) into the above matrix equation.
[0084] For example, for a device monomer model with more than two connection points, the nonlinear least squares method is used to estimate the parameters of the two matrix equations corresponding to the two sample points a and b, and the device model parameter set θ=[x 0 ,y 0 ,z 0 ,ψ,s], and define the cost function: Cost(θ) = Σ||Pi - T(Pi',θ)||^2. Pi represents the measured position of the i-th connection point, Pi' is the theoretical position of the same point in the reference coordinate system, and T(Pi',θ) represents the position of Pi' transformed to the actual space according to θ. Through Gauss-Newton method, Levenberg-Marquardt algorithm, etc., the cost function Cost(θ) is minimized, that is, the error with the observed data is minimized.
[0085] In this embodiment, the posture and size of the individual model of the equipment are adjusted according to the relative position of the local coordinate system and the global coordinate system, the size scaling scale and the rotation angle; the adjusted individual model of the equipment is used as the three-dimensional model of the power equipment to replace the corresponding power equipment point cloud. For example, when the three-dimensional model of the power equipment is in the three-dimensional area occupied by the corresponding power equipment point cloud, the power equipment point cloud occupying the three-dimensional area can be deleted or hidden.
[0086] It should be noted that since the support 3D model, conductor 3D model and equipment 3D model replace the areas occupied by the corresponding support point cloud, conductor point cloud and equipment point cloud in the substation scene point cloud data, and the substation scene point cloud data itself has clearly defined the distribution of the three point cloud data in three-dimensional space, the three monomer 3D models can be naturally arranged in three-dimensional space with the help of segmented point clouds to establish a substation scene model to represent the complete substation scene.
[0087] In this embodiment, the present invention also includes: performing quality detection on the connection status between the three-dimensional model of the power equipment, the three-dimensional model of the bracket and the three-dimensional model of the wire based on the multi-view image data corresponding to the substation, and correcting at least one of the three-dimensional model of the power equipment, the three-dimensional model of the bracket and the three-dimensional model of the wire with abnormal connection based on the corresponding quality detection results.
[0088] Specifically, for the case where they are close in the vertical direction and the equipment is located above the bracket, the bracket is usually the base of the substation equipment. If there is a gap in the wiring between the equipment and the bracket, the two can be made to fit together by increasing the bracket height or lowering the equipment height. For the connection points of the wires and the equipment, they coincide in the spatial coordinates, but the preset positions of the connection points in the equipment model are usually components such as equipment terminal blocks. If there is an overlap between the equipment model and the wire model, the two can be made to fit together by deleting the repeated parts of the wire model and the equipment model.
[0089] By adopting the above-mentioned three-dimensional reconstruction method of substation facilities based on point cloud segmentation, after point cloud segmentation, several point clouds in the point cloud segmentation results and their distribution conditions are made clearer and more concrete, wherein several point clouds respectively represent substation facilities such as at least one power equipment, at least one bracket and multiple wires. According to the distribution of the point clouds, three-dimensional modeling is first performed for each bracket and each wire respectively, and then, taking into account the wiring relationship between at least one power equipment and the multiple wires, a three-dimensional model of the power equipment connected to the three-dimensional model of the bracket and / or the three-dimensional model of the wire is formed through a monomer model adjustment method, which not only meets the demand for representing the substation scene from point cloud to three-dimensional model, but also avoids the use of traditional three-dimensional reconstruction technology to reconstruct the geometry and texture of the power equipment in the substation scene, which helps to take into account the adaptability, accuracy and efficiency of three-dimensional reconstruction of the substation scene.
[0090] The present invention also provides another step flow of a three-dimensional reconstruction method for substation facilities based on point cloud segmentation, including: segmenting the substation scene point cloud data to obtain a point cloud segmentation result; performing three-dimensional reconstruction on multiple conductor point clouds in the point cloud segmentation result to obtain corresponding three-dimensional models of each conductor; determining a wiring point set corresponding to at least one power equipment point cloud in the point cloud segmentation result based on preset equipment wiring data and multiple three-dimensional models of the conductor; searching for a device monomer model corresponding to each of the power equipment point clouds from a preset model library, and adjusting the at least one device monomer model found to a corresponding three-dimensional model of the power equipment based on the wiring point set.
[0091] After point cloud segmentation, several point clouds in the point cloud segmentation results and their distribution conditions are made clearer and more concrete. Among them, several point clouds respectively represent at least one power equipment and multiple wires and other substation facilities. According to the point cloud distribution conditions, three-dimensional modeling is first performed for each wire separately. Then, taking into account the wiring relationship between at least one power equipment and multiple wires, a three-dimensional model of the power equipment connected to the three-dimensional model of multiple wires is formed through a monomer model adjustment method. This not only meets the demand for representing the power equipment in the substation scene and the multiple wires connected thereto from point clouds to three-dimensional models, but also avoids the use of traditional three-dimensional reconstruction technology to reconstruct the geometry and texture of the aforementioned power equipment, which helps to take into account the adaptability, accuracy and reconstruction efficiency of the three-dimensional model of the power equipment.
[0092] The present invention also provides an electronic device, comprising: a processor, a memory, and a computer program stored in the memory and suitable for being executed by the processor, wherein when the processor executes the computer program, the method for three-dimensional reconstruction of substation facilities based on point cloud segmentation as described in any of the above embodiments is implemented.
[0093] The electronic device of this embodiment includes: a processor, a memory, and a computer program and computer instructions stored in the memory and executable on the processor. When the processor executes the computer program, each step in the above-mentioned embodiment 1 is implemented, for example: Figure 1 Alternatively, the processor implements the functions of each module / unit in the above system embodiment when executing the computer program.
[0094] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the electronic device. For example, Figure 1The module corresponding to S104 shown can search for the individual equipment model corresponding to each of the power equipment point clouds from the preset model library, and adjust at least one of the individual equipment models found into a corresponding three-dimensional model of the power equipment according to the wiring point set, wherein the three-dimensional model of the power equipment is suitable for being connected to at least one of the corresponding three-dimensional model of the bracket and the three-dimensional model of the wire.
[0095] The electronic device may be a computing device such as a desktop computer, a notebook, a PDA, and a cloud server. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the schematic diagram is merely an example of an electronic device and does not constitute a limitation on the electronic device. The electronic device may include more or fewer components than shown in the diagram, or may combine certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.
[0096] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the electronic device, and uses various interfaces and lines to connect various parts of the entire electronic device.
[0097] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the electronic device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created according to the use of the mobile terminal, etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.
[0098] Wherein, if the module / unit integrated in the electronic device 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 such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.
[0099] The present invention also provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the method for three-dimensional reconstruction of substation facilities based on point cloud segmentation as described in any of the above embodiments is implemented.
[0100] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.
Claims
1. A three-dimensional reconstruction method for substation facilities based on point cloud segmentation, characterized in that: include: Segment the point cloud data of the substation scene to obtain the point cloud segmentation result; Reconstructing three-dimensionally at least one bracket point cloud and multiple wire point clouds in the point cloud segmentation results to obtain corresponding three-dimensional models of each bracket and each wire; Determine a set of wiring points corresponding to at least one power equipment point cloud in the point cloud segmentation result according to preset equipment wiring data and the plurality of three-dimensional wire models; Searching for a device monomer model corresponding to each of the power equipment point clouds from a preset model library, and adjusting at least one of the found device monomer models into a corresponding power equipment three-dimensional model according to the wiring point set, wherein the power equipment three-dimensional model is suitable for being connected to at least one of the corresponding bracket three-dimensional model and the wire three-dimensional model; The step of determining a set of connection points corresponding to at least one power equipment point cloud in the point cloud segmentation result according to the preset equipment connection data and the plurality of three-dimensional wire models specifically includes: Treating at least one electric power equipment point cloud in the point cloud segmentation result as a virtual connection object with input and output sides, and selecting a plurality of three-dimensional wire models suitable for matching the same virtual connection object according to the equipment wiring data; The center points of the ends of the wire models distributed on the selected three-dimensional wire models and suitable for connecting with the virtual connection object are respectively set, so that the corresponding at least one input side wiring point and at least one output side wiring point are in the wiring point set.
2. The method for 3D reconstruction of substation facilities based on point cloud segmentation according to claim 1, characterized in that: The three-dimensional reconstruction of at least one bracket point cloud and multiple wire point clouds in the point cloud segmentation results to obtain corresponding three-dimensional models of each bracket and each wire specifically includes: Performing surface mesh modeling on any of the bracket point clouds to form a corresponding three-dimensional model of the bracket, wherein the three-dimensional model of the bracket is suitable for replacing the corresponding bracket point cloud; Segmenting any of the wire point clouds, performing cylinder fitting on each of the divided wire point cloud segments, and splicing the cylinder axes that match each of the wire point cloud segments; According to the cylinder radius adapted to any of the wire point cloud segments, three-dimensional expansion is performed toward the periphery of the overall axis obtained by splicing, so that a corresponding three-dimensional model of the wire is formed, and the three-dimensional model of the wire is suitable for replacing the corresponding wire point cloud.
3. The method for 3D reconstruction of substation facilities based on point cloud segmentation according to claim 2, characterized in that: The splicing process of the cylindrical axes adapted to the respective wire point cloud segments specifically includes: Performing spatial relationship detection on two cylindrical axis sections that are matched one-to-one with any two adjacent wire point cloud sections; If the two cylindrical axes are separated, two first target positions that are distributed on the two cylindrical axes and are closest to each other are connected by a straight line segment; Performing spatial relationship detection on each of the first target positions and each of the line end positions that are suitable for being located on the same cylindrical axis as the first target position; If any of the first target positions is far away from the corresponding line end positions, the line end position closest to the first target position will be taken as the corresponding second target position, and the axis part from the second target position to the first target position on the corresponding cylindrical axis will be deleted; otherwise, the corresponding cylindrical axis will be kept unchanged.
4. The method for 3D reconstruction of substation facilities based on point cloud segmentation according to claim 1, characterized in that: The adjusting the at least one device monomer model found according to the wiring point set into a corresponding three-dimensional model of the power device specifically includes: The three-dimensional coordinate system where the point cloud segmentation result is located is used as the global coordinate system, and the three-dimensional coordinate system on the currently found monomer model of the device is used as the local coordinate system; Selecting an input-side wiring point and an output-side wiring point corresponding to the same device monomer model from the wiring point set; The selected input-side wiring point and the selected output-side wiring point are respectively regarded as sample points, and two three-dimensional coordinates distributed in the global coordinate system and the local coordinate system are obtained for the sample points, so that a corresponding three-dimensional coordinate set can be set; Performing multiple parameter solutions according to the multiple three-dimensional coordinate sets corresponding to the same device monomer model to obtain a corresponding device model parameter set; According to the equipment model parameter set, the corresponding equipment monomer model is adjusted to the electric power equipment three-dimensional model, so that the corresponding electric power equipment point cloud is replaced.
5. The method for 3D reconstruction of substation facilities based on point cloud segmentation according to claim 4, characterized in that: The device model parameter set includes a relative position, a size scaling, and a rotation angle between the local coordinate system and the global coordinate system; The adjusting the corresponding device monomer model to the power device three-dimensional model according to the device model parameter set so that the corresponding power device point cloud is replaced specifically includes: Adjusting the pose and size of the device monomer model according to the relative position between the local coordinate system and the global coordinate system, the size scaling scale and the rotation angle; The adjusted individualized model of the equipment is used as the three-dimensional model of the electric power equipment to replace the corresponding point cloud of the electric power equipment.
6. The method for three-dimensional reconstruction of substation facilities based on point cloud segmentation according to any one of claims 1 to 5, characterized in that: Also includes: Based on the multi-view image data corresponding to the substation, the connection status between the three-dimensional model of the power equipment, the three-dimensional model of the bracket and the three-dimensional model of the wire is quality checked, and at least one of the three-dimensional model of the power equipment, the three-dimensional model of the bracket and the three-dimensional model of the wire with abnormal connection is corrected according to the corresponding quality check result.
7. A method for 3D reconstruction of substation facilities based on point cloud segmentation, characterized in that: include: Perform segmentation processing on the substation scene point cloud data to obtain the point cloud segmentation result; Reconstructing the point clouds of multiple conductors in the point cloud segmentation results in three dimensions to obtain corresponding three-dimensional models of each conductor; Determine a set of wiring points corresponding to at least one power equipment point cloud in the point cloud segmentation result according to preset equipment wiring data and the plurality of three-dimensional wire models; Searching for a device monomer model corresponding to each of the power equipment point clouds from a preset model library, and adjusting at least one of the found device monomer models into a corresponding power equipment three-dimensional model according to the wiring point set, wherein the power equipment three-dimensional model is suitable for connection with the corresponding wire three-dimensional model; The step of determining a set of connection points corresponding to at least one power equipment point cloud in the point cloud segmentation result according to the preset equipment connection data and the plurality of three-dimensional wire models specifically includes: Treating at least one electric power equipment point cloud in the point cloud segmentation result as a virtual connection object with input and output sides, and selecting a plurality of three-dimensional wire models suitable for matching the same virtual connection object according to the equipment wiring data; The center points of the ends of the wire models distributed on the selected three-dimensional wire models and suitable for connecting with the virtual connection object are respectively set, so that the corresponding at least one input side wiring point and at least one output side wiring point are in the wiring point set.
8. An electronic device, characterized in that: The method comprises a processor, a memory and a computer program stored in the memory and suitable for being executed by the processor, wherein when the processor executes the computer program, the method for three-dimensional reconstruction of substation facilities based on point cloud segmentation as described in any one of claims 1 to 7 is implemented.
9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the method for three-dimensional reconstruction of substation facilities based on point cloud segmentation as described in any one of claims 1 to 7 is implemented.
Citation Information
Patent Citations
Three-dimensional model layout method for power transformation equipment in transformer substation
CN117115390A
