Transmission tower model construction method and device

Abstract

The invention discloses a power transmission tower model construction method and device. The method comprises the following steps: acquiring laser point cloud data of a power transmission tower at different moments; grouping the laser point cloud data in pairs according to the sequence of the moments; repeating the steps on each group of laser point cloud data until all the laser point cloud data are merged into one piece of laser point cloud data; constructing a transmission tower model according to the combined laser point cloud data, and determining a three-dimensional model of the transmission tower; the method comprises the steps of inputting the first laser point cloud data and the second laser point cloud data into a pre-trained feature extraction network to obtain a first feature descriptor and a second feature descriptor; combining the second laser point cloud data and the first laser point cloud data according to the first feature descriptor and the second feature descriptor; and grouping the combined laser point cloud data in pairs. According to the invention, the construction cost of the transmission tower model is reduced, and the transmittance test of the infrared lens at different temperatures is realized.

Description

Technical Field

[0001] The present invention relates to the field of transmittance testing technology, and in particular to a method and device for constructing a transmission tower model. Background Art

[0002] This section is intended to provide a background or context for the embodiments of the invention described herein. No description herein is admitted to be prior art by virtue of its inclusion in this section.

[0003] With the rapid development of smart grid systems, the requirements for refined and intelligent management of power grids are increasing. As a key component of the power grid, the three-dimensional modeling of transmission towers has become an important part of building digital and intelligent power grids.

[0004] Traditional transmission tower modeling methods mainly rely on manual measurement and two-dimensional drawings, including manual on-site measurement of tower height, cross-sectional dimensions, structural form, connection nodes and other detailed parameters, or obtaining the above detailed parameters based on the two-dimensional drawings of the transmission tower, and then inputting the above parameters into the three-dimensional model construction software to obtain the three-dimensional model of the transmission tower. There are problems such as large workload, low efficiency and insufficient accuracy. Summary of the invention

[0005] The embodiment of the present invention provides a method for constructing a transmission tower model, which is used to improve the efficiency and accuracy of constructing a transmission tower model and reduce the workload. The method includes:

[0006] Obtain laser point cloud data of transmission towers at different times;

[0007] The laser point cloud data are grouped in pairs according to the sequence of time; wherein the laser point cloud data at two adjacent time points is a group; the grouped laser point cloud data includes the first laser point cloud data and the second laser point cloud data; the first laser point cloud data is the laser point cloud data at the earlier time point in each group; the second laser point cloud data is the laser point cloud data at the later time point in each group;

[0008] Repeating the steps for each set of laser point cloud data until all laser point cloud data are merged into one laser point cloud data;

[0009] The transmission tower model is constructed based on the merged laser point cloud data to determine the three-dimensional model of the transmission tower;

[0010] The repeating steps include:

[0011] The first laser point cloud data and the second laser point cloud data are respectively input into a pre-trained feature extraction network to obtain a first feature descriptor and a second feature descriptor; the first feature descriptor is a feature descriptor of each point in the first laser point cloud data; the second feature descriptor is a feature descriptor of each point in the second laser point cloud data; the feature extraction network is obtained by training a deep learning network based on historical laser point cloud data;

[0012] According to the first feature descriptor and the second feature descriptor, the second laser point cloud data is transformed into the coordinate system of the first laser point cloud data;

[0013] Merging the transformed second laser point cloud data with the first laser point cloud data;

[0014] The merged laser point cloud data are grouped in pairs according to the order of time.

[0015] The embodiment of the present invention further provides a transmission tower model construction device, which is used to improve the efficiency and accuracy of transmission tower model construction and reduce the workload. The device includes:

[0016] An acquisition module is used to acquire laser point cloud data of transmission towers at different times;

[0017] The grouping module is used to group the laser point cloud data in pairs according to the order of time; wherein the laser point cloud data at two adjacent time points is a group; the grouped laser point cloud data includes the first laser point cloud data and the second laser point cloud data; the first laser point cloud data is the laser point cloud data at the earlier time point in each group; the second laser point cloud data is the laser point cloud data at the later time point in each group;

[0018] A merging module, used for performing repeated steps on each set of laser point cloud data until all laser point cloud data are merged into one laser point cloud data;

[0019] A model building module is used to build a transmission tower model based on the merged laser point cloud data and determine the three-dimensional model of the transmission tower;

[0020] The repeating steps include:

[0021] The first laser point cloud data and the second laser point cloud data are respectively input into a pre-trained feature extraction network to obtain a first feature descriptor and a second feature descriptor; the first feature descriptor is a feature descriptor of each point in the first laser point cloud data; the second feature descriptor is a feature descriptor of each point in the second laser point cloud data; the feature extraction network is obtained by training a deep learning network based on historical laser point cloud data;

[0022] According to the first feature descriptor and the second feature descriptor, the second laser point cloud data is transformed into the coordinate system of the first laser point cloud data;

[0023] Merging the transformed second laser point cloud data with the first laser point cloud data;

[0024] The merged laser point cloud data are grouped in pairs according to the order of time.

[0025] An embodiment of the present invention further provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned transmission tower model construction method when executing the computer program.

[0026] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned transmission tower model construction method is implemented.

[0027] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the above-mentioned transmission tower model construction method is implemented.

[0028] The transmission tower model construction method in the embodiment of the present invention obtains laser point cloud data of the transmission tower at different times; groups the laser point cloud data in pairs according to the order of the times; wherein the laser point cloud data at two adjacent times is a group; the grouped laser point cloud data includes first laser point cloud data and second laser point cloud data; the first laser point cloud data is the laser point cloud data at the earlier time in each group; the second laser point cloud data is the laser point cloud data at the later time in each group; repeating steps are performed on each group of laser point cloud data until all laser point cloud data are merged into one laser point cloud data; constructing the transmission tower model according to the merged laser point cloud data, and determining the three-dimensional model of the transmission tower; the repeating steps include: inputting the first laser point cloud data and the second laser point cloud data into a pre-trained In the trained feature extraction network, a first feature descriptor and a second feature descriptor are obtained; the first feature descriptor is a feature descriptor of each point in the first laser point cloud data; the second feature descriptor is a feature descriptor of each point in the second laser point cloud data; the feature extraction network is obtained by training a deep learning network based on historical laser point cloud data; according to the first feature descriptor and the second feature descriptor, the second laser point cloud data is transformed into the coordinate system of the first laser point cloud data; the transformed second laser point cloud data and the first laser point cloud data are merged; the merged laser point cloud data are grouped in pairs according to the sequence of time. Compared with the model construction method through manual measurement and two-dimensional drawings in the prior art, the efficiency and accuracy of transmission tower model construction can be improved and the workload can be reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0030] Figure 1 It is a schematic diagram of the flow of a method for constructing a transmission tower model in an embodiment of the present invention;

[0031] Figure 2 It is a flowchart of a specific example of a method for constructing a transmission tower model in an embodiment of the present invention;

[0032] Figure 3 Schematic diagram of a transmission tower model construction device according to an embodiment of the present invention;

[0033] Figure 4 A schematic diagram of a specific example of a transmission tower model construction device in an embodiment of the present invention;

[0034] Figure 5 Schematic diagram of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0035] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0036] In view of the problems of large workload, low efficiency and insufficient precision in the transmission tower modeling method in the prior art, an embodiment of the present invention proposes a transmission tower model construction method. The transmission tower model is constructed by using this method, which can improve the efficiency and precision of the transmission tower model construction and reduce the workload. Figure 1 FIG. 1 is a flow chart of a method for constructing a transmission tower model in an embodiment of the present invention. Figure 1 As shown, the method may include:

[0037] Step 101, obtaining laser point cloud data of a transmission tower at different times;

[0038] Step 102, grouping the laser point cloud data in pairs according to the order of time; wherein the laser point cloud data at two adjacent time points is a group; the grouped laser point cloud data includes the first laser point cloud data and the second laser point cloud data; the first laser point cloud data is the laser point cloud data at the earlier time point in each group; the second laser point cloud data is the laser point cloud data at the later time point in each group;

[0039] Step 103, performing repeated steps for each set of laser point cloud data until all laser point cloud data are merged into one laser point cloud data;

[0040] Step 104: construct a transmission tower model based on the merged laser point cloud data to determine a three-dimensional model of the transmission tower.

[0041] Figure 2 FIG. 1 is a flow chart of a specific example of a method for constructing a transmission tower model according to an embodiment of the present invention. Figure 2 As shown, the repeated steps in step 103 may include:

[0042] Step 1031, inputting the first laser point cloud data and the second laser point cloud data into a pre-trained feature extraction network respectively to obtain a first feature descriptor and a second feature descriptor; the first feature descriptor is a feature descriptor of each point in the first laser point cloud data; the second feature descriptor is a feature descriptor of each point in the second laser point cloud data; the feature extraction network is obtained by training a deep learning network based on historical laser point cloud data;

[0043] Step 1032, transforming the second laser point cloud data into the coordinate system of the first laser point cloud data according to the first feature descriptor and the second feature descriptor;

[0044] Step 1033, merging the transformed second laser point cloud data and the first laser point cloud data;

[0045] Step 1034, grouping the merged laser point cloud data in pairs according to the order of time.

[0046] In order to make the final transmission tower model more accurate, it is necessary to denoise and downsample the original data to reduce the number of points in the laser point cloud data and retain the key features of the laser point cloud data. In one embodiment, after acquiring the laser point cloud data of the transmission tower at different times, it can also include: denoising the laser point cloud data of the transmission tower at different times; downsampling the denoised laser point cloud data; the downsampling process includes filtering the points in the laser point cloud data to retain the key features of the laser point cloud data.

[0047] In one embodiment, obtaining laser point cloud data of a transmission tower at different times may include: using a three-dimensional laser scanner to obtain laser point cloud data of the transmission tower at different angles at different times; wherein the angle difference between the laser point cloud data at a later time and the initial laser point cloud data is greater than the angle difference between the laser point cloud data at a previous time and the initial laser point cloud data; the initial laser point cloud data is the first laser point cloud data collected by the three-dimensional laser scanner. It should be noted that: the embodiment of the present invention uses a three-dimensional laser scanner to scan the transmission tower to obtain laser point cloud data at different times. The three-dimensional laser scanner obtains laser point cloud data at different angles at different times, specifically: (angle difference between the laser point cloud data at time T and the initial laser point cloud data)>(angle difference between the laser point cloud data at time T-1 and the initial laser point cloud data). The initial laser point cloud data is the first laser point cloud data collected by the three-dimensional laser scanner.

[0048] In step 103, the laser point cloud data at different times are aligned and merged, which specifically includes steps 1031 to 1034, wherein the deep learning network inputted by the first laser point cloud data and the second laser point cloud data can be PointNet or DGCNN (Dynamic Graph CNN, dynamic graph edge convolutional network).

[0049] Before step 1031, a step of training a deep learning network based on historical laser point cloud data and corresponding feature descriptors may also be included. The method for determining the feature descriptors and the method for training, testing and verifying the deep learning network may use existing methods, which will not be repeated here.

[0050] In one embodiment, transforming the second laser point cloud data into the coordinate system of the first laser point cloud data according to the first feature descriptor and the second feature descriptor may include: comparing the first feature descriptor in the first feature descriptor with each feature descriptor in the second feature descriptor to determine the most similar feature descriptor; transforming the second laser point cloud data into the coordinate system of the first laser point cloud data according to the first feature descriptor and the most similar feature descriptor.

[0051] In this embodiment, transforming the second laser point cloud data into the coordinate system of the first laser point cloud data according to the first feature descriptor and the most similar feature descriptor may include: determining a transformation matrix that transforms the position of the most similar feature descriptor to the position of the first feature descriptor according to the first feature descriptor and the most similar feature descriptor; and transforming the second laser point cloud data into the coordinate system of the first laser point cloud data according to the transformation matrix.

[0052] The following is an example of laser point cloud data at eight consecutive moments: t, t+1, t+2, t+3, t+4, t+5, t+6, and t+7 to illustrate the alignment and merging method in step 103 of the embodiment of the present invention:

[0053] 1) Group the laser point cloud data at time t, t+1, t+2, t+3, t+4, t+5, t+6, and t+7 in pairs according to the order of acquisition time:

[0054] That is, the laser point cloud data at time t and time t+1 are one group; the laser point cloud data at time t+2 and time t+3 are one group; the laser point cloud data at time t+4 and time t+5 are one group; the laser point cloud data at time t+6 and time t+7 are one group;

[0055] 2) Inputting the laser point cloud data at time t and time t+1 into the feature extraction network to obtain the feature descriptors of each point in the laser point cloud data at time t and time t+1;

[0056] The first feature descriptor in the laser point cloud data at time t is compared with all feature descriptors in the laser point cloud data at time t+1, and the feature descriptor at time t+1 that is most similar to the first feature descriptor in the laser point cloud data at time t (i.e., the most similar feature descriptor) is obtained;

[0057] Calculate a first transformation matrix based on the positions of the first feature descriptor and the most similar feature descriptor in the laser point cloud data at time t;

[0058] The laser point cloud data at time t+1 is transformed into the coordinate system of the laser point cloud data at time t using the first transformation matrix;

[0059] Merge the transformed laser point cloud data at time t+1 and the laser point cloud data at time t to obtain the laser point cloud data at time t||(t+1);

[0060] Similarly, the laser point cloud data at time (t+2)||(t+3), the laser point cloud data at time (t+4)||(t+5), and the laser point cloud data at time (t+6)||(t+7) can be obtained;

[0061] 3) Group the merged laser point cloud data in pairs according to the order of acquisition time:

[0062] That is, the laser point cloud data at time t||(t+1) and the laser point cloud data at time (t+2)||(t+3) are a group; the laser point cloud data at time (t+4)||(t+5) and the laser point cloud data at time (t+6)||(t+7) are a group;

[0063] The laser point cloud data at time t||(t+1) and time (t+2)||(t+3) are input into the feature extraction network to obtain the feature descriptors of each point of the laser point cloud data at time t||(t+1) and time (t+2)||(t+3);

[0064] The first feature descriptor in the laser point cloud data at time t||(t+1) is compared with all feature descriptors in the laser point cloud data at time (t+2)||(t+3) to obtain the feature descriptor at time (t+2)||(t+3) that is most similar to the first feature descriptor in the laser point cloud data at time t||(t+1) (i.e., the most similar feature descriptor);

[0065] Calculate a second transformation matrix based on the positions of the first feature descriptor and the most similar feature descriptor in the laser point cloud data at time t||(t+1);

[0066] The laser point cloud data at time (t+2)||(t+3) is transformed into the coordinate system of the laser point cloud data at time t||(t+1) by using the second transformation matrix;

[0067] The transformed laser point cloud data at time (t+2)||(t+3) and the laser point cloud data at time t||(t+1) are combined to obtain the laser point cloud data at time t||(t+1)||(t+2)||(t+3);

[0068] Similarly, the laser point cloud data at time (t+4)||(t+5)||(t+6)||(t+7) can be obtained;

[0069] Input the laser point cloud data at time t||(t+1)||(t+2)||(t+3) and time (t+4)||(t+5)||(t+6)||(t+7) into the feature extraction network to obtain the feature descriptors of each point of the laser point cloud data at time t||(t+1)||(t+2)||(t+3) and time (t+4)||(t+5)||(t+6)||(t+7);

[0070] The first feature descriptor in the laser point cloud data at time t||(t+1)||(t+2)||(t+3) is compared with all feature descriptors in the laser point cloud data at time (t+4)||(t+5)||(t+6)||(t+7) to obtain the feature descriptor at time (t+4)||(t+5)||(t+6)||(t+7) that is most similar to the first feature descriptor in the laser point cloud data at time t||(t+1)||(t+2)||(t+3) (i.e., the most similar feature descriptor);

[0071] Calculate a third transformation matrix based on the positions of the first feature descriptor and the most similar feature descriptor in the laser point cloud data at time t||(t+1)||(t+2)||(t+3);

[0072] The laser point cloud data at time (t+4)||(t+5)||(t+6)||(t+7) is transformed into the coordinate system of the laser point cloud data at time t||(t+1)||(t+2)||(t+3) by using the third transformation matrix;

[0073] The transformed laser point cloud data at time (t+4)||(t+5)||(t+6)||(t+7) and the laser point cloud data at time t||(t+1)||(t+2)||(t+3) are merged to obtain the laser point cloud data at time t||(t+1)||(t+2)||(t+3)||(t+4)||(t+5)||(t+6)||(t+7); the laser point cloud data at time t||(t+1)||(t+2)||(t+3)||(t+4)||(t+5)||(t+6)||(t+7) is the final merged laser point cloud data.

[0074] The alignment and merging method of less than or more than 8 consecutive moments is similar, and will not be described in detail here. The consecutive moments can be the scanning frequency of the 3D laser scanner, or can be pre-set, which is not limited here.

[0075] In one embodiment, constructing a transmission tower model based on the merged laser point cloud data and determining the three-dimensional model of the transmission tower may include: removing duplicate points from the merged laser point cloud data; smoothing the laser point cloud data from which the duplicate points have been removed; constructing a transmission tower model based on the smoothed laser point cloud data and determining the three-dimensional model of the transmission tower. In order to improve the accuracy of the three-dimensional model of the transmission tower, it is necessary to remove duplicate points from the merged laser point cloud data, smooth the laser point cloud data from which the duplicate points have been removed, and reconstruct the surface of the smoothed laser point cloud data to obtain the three-dimensional model of the transmission tower.

[0076] The present invention also provides a transmission tower model construction device, as described in the following embodiments. Since the principle of the device to solve the problem is similar to the transmission tower model construction method, the implementation of the device can refer to the implementation of the transmission tower model construction method, and the repeated parts will not be repeated.

[0077] Figure 3 Schematic diagram of a transmission tower model construction device according to an embodiment of the present invention. Figure 3 As shown, the device comprises:

[0078] An acquisition module 301 is used to acquire laser point cloud data of a transmission tower at different times;

[0079] The grouping module 302 is used to group the laser point cloud data in pairs according to the order of time; wherein the laser point cloud data at two adjacent time points is a group; the grouped laser point cloud data includes the first laser point cloud data and the second laser point cloud data; the first laser point cloud data is the laser point cloud data at the earlier time point in each group; the second laser point cloud data is the laser point cloud data at the later time point in each group;

[0080] A merging module 303 is used to perform repeated steps on each set of laser point cloud data until all laser point cloud data are merged into one laser point cloud data;

[0081] The model building module 304 is used to build a transmission tower model according to the merged laser point cloud data to determine the three-dimensional model of the transmission tower.

[0082] The merging module 303 is specifically used for:

[0083] The first laser point cloud data and the second laser point cloud data are respectively input into a pre-trained feature extraction network to obtain a first feature descriptor and a second feature descriptor; the first feature descriptor is a feature descriptor of each point in the first laser point cloud data; the second feature descriptor is a feature descriptor of each point in the second laser point cloud data; the feature extraction network is obtained by training a deep learning network based on historical laser point cloud data;

[0084] According to the first feature descriptor and the second feature descriptor, the second laser point cloud data is transformed into the coordinate system of the first laser point cloud data;

[0085] Merging the transformed second laser point cloud data with the first laser point cloud data;

[0086] The merged laser point cloud data are grouped in pairs according to the order of time.

[0087] In one embodiment, the merging module 303 is specifically configured to:

[0088] Comparing the first feature descriptor in the first feature descriptor with each feature descriptor in the second feature descriptor to determine the most similar feature descriptor;

[0089] According to the first feature descriptor and the most similar feature descriptor, the second laser point cloud data is transformed into the coordinate system of the first laser point cloud data.

[0090] In one embodiment, the merging module 303 is specifically configured to:

[0091] Determine, according to the first feature descriptor and the most similar feature descriptor, a transformation matrix for transforming the position of the most similar feature descriptor to the position of the first feature descriptor;

[0092] According to the transformation matrix, the second laser point cloud data is transformed into the coordinate system of the first laser point cloud data.

[0093] Figure 4 Schematic diagram of a transmission tower model construction device according to an embodiment of the present invention. Figure 4 As shown, in one embodiment, the transmission tower model building device further includes: a preprocessing module 401, which is used to:

[0094] De-noising the laser point cloud data of transmission towers at different times;

[0095] The denoised laser point cloud data is downsampled; the downsampling process includes filtering the points in the laser point cloud data to retain the key features of the laser point cloud data.

[0096] In one embodiment, the acquisition module 301 is specifically configured to:

[0097] A three-dimensional laser scanner is used to obtain laser point cloud data of a transmission tower at different viewing angles at different times; wherein the viewing angle difference between the laser point cloud data at a later time and the initial laser point cloud data is greater than the viewing angle difference between the laser point cloud data at an earlier time and the initial laser point cloud data; the initial laser point cloud data is the first laser point cloud data collected by the three-dimensional laser scanner.

[0098] In one embodiment, the model building module 304 is specifically configured to:

[0099] Eliminate duplicate points in the merged laser point cloud data;

[0100] Smoothing the laser point cloud data to remove duplicate points;

[0101] The transmission tower model is constructed based on the smoothed laser point cloud data to determine the three-dimensional model of the transmission tower.

[0102] Figure 5 Schematic diagram of a computer device in an embodiment of the present invention. Figure 5 As shown, an embodiment of the present invention further provides a computer device 500, including a processor 501, a memory 502, and a computer program 503 stored in the memory 502 and executable on the processor 501, and the processor 501 implements the above-mentioned transmission tower model construction method when executing the computer program 503.

[0103] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned transmission tower model construction method is implemented.

[0104] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the above-mentioned transmission tower model construction method is implemented.

[0105] The transmission tower model construction method in the embodiment of the present invention obtains laser point cloud data of the transmission tower at different times; groups the laser point cloud data in pairs according to the order of the times; wherein the laser point cloud data at two adjacent times is a group; the grouped laser point cloud data includes first laser point cloud data and second laser point cloud data; the first laser point cloud data is the laser point cloud data at the earlier time in each group; the second laser point cloud data is the laser point cloud data at the later time in each group; repeating steps are performed on each group of laser point cloud data until all laser point cloud data are merged into one laser point cloud data; constructing the transmission tower model according to the merged laser point cloud data, and determining the three-dimensional model of the transmission tower; the repeating steps include: inputting the first laser point cloud data and the second laser point cloud data into a pre-trained In the trained feature extraction network, a first feature descriptor and a second feature descriptor are obtained; the first feature descriptor is a feature descriptor of each point in the first laser point cloud data; the second feature descriptor is a feature descriptor of each point in the second laser point cloud data; the feature extraction network is obtained by training a deep learning network based on historical laser point cloud data; according to the first feature descriptor and the second feature descriptor, the second laser point cloud data is transformed into the coordinate system of the first laser point cloud data; the transformed second laser point cloud data and the first laser point cloud data are merged; the merged laser point cloud data are grouped in pairs according to the sequence of time. Compared with the model construction method through manual measurement and two-dimensional drawings in the prior art, the efficiency and accuracy of transmission tower model construction can be improved and the workload can be reduced.

[0106] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0107] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0108] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0109] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0110] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. 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. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for constructing a transmission tower model, characterized in that: include: Obtain laser point cloud data of transmission towers at different times; The laser point cloud data are grouped in pairs according to the sequence of time; wherein the laser point cloud data at two adjacent time points is a group; the grouped laser point cloud data includes the first laser point cloud data and the second laser point cloud data; the first laser point cloud data is the laser point cloud data at the earlier time point in each group; the second laser point cloud data is the laser point cloud data at the later time point in each group; Repeating the steps for each set of laser point cloud data until all laser point cloud data are merged into one laser point cloud data; The transmission tower model is constructed based on the merged laser point cloud data to determine the three-dimensional model of the transmission tower; The repeating steps include: The first laser point cloud data and the second laser point cloud data are respectively input into a pre-trained feature extraction network to obtain a first feature descriptor and a second feature descriptor; the first feature descriptor is a feature descriptor of each point in the first laser point cloud data; the second feature descriptor is a feature descriptor of each point in the second laser point cloud data; the feature extraction network is obtained by training a deep learning network based on historical laser point cloud data; According to the first feature descriptor and the second feature descriptor, the second laser point cloud data is transformed into the coordinate system of the first laser point cloud data; Merging the transformed second laser point cloud data with the first laser point cloud data; The merged laser point cloud data are grouped in pairs according to the order of time.

2. The method for constructing a transmission tower model according to claim 1, characterized in that: According to the first feature descriptor and the second feature descriptor, the second laser point cloud data is transformed into a coordinate system of the first laser point cloud data, including: Comparing the first feature descriptor in the first feature descriptor with each feature descriptor in the second feature descriptor to determine the most similar feature descriptor; According to the first feature descriptor and the most similar feature descriptor, the second laser point cloud data is transformed into the coordinate system of the first laser point cloud data.

3. The method for constructing a transmission tower model according to claim 2, wherein: According to the first feature descriptor and the most similar feature descriptor, the second laser point cloud data is transformed into the coordinate system of the first laser point cloud data, including: Determine, according to the first feature descriptor and the most similar feature descriptor, a transformation matrix for transforming the position of the most similar feature descriptor to the position of the first feature descriptor; According to the transformation matrix, the second laser point cloud data is transformed into the coordinate system of the first laser point cloud data.

4. The method for constructing a transmission tower model according to claim 1, wherein: After obtaining the laser point cloud data of the transmission tower at different times, it also includes: De-noising the laser point cloud data of transmission towers at different times; The denoised laser point cloud data is downsampled; the downsampling process includes filtering the points in the laser point cloud data to retain the key features of the laser point cloud data.

5. The method for constructing a transmission tower model according to claim 1, wherein: Obtain laser point cloud data of transmission towers at different times, including: A three-dimensional laser scanner is used to obtain laser point cloud data of a transmission tower at different viewing angles at different times; wherein the viewing angle difference between the laser point cloud data at a later time and the initial laser point cloud data is greater than the viewing angle difference between the laser point cloud data at an earlier time and the initial laser point cloud data; the initial laser point cloud data is the first laser point cloud data collected by the three-dimensional laser scanner.

6. The method for constructing a transmission tower model according to claim 1, wherein: The transmission tower model is constructed based on the merged laser point cloud data to determine the three-dimensional model of the transmission tower, including: Eliminate duplicate points in the merged laser point cloud data; Smoothing the laser point cloud data to remove duplicate points; The transmission tower model is constructed based on the smoothed laser point cloud data to determine the three-dimensional model of the transmission tower.

7. A transmission tower model construction device, characterized in that: include: An acquisition module is used to acquire laser point cloud data of transmission towers at different times; The grouping module is used to group the laser point cloud data in pairs according to the order of time; wherein the laser point cloud data at two adjacent time points is a group; the grouped laser point cloud data includes the first laser point cloud data and the second laser point cloud data; the first laser point cloud data is the laser point cloud data at the earlier time point in each group; the second laser point cloud data is the laser point cloud data at the later time point in each group; A merging module, used for performing repeated steps on each set of laser point cloud data until all laser point cloud data are merged into one laser point cloud data; A model building module is used to build a transmission tower model based on the merged laser point cloud data and determine the three-dimensional model of the transmission tower; The repeating steps include: The first laser point cloud data and the second laser point cloud data are respectively input into a pre-trained feature extraction network to obtain a first feature descriptor and a second feature descriptor; the first feature descriptor is a feature descriptor of each point in the first laser point cloud data; the second feature descriptor is a feature descriptor of each point in the second laser point cloud data; the feature extraction network is obtained by training a deep learning network based on historical laser point cloud data; According to the first feature descriptor and the second feature descriptor, the second laser point cloud data is transformed into the coordinate system of the first laser point cloud data; Merging the transformed second laser point cloud data with the first laser point cloud data; The merged laser point cloud data are grouped in pairs according to the order of time.

8. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method for constructing a transmission tower model according to any one of claims 1 to 6 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 a processor, the method for constructing a transmission tower model according to any one of claims 1 to 6 is implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method for constructing a transmission tower model according to any one of claims 1 to 6 is implemented.

Images