Method and system for three-dimensional reconstruction of power transmission channels using multiple parallel laser lines

Abstract

The present application provides a method and system for three-dimensional reconstruction of a power transmission channel using multiple parallel line lasers, and relates to the field of three-dimensional reconstruction technology, including: assembling multiple parallel line laser transmitters and a monocular camera and performing position calibration; determining the device posture relationship and constructing a dynamic acquisition module; interactive three-dimensional reconstruction tasks, performing multiple parallel laser projections and imaging acquisition, and determining a laser image set; combining the three-dimensional reconstruction space to perform step-by-step multi-level image feature screening and processing, and performing feature three-dimensional reconstruction to determine a three-dimensional power transmission channel; performing terminal visualization and reverse guidance of the operation and maintenance management of the target power transmission channel. The present application can solve the technical problem of low efficiency in the prior art due to the use of a single line laser or point laser scanning in traditional three-dimensional reconstruction methods when processing complex or large structures. By combining a multi-parallel line laser transmitter with a monocular camera and performing precise position calibration, the efficiency of three-dimensional reconstruction is improved.

Description

Technical Field

[0001] The present application relates to the field of three-dimensional reconstruction technology, and in particular to a three-dimensional reconstruction method and system for a power transmission channel of multiple parallel laser lines. Background Art

[0002] The 3D reconstruction of power transmission channels is a process of collecting the actual spatial data of power transmission channels and using computer technology to perform 3D modeling and visualization. It is often used in the maintenance, monitoring, and planning and design of power systems. Among the existing 3D reconstruction technologies, single line laser scanning and point laser scanning are two common methods. They show good performance when dealing with small or simple structures, but their efficiency and accuracy are limited when facing complex or large structures.

[0003] Single-line laser scanning technology can only scan one point or one line at a time, so it takes a long time to scan large structures. Point laser scanning can quickly acquire a large number of data points, but when dealing with complex structures, the lack of point cloud density requires multiple scans and data processing to improve accuracy, which also increases time and cost. In complex environments, single-line laser or point laser scanning is easily affected by occlusion or uneven reflection, affecting the reconstruction results. In addition, single-line laser scanning or point laser scanning has limitations in spatial coverage, making it difficult to fully capture the three-dimensional information of an object.

[0004] In summary, the prior art has a technical problem of low efficiency in three-dimensional reconstruction when processing complex or large structures due to the slow scanning speed of traditional single line laser or point laser. Summary of the invention

[0005] The purpose of this application is to provide a three-dimensional reconstruction method and system for power transmission channels using multiple parallel line lasers, in order to solve the technical problem in the prior art of low efficiency of three-dimensional reconstruction when processing complex or large structures due to the slow scanning speed of traditional single line lasers or point lasers.

[0006] In view of the above problems, the present application provides a method and system for three-dimensional reconstruction of power transmission channels of multiple parallel line lasers.

[0007] In the first aspect, the present application provides a three-dimensional reconstruction method for a power transmission channel using multiple parallel line lasers, which is implemented by a three-dimensional reconstruction system for a power transmission channel using multiple parallel line lasers, wherein the three-dimensional reconstruction method for a power transmission channel using multiple parallel line lasers comprises: assembling a multi-parallel line laser transmitter and a monocular camera and performing position calibration, wherein the multi-parallel line laser transmitter and the monocular camera are assembled on a transmission tower based on a preset relative spatial position; based on the position calibration, combined with a triangular feature relationship, relative motion verification is performed with the laser projection domain as the center, the device posture relationship is determined, and a motion map is constructed. state acquisition module; interactive 3D reconstruction task, triggering the multi-parallel line laser transmitter to perform multi-parallel laser projection on the target power transmission channel, combining with the dynamic acquisition module to perform imaging acquisition based on posture relationship, and determining the laser image set; combining with the 3D reconstruction space, performing step-by-step multi-level image feature screening and processing on the laser image set to determine the feature pyramid, performing feature 3D reconstruction to determine the 3D power transmission channel, wherein the 3D reconstruction space is connected with a feature extraction unit based on the pyramid principle; performing terminal visualization display on the 3D power transmission channel, and reversely guiding the operation and maintenance management of the target power transmission channel.

[0008] In the second aspect, the present application also provides a three-dimensional reconstruction system for a power transmission channel of multiple parallel line lasers, which is used to execute the three-dimensional reconstruction method for a power transmission channel of multiple parallel line lasers as described in the first aspect, wherein the three-dimensional reconstruction system for a power transmission channel of multiple parallel line lasers includes: a position calibration module, the position calibration module is used to assemble a multi-parallel line laser transmitter and a monocular camera and perform position calibration, wherein the multi-parallel line laser transmitter and the monocular camera are assembled on a transmission tower based on a preset relative spatial position; a relative motion verification module, the relative motion verification module is used to perform relative motion verification based on the position calibration, combined with a triangular feature relationship, with the laser projection domain as the center, determine the device posture relationship, and construct a dynamic acquisition module; an imaging acquisition module, the The imaging acquisition module is used for interactive three-dimensional reconstruction tasks, triggering the multi-parallel line laser transmitter to perform multi-parallel laser projection on the target transmission channel, combining with the dynamic acquisition module to perform imaging acquisition based on posture relationship, and determining the laser image set; the transmission channel determination module, the transmission channel determination module is used to combine the three-dimensional reconstruction space, perform step-by-step multi-level image feature screening and processing on the laser image set to determine the feature pyramid, perform feature three-dimensional reconstruction to determine the three-dimensional transmission channel, wherein the three-dimensional reconstruction space is connected to a feature extraction unit based on the pyramid principle; the operation and maintenance management module, the operation and maintenance management module is used to perform terminal visualization display on the three-dimensional transmission channel, and reversely guide the operation and maintenance management of the target transmission channel.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages: By assembling a multi-parallel line laser transmitter and a monocular camera and performing position calibration, wherein the multi-parallel line laser transmitter and the monocular camera are assembled on the transmission tower based on a preset relative spatial position; based on the position calibration, combined with the triangular feature relationship, relative motion verification is performed with the laser projection domain as the center, the device posture relationship is determined, and a dynamic acquisition module is constructed; an interactive three-dimensional reconstruction task is performed, the multi-parallel line laser transmitter is triggered to perform multi-parallel laser projection on the target transmission channel, and imaging acquisition based on the posture relationship is performed in combination with the dynamic acquisition module to determine the laser image set; combined with the three-dimensional reconstruction space, the laser image set is subjected to step-by-step multi-level image feature screening and processing to determine the feature pyramid, and feature three-dimensional reconstruction is performed to determine the three-dimensional transmission channel, wherein the three-dimensional reconstruction space is connected with a feature extraction unit based on the pyramid principle; the three-dimensional transmission channel is terminal visualized, and the operation and maintenance management of the target transmission channel is reversely guided. In other words, through the combination of a multi-parallel line laser transmitter and a monocular camera, accurate position calibration, dynamic acquisition module and feature pyramid technology can achieve more accurate three-dimensional reconstruction in a complex environment, and improve the efficiency of three-dimensional reconstruction of the transmission channel.

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0012] Figure 1 A schematic diagram of the flow of a three-dimensional reconstruction method of a transmission channel of multiple parallel laser lines of this application; Figure 2 This is a schematic diagram of the structure of the three-dimensional reconstruction system of the transmission channel of multiple parallel line lasers in this application.

[0013] Explanation of reference numerals: position calibration module 11 , relative motion verification module 12 , imaging acquisition module 13 , power transmission channel determination module 14 , operation and maintenance management module 15 . DETAILED DESCRIPTION

[0014] This application solves the technical problem in the prior art that the efficiency of three-dimensional reconstruction is low when processing complex or large structures due to the slow scanning speed of traditional single-line laser or point laser by providing a method and system for three-dimensional reconstruction of power transmission channels using multiple parallel-line lasers. Through the combination of multi-parallel-line laser transmitters and monocular cameras, precise position calibration, dynamic acquisition modules and feature pyramid technology can achieve more accurate three-dimensional reconstruction in complex environments, thereby improving the efficiency of three-dimensional reconstruction of power transmission channels.

[0015] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0016] For example, please refer to the attached Figure 1 The present application provides a three-dimensional reconstruction method for a power transmission channel of multiple parallel line lasers, wherein the three-dimensional reconstruction method for a power transmission channel of multiple parallel line lasers is applied to a three-dimensional reconstruction system for a power transmission channel of multiple parallel line lasers, and the three-dimensional reconstruction method for a power transmission channel of multiple parallel line lasers specifically comprises the following steps: Step 1: Assemble a multi-parallel line laser transmitter and a monocular camera and perform position calibration, wherein the multi-parallel line laser transmitter and the monocular camera are assembled on the transmission tower based on a preset relative spatial position.

[0017] Specifically, the multi-parallel line laser emitter and the monocular camera are fixed on the transmission tower according to the preset relative spatial position to ensure the accuracy of the relative position between the two. The multi-parallel line laser emitter is a device that can emit multiple parallel laser lines, which is usually used for three-dimensional scanning and measurement. Each laser line forms a line on the surface of the object. After being captured by the camera, these lines form a set of parallel light rays in space, which are used to scan and obtain the three-dimensional information of the target object. The monocular camera is a camera that uses only one lens to capture images. It cannot directly obtain the depth information of the object, but can indirectly achieve three-dimensional reconstruction by cooperating with the laser emitter. In three-dimensional reconstruction, the monocular camera is used to capture the projection of the laser line on the target object, thereby helping to determine the three-dimensional position of the object. The spatial position and posture relationship between the laser emitter and the camera are determined by standard methods (such as checkerboard calibration) to ensure the accurate correspondence between laser projection and image acquisition. For example, assume that a multi-parallel line laser emitter and a monocular camera are installed on a transmission tower. The relative position between the laser emitter and the camera is determined according to the design requirements. The laser emitter is located 1 meter in front of the camera and at the same height. Then, high-precision measuring tools (such as laser rangefinders and angle meters) are used to adjust the actual position of the equipment until the preset requirements are met. The distance between the laser emitter and the camera is 1000 mm, and the error does not exceed 1 mm. By assembling these devices on the transmission tower, the transmission lines in the sky can be remotely and dynamically monitored and reconstructed in three dimensions without being affected by obstacles on the ground, so that the laser scanning data and the camera image data can be accurately matched, thereby improving the accuracy and reliability of the three-dimensional reconstruction.

[0018] Step 2: Based on the position calibration and combined with the triangular feature relationship, relative motion verification is performed with the laser projection domain as the center, the device posture relationship is determined, and a dynamic acquisition module is constructed.

[0019] Specifically, according to the position calibration and triangulated feature relationship, the laser projection domain is taken as the center, and all motion verification and image acquisition are carried out around the area where the laser is projected onto the object. The preset distribution characteristics of the laser projection part and the background part in the image are used to determine the layout structure of the image, and this is used as a constraint condition to identify and extract the key information of the laser projection part, while suppressing or ignoring the interference information of the background part. Combined with the triangulated feature relationship, that is, the relative position and posture relationship between the laser emitter and the camera, as well as the geometric features of the laser projection on the target object, are used to determine the device posture relationship. The triangulated feature relationship is based on the principle of triangulation, and the three-dimensional position of the object surface is calculated by analyzing the projection point of the laser on the object and the image captured by the camera. The posture relationship between the devices is verified by controlling the relative movement of the laser emitter and the camera. For example, rotate or translate the laser emitter while keeping the camera stable, or vice versa, and observe the changes of the laser line in the camera image. With the laser projection domain as the center, the relative motion relationship of the device in space is verified by analyzing the position and angle changes of the laser line in the camera image. The device posture relationship refers to the relative position and posture relationship between the multi-parallel line laser emitter and the monocular camera in space. The dynamic acquisition module is a specially designed hardware and software system for dynamically acquiring and processing data. It can automatically adjust acquisition parameters according to different environments and conditions to ensure the quality and accuracy of the acquired data. In the acquired image, the position of the laser line in the image is fixed because it is constrained by the device posture relationship. Through relative motion verification, the posture of the laser emitter and the camera is precisely controlled to ensure that the acquired data is accurate. The construction of the dynamic acquisition module ensures the stability of the data acquired during relative motion and reduces the error caused by motion.

[0020] Step 3: Interactive 3D reconstruction task, triggering the multi-parallel line laser transmitter to perform multi-parallel laser projection on the target power transmission channel, combining with the dynamic acquisition module to perform imaging acquisition based on posture relationship, and determining the laser image set.

[0021] Specifically, the laser emitter is activated by the control system to project laser light onto the target transmission channel. This can be triggered manually or automatically according to a preset program or external instruction. For example, the operation and maintenance personnel regularly reconstruct the transmission channel in three dimensions, or the system automatically triggers the reconstruction task according to preset conditions. The laser emitter emits multiple parallel laser lines according to preset parameters, covering various parts of the target transmission channel, which is convenient for subsequent imaging acquisition. The dynamic acquisition module will perform imaging acquisition according to the device posture relationship, that is, according to the relative position and posture relationship between the laser emitter and the camera, the acquisition parameters are automatically adjusted to ensure that the collected image data is accurate, which means that the camera's shooting angle, focal length and other parameters are adjusted according to the position and direction of the laser emitter. A laser image set refers to a set of camera images containing laser points or lines formed by laser projection on a target object. Through multi-parallel laser projection and dynamic acquisition modules, a high-quality laser image set can be obtained, providing a reliable data basis for subsequent three-dimensional reconstruction.

[0022] Step 4: Combined with the three-dimensional reconstruction space, the laser image set is subjected to step-by-step multi-level image feature screening and processing to determine a feature pyramid, and feature three-dimensional reconstruction is performed to determine a three-dimensional power transmission channel, wherein the three-dimensional reconstruction space is connected to a feature extraction unit based on the pyramid principle.

[0023] Specifically, in the 3D reconstruction space, the reconstruction element sequence is set according to the 3D reconstruction task. The laser image set is preliminarily processed and segmented into the projection part and the background part according to the image layout structure. The background part can be selectively extracted or removed to emphasize the main features. According to the reconstruction element sequence, the laser image set is subjected to multi-level feature extraction and screening to gradually extract feature information useful for 3D reconstruction. Feature pyramid refers to the hierarchical organization of processed features to form a hierarchical structure, in which each layer contains feature information of different levels. Basic features such as low-level features such as edges and corners are extracted from the laser image, and the screened features are strengthened step by step. Each level is based on the results of the previous level for more refined processing. For key features, algorithms such as scale-invariant feature transform (SIFT) and speeded up robust features (SURF) are used for enhancement. For non-key features, methods such as principal component analysis (PCA) are used for dimensionality reduction to reduce computational complexity. The features processed at each level are mapped into the 3D space to construct a feature pyramid, in which the bottom layer of the pyramid is the original image features and the top layer is the highly abstract features.

[0024] The three-dimensional point cloud is constructed in the three-dimensional reconstruction space by using the features at each level in the feature pyramid. Specifically, the features at each level in the first feature pyramid are integrated by using the spatial phase relationship. The position of each feature point in the three-dimensional space is determined through the integrated feature pyramid to form a three-dimensional point cloud feature distribution. The feature distribution of the three-dimensional point cloud is analyzed to identify the key structure and features of the transmission channel, and a three-dimensional model of the first local transmission channel is constructed in the three-dimensional reconstruction space. The laser image set is traversed, and the above steps are repeated to construct three-dimensional models of multiple local transmission channels until the Nth local transmission channel. Based on the spatial phase relationship, the first local transmission channel to the Nth local transmission channel are spliced ​​to form a complete three-dimensional model of the transmission channel. The three-dimensional reconstruction space is connected with a feature extraction unit based on the pyramid principle, and the feature extraction unit refers to integrating the feature extraction unit based on the pyramid principle into the three-dimensional reconstruction space so as to extract and process features in the three-dimensional space. By combining the three-dimensional reconstruction space, the laser image set is subjected to step-by-step multi-level image feature screening and processing to determine the feature pyramid, and feature three-dimensional reconstruction is performed to determine the three-dimensional transmission channel, thereby achieving three-dimensional reconstruction of the target transmission channel. Through step-by-step multi-level image feature screening and processing, the features in the laser image are accurately represented, thereby improving the accuracy of three-dimensional reconstruction.

[0025] Step 5: Perform terminal visualization display on the three-dimensional power transmission channel, and reversely guide the operation and maintenance management of the target power transmission channel.

[0026] Specifically, the reconstructed three-dimensional transmission channel model is rendered through professional three-dimensional visualization software, including lighting effects, texture mapping, color adjustment, etc., so that it can be displayed in an intuitive three-dimensional form on the terminal device. Users can view the three-dimensional model from different angles and scales through operations such as rotation, scaling, and translation to better understand the structure and status of the transmission channel. The names, numbers, technical parameters and other information of key components are marked on the three-dimensional model, and areas where problems may occur, such as wear, corrosion, and deformation, are highlighted. Through the three-dimensional model, operation and maintenance personnel can quickly locate the fault point, analyze the cause of the fault, and formulate a maintenance plan to help technicians conduct detailed inspections of the transmission channel without physical contact. Using the three-dimensional model, maintenance work can be planned more accurately, including maintenance time, required materials, staffing, etc., to discover possible problems in advance and reduce risks in actual operations. The data collected during the operation and maintenance process compares the differences between the actual operating status and the design model, and is used to optimize the design of the transmission channel and improve its performance and reliability. The 3D visualization display enables operation and maintenance personnel to quickly understand and analyze the status of the transmission channel, improve work efficiency, identify potential fault points in advance, reduce unexpected downtime, and reduce unnecessary repair and replacement costs.

[0027] Further, step 2 of this application includes: The triangular feature relationship is determined based on the laser projection area and the position calibration, wherein the multi-parallel line laser emitter and the monocular camera have a movement based on the spatial perspective; the image layout structure is determined, wherein the image layout structure is determined by the preset distribution characteristics of the projection part and the background part; with the image layout structure as a constraint, the triangular feature relationship is combined to perform relative motion verification to mine the device posture relationship, wherein the device posture relationship is the relationship between the monocular camera's follow-up posture under the multi-parallel line laser emitter's perspective posture movement.

[0028] Specifically, the triangular feature relationship and image layout structure are used to determine the posture relationship between the multi-parallel line laser emitter and the monocular camera. Among them, the triangular feature relationship is determined by the laser projection area and position calibration; the second is to mine the device posture relationship according to the image layout structure. The triangular feature relationship is a principle based on triangulation, which measures the position of the same object through two observation points at different positions. After the calibration is completed, when the laser is projected onto the object, a series of light points will be formed on the surface of the object, and the camera will capture the image of these points from another angle. The three-dimensional coordinates of each point are calculated by the image position of these points and the known relative position relationship between the laser emitter and the camera. The laser projection area is the spatial area where the laser emitter projects the laser line. In this area, the laser line intersects with the target object to form data points that can be used for three-dimensional reconstruction. The movement of the spatial perspective refers to the relative position change of the laser emitter and the camera in space. It can be mechanical, such as changing the perspective by moving the device, or it can be fixed, such as simulating the change of perspective by changing the focal length of the lens.

[0029] The layout structure of the image is determined according to the preset distribution characteristics of the projection part and the background part, that is, the spatial distribution and arrangement of different parts in the image (such as laser projection and background) are determined. The projection part usually refers to the projection of the laser line on the target object, while the background part refers to the other areas in the image except the laser projection. Combining the image layout structure with the triangular feature relationship, the relative movement of the laser emitter and the camera under different postures is analyzed. During the movement, the perspective posture of the laser emitter affects the position of the projected laser line on the object, and the follow-up posture of the camera determines the position of these laser lines in the image. The device posture relationship refers to the relative position and direction relationship between the laser emitter and the camera in space. The perspective posture movement refers to the change of the perspective of the laser emitter and the camera in space. The laser emitter projects the laser to the transmission channel, and the camera captures the projection of the laser line on the channel. The position and width of the laser line in the image are set as the constraints of the image layout structure. When the laser emitter moves in the perspective posture, the camera adjusts its posture to keep the position of the laser line in the image stable. For example, when the laser transmitter moves up 10 degrees, the camera adjusts its pitch angle accordingly to keep the laser line in the center of the image. By combining the triangulated feature relationship and the image layout structure, the posture relationship of the device can be accurately determined, thereby improving the accuracy of 3D reconstruction.

[0030] Furthermore, the present application also includes the following steps: Determine the acquisition free parallax, wherein the acquisition free parallax is the allowed overlapping field of view of the upper acquisition node and the lower acquisition node; use the device posture relationship as a single-point acquisition constraint and the acquisition free parallax as a continuous scanning constraint to perform continuous scanning projection and image acquisition management on the target power transmission channel.

[0031] Specifically, acquisition free parallax means that in a multi-node acquisition system, different acquisition nodes (such as different laser emitter and camera combinations) have partially overlapping fields of view when scanning. This overlapping area ensures that the data collected by different nodes can be smoothly transitioned during the scanning process, avoiding information loss. By setting a suitable acquisition free parallax, it can be ensured that there is sufficient overlap between the data collected at different positions, thereby improving the continuity and integrity of the three-dimensional reconstruction. The upper acquisition node and the lower acquisition node refer to acquisition devices at different positions in the vertical or horizontal direction. The upper acquisition node may be located at a higher position, while the lower acquisition node is located at a lower position.

[0032] Taking the device posture relationship as a single-point acquisition constraint ensures that the relative posture between devices is known at each acquisition, thereby improving the accuracy of data acquisition. When collecting data at each point, the device posture relationship, that is, the relative position and direction of the laser emitter and the camera, needs to be taken into account to ensure the accuracy of data at a single acquisition point. During continuous scanning, it is necessary to follow the rules of acquisition free parallax to ensure that the entire scanning process is smooth and without omissions, which helps to establish good data continuity between multiple acquisition nodes. For example, assuming that the length of the transmission channel is 1000 meters, two acquisition nodes are set, located at 500 meters and 600 meters of the channel respectively. Setting the acquisition free parallax to 50 meters means that the overlapping field of view between the two nodes is 50 meters. During the scanning process, the laser emitter and the camera perform single-point acquisition according to the device posture relationship, while complying with the acquisition free parallax as a constraint for continuous scanning. When the first node scans to 450 meters, the second node starts scanning until it scans to 650 meters, ensuring a smooth transition of data between the two nodes.

[0033] Analyze the characteristics of the target transmission channel, including its size, shape, environmental conditions, etc., and determine the performance indicators that need to be met, such as acquisition speed, accuracy, resolution, etc. Design the basic architecture of the module, including hardware selection (such as laser emitters, cameras, sensors, etc.) and software framework (such as control algorithms, data processing processes, etc.). Select appropriate hardware components, such as high-precision laser emitters, high-resolution cameras, stable brackets, etc., integrate the components into a unified system, and conduct preliminary tests to ensure that they can work together. Collect a large amount of representative data, including laser scanning data and corresponding camera image data under different conditions, such as different lighting conditions, weather conditions, and the geometry of the transmission line. Preprocess the collected data, including denoising, normalization, data enhancement, etc., to improve the quality of the data and the training effect of the model. Select an appropriate machine learning or deep learning model based on the characteristics of the reconstruction task. Use the preprocessed data to train the model. During the training process, the parameters of the model, such as learning rate, batch size, etc., need to be adjusted to optimize the performance of the model. The dynamic acquisition module is a specially designed hardware and software system for dynamically acquiring and processing data, usually including sensors, data acquisition cards, computers and corresponding software algorithms. The dynamic acquisition module can automatically adjust acquisition parameters such as sampling frequency and sampling time according to different environments and conditions to ensure the quality and accuracy of the acquired data. By acquiring free parallax and continuous scanning constraints, the system ensures continuous and consistent 3D data between different acquisition nodes.

[0034] Furthermore, the present application also includes the following steps: Determine a preset laser projection state and a preset image state; configure laser parameters based on the preset laser projection state, wherein the same projection state based on the difference in projection viewing angle distance is used as a standard; configure camera parameters based on the preset image state, wherein the camera parameters include internal parameters and external parameters, wherein the same image state based on the difference in acquisition viewing angle distance is used as a standard.

[0035] Specifically, the laser projection standard state is preset according to the projection viewing angle spacing, and the image acquisition standard state is preset according to the acquisition viewing angle spacing. The preset laser projection state refers to the projection parameters of the laser transmitter, including the number, spacing, projection angle, etc. of laser lines, which are preset according to the characteristics of the target power transmission channel and the required measurement accuracy before starting the 3D reconstruction task. The laser parameters are configured based on the same projection state with different projection viewing angle spacing. Under the same projection state, the laser parameters will be adjusted according to the different projection distances to compensate for the laser diffusion and weakening of light intensity due to the increase in distance. For example, if the projection distance increases, adjust the focus of the laser transmitter, increase the intensity of the laser, or adjust the spacing of the laser lines to ensure that the laser points or lines formed on the target are still clearly discernible.

[0036] The preset image state refers to a series of parameters set for the monocular camera according to the characteristics of the laser projection state and the target power transmission channel in the 3D reconstruction task, including internal parameters (such as focal length, aperture, shutter speed) and external parameters (such as the position and direction of the camera). Based on the preset image state, the camera parameters are configured, where the camera parameters include internal parameters and external parameters, with the same image state under the difference in acquisition viewing angle distance as the standard. Under the same image state, the camera parameters will be adjusted according to the different acquisition distances to compensate for image blur or underexposure caused by increased distance. By presetting the laser projection state and image state, a unified standard is established, and the laser and camera parameters are adjusted based on the distance, which optimizes the parameter configuration during the acquisition process and improves the quality of the data.

[0037] Further, step 4 of this application includes: The three-dimensional reconstruction task is identified, and a reconstruction element sequence is set, wherein each reconstruction element sequence includes element feature, feature dimension, and feature intensity; a first laser image is received and identified, and based on the image layout structure, boundary segmentation and primary feature extraction are performed on the first laser image to determine a first feature layer, wherein the segmentation boundary is an adjacent boundary between a projection part and a background part; based on the reconstruction element sequence, a multi-level image feature screening process is performed on the first feature layer to determine a first feature pyramid.

[0038] Furthermore, the present application also includes the following steps: A pyramid architecture is constructed by multi-level feature enhancement; the first feature layer is traversed, and feature feature transformation and enhancement processing is performed based on the reconstructed feature sequence until the Nth feature layer is determined; inter-layer mapping is performed on the first feature layer to the Nth feature layer to generate the first feature pyramid.

[0039] Specifically, the identification of three-dimensional reconstruction tasks includes modeling of transmission lines, towers or other related structures, and determining the target and scope of reconstruction. The various elements to be processed (such as laser lines, edges of objects, etc.) are arranged in a certain order, and each element contains its features, feature dimensions and feature strengths. Feature features are specific features of objects in laser images, such as shape, edge, texture, etc.; feature dimensions refer to the spatial dimensions of the features, such as the coordinate position in three-dimensional space; feature strength refers to the prominence or intensity of the features, which can usually be measured by the intensity of laser reflection. For example, for the reconstruction of transmission lines, feature features may include the diameter, material, curvature, etc. of the line, feature dimensions may include length, width, height, etc., and feature strength may include the strength and stability of the line. A laser image collected from a multi-parallel line laser transmitter and a monocular camera is used as the first laser image, and the projection part and the background part are segmented according to the set image layout structure. The segmentation boundary is usually the adjacent boundary between the projection part and the background part. After the boundary segmentation of the first laser image, useful features of the laser projection part, such as the position, shape, and intensity of the laser line, are extracted to form the first feature layer. The background part can be extracted weakly, extracted selectively or removed completely to reduce interference with subsequent processing.

[0040] When constructing a feature pyramid, the importance of a feature is positively correlated with the level of enhancement. Feature enhancement includes dimensionality transformation, perspective transformation, and magnification enhancement processing for strongly correlated features; for weakly correlated features, dimensionality reduction and weakening processing are performed. Dimension transformation refers to the reorganization of features in different coordinate systems or reference frames, such as mapping three-dimensional features to a two-dimensional plane; perspective transformation refers to changing the observation angle to obtain more comprehensive feature information; magnification enhancement processing refers to enhancing the saliency of features to make them easier to identify and use during reconstruction. The processed features are organized in layers to form a hierarchical structure, namely the pyramid architecture, in which each layer contains feature information at different levels.

[0041] Traverse the first feature layer, and transform and enhance each feature according to the feature features, feature dimensions, and feature strengths of the reconstructed feature sequence, including dimensional transformation, perspective transformation, and magnification enhancement for strongly correlated features, and weakening treatments such as dimensionality reduction for weakly correlated features. Check each feature in the first feature layer and perform corresponding enhancement processing according to its importance. Repeat the above process until the required feature enhancement level is reached to form the Nth feature layer. Map the features between the first feature layer and the Nth feature layer, establish a hierarchical relationship, and generate a feature pyramid. Inter-layer mapping refers to determining the mapping relationship between adjacent feature layers during the construction of the feature pyramid to facilitate subsequent three-dimensional reconstruction. Through multi-level enhancement, important features are enhanced, and the establishment of inter-layer mapping and pyramid architecture provides more refined and structured data for the reconstruction process, thereby improving the accuracy of reconstruction.

[0042] Furthermore, the present application also includes the following steps: Based on the spatial phase relationship, the first feature pyramid is integrated to determine the feature distribution of the three-dimensional point cloud; the feature distribution of the three-dimensional point cloud is identified, three-dimensional reconstruction is performed in the three-dimensional reconstruction space, and a first local transmission channel is determined; the laser image set is traversed to construct the Nth local transmission channel; based on the spatial phase relationship, the first local transmission channel to the Nth local transmission channel are spliced ​​to generate the three-dimensional transmission channel.

[0043] Specifically, the spatial phase relationship is used to integrate the features of each level in the first feature pyramid. The spatial phase relationship refers to the relative position and direction relationship of the features in three-dimensional space. Through the integrated feature pyramid, the position of each feature point in three-dimensional space is determined to form a three-dimensional point cloud feature distribution. The three-dimensional point cloud feature distribution refers to the distribution of feature points in three-dimensional space. The distribution pattern of feature points in three-dimensional space is analyzed to identify the key structures and features of the transmission channel, such as towers, conductors, insulators, etc. The structure of the transmission channel is reconstructed in three-dimensional space using the identified three-dimensional point cloud feature distribution. According to the feature pyramid and the three-dimensional point cloud feature distribution, each image in the laser image set is processed one by one to construct a three-dimensional model of multiple local transmission channels until the Nth local transmission channel. Each laser image is analyzed to extract new local features, and the existing three-dimensional model is continuously superimposed and expanded. By analyzing the spatial phase relationship between the first local transmission channel and the Nth local transmission channel, they are spliced ​​to ensure seamless connection between different local channels and generate a complete three-dimensional transmission channel. Specifically, the features of the first local transmission channel are integrated with the features of the Nth local channel to generate a new three-dimensional point cloud distribution; based on the integrated point cloud features, the structure of the transmission channel is reconstructed to ensure the accuracy of the height and diameter; by analyzing the spatial phase relationship, the first local transmission channel is ensured to be accurately connected with the Nth local transmission channel in space to form a complete three-dimensional transmission channel. By integrating the feature pyramid based on the spatial phase relationship, the distribution of three-dimensional point cloud features can be determined more accurately, thereby improving the accuracy of three-dimensional reconstruction. By traversing the laser image set and splicing the local transmission channels, a complete three-dimensional transmission channel can be quickly constructed, thereby improving the efficiency of three-dimensional reconstruction.

[0044] In summary, the method for 3D reconstruction of power transmission channels using multiple parallel laser lines provided in this application has the following technical effects: By assembling a multi-parallel line laser transmitter and a monocular camera and performing position calibration, wherein the multi-parallel line laser transmitter and the monocular camera are assembled on the transmission tower based on a preset relative spatial position; based on the position calibration, combined with the triangular feature relationship, relative motion verification is performed with the laser projection domain as the center, the device posture relationship is determined, and a dynamic acquisition module is constructed; an interactive three-dimensional reconstruction task is performed, the multi-parallel line laser transmitter is triggered to perform multi-parallel laser projection on the target transmission channel, and imaging acquisition based on the posture relationship is performed in combination with the dynamic acquisition module to determine the laser image set; combined with the three-dimensional reconstruction space, the laser image set is subjected to step-by-step multi-level image feature screening and processing to determine the feature pyramid, and feature three-dimensional reconstruction is performed to determine the three-dimensional transmission channel, wherein the three-dimensional reconstruction space is connected with a feature extraction unit based on the pyramid principle; the three-dimensional transmission channel is terminal visualized, and the operation and maintenance management of the target transmission channel is reversely guided. In other words, through the combination of a multi-parallel line laser transmitter and a monocular camera, accurate position calibration, dynamic acquisition module and feature pyramid technology can achieve more accurate three-dimensional reconstruction in a complex environment, and improve the efficiency of three-dimensional reconstruction of the transmission channel.

[0045] Embodiment 2: Based on the same inventive concept as the method for 3D reconstruction of a power transmission channel using multiple parallel laser lines in the aforementioned embodiment 1, the present application also provides a 3D reconstruction system for a power transmission channel using multiple parallel laser lines. Figure 2 , the transmission channel three-dimensional reconstruction system of multiple parallel laser lines includes: The position calibration module 11 is used to assemble a multi-parallel line laser transmitter and a monocular camera and perform position calibration, wherein the multi-parallel line laser transmitter and the monocular camera are assembled on the transmission tower based on a preset relative spatial position.

[0046] The relative motion verification module 12 is used to perform relative motion verification based on the position calibration and in combination with the triangular feature relationship, with the laser projection domain as the center, determine the device posture relationship, and construct a dynamic acquisition module.

[0047] The imaging acquisition module 13 is used for interactive three-dimensional reconstruction tasks, triggering the multi-parallel line laser transmitter to perform multi-parallel laser projection on the target power transmission channel, combining with the dynamic acquisition module to perform imaging acquisition based on posture relationship, and determining the laser image set.

[0048] The transmission channel determination module 14 is used to combine the three-dimensional reconstruction space, perform step-by-step multi-level image feature screening and processing on the laser image set to determine the feature pyramid, and perform feature three-dimensional reconstruction to determine the three-dimensional transmission channel, wherein the three-dimensional reconstruction space is connected to a feature extraction unit based on the pyramid principle.

[0049] The operation and maintenance management module 15 is used to perform terminal visualization display of the three-dimensional power transmission channel and reversely guide the operation and maintenance management of the target power transmission channel.

[0050] Furthermore, the relative motion verification module 12 in the multi-parallel line laser transmission channel three-dimensional reconstruction system is also used for: The triangular feature relationship is determined based on the laser projection area and the position calibration, wherein the multi-parallel line laser emitter and the monocular camera have a movement based on the spatial perspective; the image layout structure is determined, wherein the image layout structure is determined by the preset distribution characteristics of the projection part and the background part; with the image layout structure as a constraint, the triangular feature relationship is combined to perform relative motion verification to mine the device posture relationship, wherein the device posture relationship is the relationship between the monocular camera's follow-up posture under the multi-parallel line laser emitter's perspective posture movement.

[0051] Furthermore, the relative motion verification module 12 in the multi-parallel line laser transmission channel three-dimensional reconstruction system is also used for: Determine the acquisition free parallax, wherein the acquisition free parallax is the allowed overlapping field of view of the upper acquisition node and the lower acquisition node; use the device posture relationship as a single-point acquisition constraint and the acquisition free parallax as a continuous scanning constraint to perform continuous scanning projection and image acquisition management on the target power transmission channel.

[0052] Furthermore, the three-dimensional reconstruction system of the power transmission channel of the multi-parallel line laser also includes a device configuration module for: Determine a preset laser projection state and a preset image state; configure laser parameters based on the preset laser projection state, wherein the same projection state based on the difference in projection viewing angle distance is used as a standard; configure camera parameters based on the preset image state, wherein the camera parameters include internal parameters and external parameters, wherein the same image state based on the difference in acquisition viewing angle distance is used as a standard.

[0053] Furthermore, the transmission channel determination module 14 in the transmission channel 3D reconstruction system of multiple parallel laser lines is also used for: The three-dimensional reconstruction task is identified, and a reconstruction element sequence is set, wherein each reconstruction element sequence includes element feature, feature dimension, and feature intensity; a first laser image is received and identified, and based on the image layout structure, boundary segmentation and primary feature extraction are performed on the first laser image to determine a first feature layer, wherein the segmentation boundary is an adjacent boundary between a projection part and a background part; based on the reconstruction element sequence, a multi-level image feature screening process is performed on the first feature layer to determine a first feature pyramid.

[0054] Furthermore, the transmission channel determination module 14 in the transmission channel 3D reconstruction system of multiple parallel laser lines is also used for: A pyramid architecture is constructed by multi-level feature enhancement; the first feature layer is traversed, and feature feature transformation and enhancement processing is performed based on the reconstructed feature sequence until the Nth feature layer is determined; inter-layer mapping is performed on the first feature layer to the Nth feature layer to generate the first feature pyramid.

[0055] Furthermore, the transmission channel determination module 14 in the transmission channel 3D reconstruction system of multiple parallel laser lines is also used for: Based on the spatial phase relationship, the first feature pyramid is integrated to determine the feature distribution of the three-dimensional point cloud; the feature distribution of the three-dimensional point cloud is identified, three-dimensional reconstruction is performed in the three-dimensional reconstruction space, and a first local transmission channel is determined; the laser image set is traversed to construct the Nth local transmission channel; based on the spatial phase relationship, the first local transmission channel to the Nth local transmission channel are spliced ​​to generate the three-dimensional transmission channel.

[0056] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1 The three-dimensional reconstruction method and specific examples of the transmission channel of multiple parallel line lasers in the first embodiment are also applicable to the three-dimensional reconstruction system of the transmission channel of multiple parallel line lasers in this embodiment. Through the detailed description of the three-dimensional reconstruction method of the transmission channel of multiple parallel line lasers, those skilled in the art can clearly know the three-dimensional reconstruction system of the transmission channel of multiple parallel line lasers in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0057] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0058] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A three-dimensional reconstruction method of a power transmission channel using multiple parallel laser lines, characterized in that: include: Assembling a multi-parallel-line laser transmitter and a monocular camera and performing position calibration, wherein the multi-parallel-line laser transmitter and the monocular camera are assembled on a transmission tower based on a preset relative spatial position; Based on the position calibration, combined with the triangular feature relationship, relative motion verification is performed with the laser projection domain as the center, the device posture relationship is determined, and a dynamic acquisition module is constructed; Interactive 3D reconstruction task, triggering the multi-parallel line laser transmitter to perform multi-parallel laser projection on the target power transmission channel, combining with the dynamic acquisition module to perform imaging acquisition based on posture relationship, and determining the laser image set; Combined with the three-dimensional reconstruction space, the laser image set is subjected to step-by-step multi-level image feature screening and processing to determine a feature pyramid, and feature three-dimensional reconstruction is performed to determine a three-dimensional power transmission channel, wherein the three-dimensional reconstruction space is connected to a feature extraction unit based on the pyramid principle; The three-dimensional power transmission channel is displayed visually on the terminal, and reverse guidance is provided on the operation and maintenance management of the target power transmission channel.

2. The method for three-dimensional reconstruction of a transmission channel of multiple parallel laser lines according to claim 1, characterized in that: The determining of the device posture relationship comprises: The triangular feature relationship is determined based on the laser projection area and the position calibration, wherein the multi-parallel line laser emitter and the monocular camera have a motion based on the spatial perspective; Determining an image layout structure, wherein the image layout structure is determined by preset distribution characteristics of the projection part and the background part; Taking the image layout structure as a constraint and combining the triangular feature relationship, relative motion-based verification is performed to mine the device posture relationship, wherein the device posture relationship is the relationship between the monocular camera's follow-up posture under the multi-parallel line laser transmitter's viewing angle posture movement.

3. The method for three-dimensional reconstruction of a transmission channel of multiple parallel laser lines according to claim 2, characterized in that: The construction of the dynamic acquisition module includes: Determine a collection free parallax, wherein the collection free parallax is an allowed overlapping field of view between an upper collection node and a lower collection node; The device posture relationship is used as a single-point acquisition constraint, and the acquisition free parallax is used as a continuous scanning constraint to perform continuous scanning projection and image acquisition management on the target power transmission channel.

4. The method for three-dimensional reconstruction of a transmission channel of multiple parallel laser lines according to claim 1, characterized in that: Also includes: Determine a preset laser projection state and a preset image state; Based on the preset laser projection state, laser parameter configuration is performed, wherein the same projection state based on the difference in projection viewing angle interval is used as a standard; Based on the preset image state, camera parameters are configured, wherein the camera parameters include internal parameters and external parameters, with the same image state under different acquisition viewing angle intervals as the standard.

5. The method for three-dimensional reconstruction of a transmission channel of multiple parallel laser lines according to claim 2, characterized in that: Perform step-by-step multi-level image feature screening and processing to determine the feature pyramid, including: Identify the three-dimensional reconstruction task and set a reconstruction element sequence, wherein each reconstruction element sequence includes element feature-feature dimension-feature intensity; Receive and identify a first laser image, perform boundary segmentation and primary feature extraction on the first laser image based on the image layout structure, and determine a first feature layer, wherein the segmentation boundary is an adjacent boundary between a projection part and a background part; Based on the reconstruction element sequence, a multi-level image feature screening process is performed on the first feature layer to determine a first feature pyramid.

6. The method for three-dimensional reconstruction of a transmission channel of multiple parallel laser lines according to claim 5, characterized in that: The determining of the first feature pyramid includes: Build a pyramid structure with multi-level feature enhancement; Traversing the first feature layer, performing feature transformation and enhancement processing based on the reconstructed feature sequence until the Nth feature layer is determined; Inter-layer mapping is performed on the first feature layer to the Nth feature layer to generate the first feature pyramid.

7. The method for three-dimensional reconstruction of a transmission channel of multiple parallel laser lines according to claim 6, characterized in that: The performing of feature three-dimensional reconstruction to determine the three-dimensional power transmission channel comprises: Based on the spatial phase relationship, integrating the first feature pyramid, determining the feature distribution of the three-dimensional point cloud; Identifying the characteristic distribution of the three-dimensional point cloud, performing three-dimensional reconstruction in the three-dimensional reconstruction space, and determining a first local power transmission channel; Traversing the laser image set to construct an Nth local power transmission channel; Based on the spatial phase relationship, the first local power transmission channel to the Nth local power transmission channel are spliced ​​to generate the three-dimensional power transmission channel.

8. A three-dimensional reconstruction system for transmission channels of multiple parallel laser lines, characterized in that: The method for implementing the three-dimensional reconstruction of the power transmission channel of multiple parallel laser lines as described in any one of claims 1 to 7, wherein the three-dimensional reconstruction system of the power transmission channel of multiple parallel laser lines comprises: A position calibration module, wherein the position calibration module is used to assemble a multi-parallel line laser transmitter and a monocular camera and perform position calibration, wherein the multi-parallel line laser transmitter and the monocular camera are assembled on the transmission tower based on a preset relative spatial position; A relative motion verification module, which is used to perform relative motion verification based on the position calibration and in combination with the triangular feature relationship, with the laser projection domain as the center, determine the device posture relationship, and construct a dynamic acquisition module; An imaging acquisition module, which is used for interactive three-dimensional reconstruction tasks, triggers the multi-parallel line laser transmitter to perform multi-parallel laser projection on the target power transmission channel, combines with the dynamic acquisition module to perform imaging acquisition based on posture relationship, and determines a laser image set; A power transmission channel determination module, the power transmission channel determination module is used to combine the three-dimensional reconstruction space, perform step-by-step multi-level image feature screening and processing on the laser image set to determine a feature pyramid, perform feature three-dimensional reconstruction to determine a three-dimensional power transmission channel, wherein the three-dimensional reconstruction space is connected to a feature extraction unit based on the pyramid principle; An operation and maintenance management module, wherein the operation and maintenance management module is used to perform terminal visualization display of the three-dimensional power transmission channel and reversely guide the operation and maintenance management of the target power transmission channel.

Images