An underground cable automatic detection and trajectory mapping method

Abstract

The application discloses an underground cable automatic detection and track drawing method, relates to the cable detection field, and comprises the following steps: obtaining cable relative position information and detection equipment absolute space coordinates, fusing to obtain absolute space coordinate data of a directly buried cable under a national geodetic coordinate system; acquiring street view images and laser point cloud data, and generating a real scene three-dimensional point cloud model; performing semantic segmentation and identification on a ground object marker, registering the absolute space coordinate data and the real scene three-dimensional point cloud model, establishing a mapping relationship between the underground cable and the ground object marker, and generating a three-dimensional space model; and drawing a cable three-dimensional space track curve and visually presenting the burial depth, trend and relative position relationship with the ground object marker. Through multi-source data fusion, real scene three-dimensional modeling, semantic segmentation registration and mobile terminal visual interaction, the application realizes cm-level accurate positioning of the directly buried cable, underground-ground integrated three-dimensional visual modeling, and on-site fault rapid tracing and accurate excavation.

Description

Technical Field

[0001] This application relates to the field of cable inspection technology, and in particular to an automatic inspection and trajectory mapping method for underground cables. Background Technology

[0002] With the accelerating pace of urbanization, urban underground pipeline networks are becoming increasingly complex. As a core component of urban power distribution networks, the safe and stable operation of buried cables directly impacts power supply reliability. However, traditional operation and maintenance models have long faced the following problems: significant deviations in cable geographic information, with missing or outdated data from early installations, leading to discrepancies between actual routes and records; limited detection methods, with existing electromagnetic induction equipment only able to acquire relative cable location information, failing to visually represent the spatial relationship between underground cables and the surface environment; difficulties in fault tracing, with manual troubleshooting being time-consuming and labor-intensive; and high risks associated with indiscriminate excavation, resulting in frequent accidents involving cable breakage during construction. The root cause of these problems lies in the lack of a technological means to effectively integrate the precise spatial location of underground cables with the actual surface environment.

[0003] In existing technologies, the detection of directly buried cables mainly relies on pipeline detectors based on the principle of electromagnetic induction. However, this method can only obtain local positional information of the cable relative to the detection equipment, and cannot establish absolute spatial coordinates. Furthermore, the detection results lack spatial correlation with the above-ground environment. Although GIS mapping technology is widely used in pipeline network management, the urban canyon effect leads to significant satellite positioning deviations, and the results are mostly two-dimensional plane displays, failing to reflect the three-dimensional spatial orientation and burial depth changes of the cable. In recent years, technologies such as 3D laser scanning, RTK / BeiDou high-precision positioning, and deep learning semantic segmentation have become increasingly mature, but they have not yet formed a systematic integrated application in the field of directly buried cable operation and maintenance. There is a lack of technical solutions that organically integrate these technologies to achieve full-scene spatial perception and visualized operation and maintenance management of underground cables. Summary of the Invention

[0004] The purpose of this application is to provide an automatic detection and trajectory mapping method for underground cables, thereby solving the aforementioned problems in the existing technology.

[0005] To achieve the above objectives, this application provides an automatic detection and trajectory mapping method for underground cables, comprising the following steps: S1: Multi-source detection equipment is used to conduct underground detection of directly buried cables in the city to obtain the relative position information of the cables; at the same time, the absolute spatial coordinates of the detection equipment are obtained through RTK real-time dynamic differential positioning technology; the relative position information and the absolute spatial coordinates are spatiotemporally aligned and fused to obtain the absolute spatial coordinate data of the directly buried cables in the national geodetic coordinate system. S2: Collect street view images of the ground environment in the area where the direct-buried cable is located, obtain street view image data and laser point cloud data, and generate a real-scene three-dimensional point cloud model based on the laser point cloud data and street view image data. S3: Extract the features of ground features from the real-world 3D point cloud model, and automatically identify and classify the ground features based on semantic segmentation technology; perform spatial registration between the absolute spatial coordinate data and the real-world 3D point cloud model to generate a 3D spatial model; S4: Based on the three-dimensional spatial model, draw the underground spatial trajectory curve of the direct-buried cable, and present the burial depth, direction and relative positional relationship with ground features in the real-scene three-dimensional point cloud model in a visual way. S5: Load the 3D spatial model into the mobile terminal device. On-site personnel can use the mobile terminal device to view the spatial location information of the cable on the ground and its relative relationship with surrounding landmarks, so as to achieve accurate positioning of the cable.

[0006] Preferably, the multi-source detection equipment includes electromagnetic induction detection equipment, ground penetrating radar, and acoustic positioning equipment, used to obtain information on the burial depth, direction, and horizontal offset of the cable relative to the detection equipment.

[0007] Preferably, the spatiotemporal alignment and fusion calculation in S1 specifically includes: Based on a unified time reference, the relative position information is matched with the absolute spatial coordinates at the corresponding time for time synchronization. The Kalman filter algorithm is used to fuse and optimize the relative position information and absolute spatial coordinates to obtain the absolute spatial coordinate data of the buried cable in the national geodetic coordinate system.

[0008] Preferably, S2 specifically includes: Using a vehicle-mounted mobile measurement system or a portable 3D laser scanning device, mobile street view data is collected along the path of the buried cable to obtain panoramic images and laser point cloud data; The laser point cloud data is denoised, filtered, downsampled, and registered and stitched to generate a three-dimensional point cloud model. The panoramic image is enhanced and geometrically corrected, and the texture information is extracted and mapped to a 3D point cloud model to generate a realistic 3D point cloud model.

[0009] Preferably, S3 specifically includes: Based on a deep learning semantic segmentation network, this method automatically identifies, classifies, and segments at least one long-term fixed landmark in a real-world 3D point cloud model, including trees, lampposts, manhole covers, traffic signs, and building facades, and extracts the geometric center coordinates and feature descriptors of each landmark. Using the geometric center coordinates of ground features as control points, a point cloud registration algorithm based on feature descriptors is used to perform high-precision spatial registration between absolute spatial coordinate data and the real-world 3D point cloud model. Establish a mapping relationship between the spatial location of underground cables and above-ground landmarks to generate a three-dimensional spatial model.

[0010] Preferably, S4 specifically includes: The B-spline curve fitting algorithm is used to smooth the absolute spatial coordinate data and generate a continuous three-dimensional spatial trajectory curve for the direct-buried cable. In the real-world 3D point cloud model, the buried cable is rendered and displayed in the form of a 3D tubular geometry, and is color-coded according to the burial depth range and line-type coded according to the voltage level. Generate a vertical projection line of the buried cable on the ground, and mark the horizontal distance, azimuth angle and cable burial depth between the projection line and surrounding landmarks, so as to realize a multi-dimensional visualization of the cable's spatial location.

[0011] Preferably, the mobile terminal device in S5 is a tablet computer, a smartphone, or augmented reality (AR) glasses.

[0012] Preferably, it also includes a model dynamic update step, specifically including: When the direct-buried cable is relocated, a new joint is added, or the wiring is changed due to fault repair, or when the surface environment changes due to road construction or greening renovation, S1 to S4 shall be re-executed. The updated detection data is collected and spatially registered and modeled again to generate an updated three-dimensional spatial model; It archives and manages historical versions, supporting version rollback and change comparison analysis.

[0013] Therefore, the above-mentioned method for automatic detection and trajectory mapping of underground cables has the following beneficial effects: (1) This application combines multi-source detection equipment with RTK positioning technology to obtain the absolute spatial coordinate data of the buried cable in the national geodetic coordinate system, which solves the problem that traditional detection methods can only obtain relative position information and cannot establish absolute spatial coordinates, and realizes the centimeter-level accurate positioning of the cable position.

[0014] (2) This application establishes a mapping relationship between underground cables and ground features by spatially registering underground cable detection data with the ground real scene three-dimensional point cloud model, thereby solving the problem of lack of spatial association between underground cables and the ground environment and realizing the integrated underground-surface full-scene spatial perception.

[0015] (3) This application uses semantic segmentation technology to automatically identify ground features and combines it with point cloud registration algorithm to achieve high-precision fusion of underground and above-ground data, reducing manual intervention and improving the automation and accuracy of spatial registration.

[0016] (4) This application generates a three-dimensional spatial model and visualizes the burial depth, direction and relative position of the cable with surrounding landmarks on a mobile terminal, enabling on-site personnel to locate the cable position intuitively and quickly, effectively solving problems such as difficulty in fault tracing and blind excavation, and significantly improving the efficiency of cable operation and maintenance and the reliability of power supply.

[0017] The technical solution of this application will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating an automatic detection and trajectory mapping method for underground cables according to this application. Figure 2 This is a flowchart of the multi-source detection and spatiotemporal alignment fusion calculation in the embodiments of this application; Figure 3 This is a flowchart of semantic segmentation and spatial registration in the embodiments of this application. Detailed Implementation

[0019] The following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0020] Unless otherwise defined, the technical or scientific terms used in this application shall have the ordinary meaning as understood by a person of ordinary skill in the art to which this application pertains.

[0021] The terms "comprising" or "including," as used in this application, mean that the element preceding the term encompasses the element listed after the term, and do not exclude the possibility of encompassing other elements as well. The terms "inner," "outer," "upper," and "lower," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application. When the absolute position of the described object changes, the relative positional relationship may also change accordingly. In this application, unless otherwise expressly specified and limited, the term "attached," etc., should be interpreted broadly. For example, it can refer to a fixed connection, a detachable connection, or an integral part; it can refer to a direct connection or an indirect connection through an intermediate medium; it can refer to the internal communication of two elements or the interaction relationship between two elements. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0022] Example 1: An automatic detection and trajectory mapping method for underground cables, such as Figure 1 As shown, it includes the following steps: S1: Multi-source detection equipment is used to conduct underground detection of directly buried cables in the city to obtain the relative position information of the cables; at the same time, the absolute spatial coordinates of the detection equipment are obtained through RTK real-time dynamic differential positioning technology; the relative position information and the absolute spatial coordinates are spatiotemporally aligned and fused to obtain the absolute spatial coordinate data of the directly buried cables in the national geodetic coordinate system. Multi-source detection equipment includes electromagnetic induction detection equipment, ground-penetrating radar, and acoustic positioning equipment, used to acquire information on the burial depth, direction, and horizontal offset of the cable relative to the detection equipment. Specifically, the electromagnetic induction detection equipment injects a specific frequency alternating current signal into the directly buried cable, utilizes the alternating electromagnetic field generated by the cable itself, and detects the electromagnetic field intensity distribution and phase characteristics through a ground receiving coil array. Based on the electromagnetic field inversion principle, it calculates the burial depth and horizontal offset of the cable. The ground-penetrating radar emits high-frequency electromagnetic waves and receives reflected echoes from different underground medium interfaces. Based on the echo delay and amplitude characteristics, it obtains the dielectric constant distribution of the medium surrounding the cable, thereby identifying the cable location and determining its burial depth. The acoustic positioning equipment applies a pulse excitation signal to the cable, utilizes the characteristic acoustic response generated by cable joints or fault points, and combines time-domain and frequency-domain analysis algorithms to identify the location of acoustic anomalies, assisting in confirming the location of key nodes of the cable. The three types of detection equipment work together, and through the complementarity and cross-verification of multi-source data, they effectively overcome the problem of reduced detection accuracy caused by electromagnetic interference, differences in geological conditions and cable materials in complex urban underground environments when using a single detection method. This enables comprehensive and accurate acquisition of information on the burial depth, direction and horizontal offset of directly buried cables.

[0023] like Figure 2 As shown, spatiotemporal alignment and fusion computation specifically includes: Based on a unified time reference, the relative position information is matched with the absolute spatial coordinates at the corresponding time for time synchronization. The Kalman filter algorithm is used to fuse and optimize the relative position information and absolute spatial coordinates to obtain the absolute spatial coordinate data of the buried cable in the national geodetic coordinate system.

[0024] Specifically, firstly, a unified time reference is configured for the electromagnetic induction detection equipment, ground-penetrating radar, acoustic positioning equipment, and RTK / BeiDou high-precision positioning module using GPS time stamps or network time protocols. Based on timestamps, the relative position information of the cable acquired by each detection device is matched one-to-one with the absolute spatial coordinates output by the RTK / BeiDou module at the corresponding time to ensure the consistency of multi-source data in the spatiotemporal dimension. Subsequently, a Kalman filter algorithm is used to recursively fuse the relative position information of the cable with the absolute spatial coordinates output by RTK / BeiDou. The prior estimate of the current time is calculated through the prediction step, and the a posteriori correction is performed by combining the measured coordinates through the update step. This compensates in real time for satellite positioning drift errors caused by urban canyon effects and system deviations of multi-source detection equipment. Finally, the high-precision absolute spatial coordinate data of the directly buried cable under the national geodetic coordinate system after fusion and optimization is output, realizing the accurate conversion from relative position information to absolute spatial coordinates.

[0025] S2: Collect street view images of the ground environment in the area where the direct-buried cable is located, obtain street view image data and laser point cloud data, and generate a real-scene three-dimensional point cloud model based on the laser point cloud data and street view image data. Mobile street view data is collected along the route of buried cables using either a vehicle-mounted mobile measurement system or a portable 3D laser scanning device, acquiring panoramic images and laser point cloud data. The acquisition method is selected based on the road conditions in the area where the buried cables are laid: for main urban roads or wide sections, a vehicle-mounted mobile measurement system integrating LiDAR, panoramic camera, and RTK / BeiDou positioning module is used to perform mobile scanning along the cable laying route at a constant speed; for sidewalks, green belts, or narrow streets, a backpack or hand-pushed portable 3D laser scanning device is used, with workers gradually advancing along the cable route to collect data; both acquisition methods simultaneously acquire high-density laser point cloud data and 360° panoramic image data.

[0026] The laser point cloud data is denoised, filtered, downsampled, and registered and stitched to generate a 3D point cloud model. The original laser point cloud data is denoised to remove outliers and noise points, and statistical filtering or radius filtering algorithms are used to remove dynamic interference points caused by pedestrian and vehicle movement. The denoised point cloud is downsampled to reduce the data volume and retain geometric features. ICP registration or NDT registration algorithms are used to align and stitch the coordinates of the multi-station scan point clouds to generate a continuous 3D point cloud model covering the entire acquisition area.

[0027] The panoramic imagery undergoes image enhancement and geometric correction, extracting texture information and mapping it to a 3D point cloud model to generate a realistic 3D point cloud model. Image enhancement processing, including white balance correction, exposure compensation, and distortion correction, is applied to the panoramic imagery. Texture information is extracted based on collinearity equations or projection transformation algorithms, and texture mapping is used to map image colors and details onto the surface of the 3D point cloud model, generating a realistic 3D point cloud model with realistic textures that intuitively reflects the details of the ground environment.

[0028] S3: Extract the features of ground features from the real-world 3D point cloud model, and automatically identify and classify the ground features based on semantic segmentation technology; perform spatial registration between the absolute spatial coordinate data and the real-world 3D point cloud model to generate a 3D spatial model; like Figure 3 As shown, based on a deep learning semantic segmentation network, at least one long-term fixed landmark in a real-world 3D point cloud model, including trees, lampposts, manhole covers, traffic signs, and building facades, is automatically identified, classified, and segmented. The geometric center coordinates and feature descriptors of each landmark are extracted. The real-world 3D point cloud model is voxelized or meshed to construct a regularized point cloud data structure. The processed point cloud data is then input into a deep learning semantic segmentation network based on PointNet++ or KPConv architecture. The encoder extracts multi-scale local geometric features and global contextual features, and the decoder performs point-by-point semantic prediction to achieve automatic identification and point-by-point classification of long-term fixed landmarks such as trees, lampposts, manhole covers, traffic signs, and building facades. Instance segmentation is used to distinguish individual landmarks of the same type, obtaining independent point cloud clusters for each landmark instance. Based on the independent point cloud clusters, the geometric center coordinates of each landmark are calculated, and local geometric feature descriptors of each landmark are extracted to provide control points and feature matching basis for subsequent spatial registration.

[0029] Using the geometric center coordinates of ground features as control points, a point cloud registration algorithm based on feature descriptors is used to perform high-precision spatial registration between absolute spatial coordinate data and the real-world 3D point cloud model. Specifically, using the geometric center coordinates of each landmark as control points, an initial correspondence is established between absolute spatial coordinate data and the real-world 3D point cloud model. Local geometric features of each landmark are extracted and feature matching is performed to select the correct matching point pairs. Based on the correctly matched control point pairs, spatial transformation parameters between the two sets of data are calculated to complete the initial alignment. Then, through iterative optimization, the registration error between the two sets of data is further reduced until the accuracy requirements are met, resulting in a high-precision spatial registration result between the absolute spatial coordinate data and the real-world 3D point cloud model, thus achieving accurate spatial mapping between the spatial location of underground cables and above-ground landmarks.

[0030] Establish a mapping relationship between the spatial location of underground cables and above-ground landmarks to generate a three-dimensional spatial model.

[0031] Specifically, based on the high-precision spatial registration results, the absolute spatial coordinate sequence of the buried cable is mapped to the same coordinate frame of the real-scene 3D point cloud model; the horizontal distance, azimuth angle and elevation difference between each positioning point of the cable and surrounding ground features are calculated, and a relative position mapping table between the spatial position of the cable and ground features is established; the 3D spatial trajectory curve of the cable is embedded into the real-scene 3D point cloud model in the form of a 3D tubular geometry, and color coding is performed according to the cable burial depth range and line type coding is performed according to the voltage level, generating a 3D spatial model that integrates the underground cable with the above-ground environment, realizing an intuitive and visual expression of the relationship between the spatial position of the cable and ground features.

[0032] S4: Based on the three-dimensional spatial model, draw the underground spatial trajectory curve of the direct-buried cable, and present the burial depth, direction and relative positional relationship with ground features in the real-scene three-dimensional point cloud model in a visual way. The B-spline curve fitting algorithm is used to smooth the absolute spatial coordinate data and generate a continuous three-dimensional spatial trajectory curve for the direct-buried cable. Specifically, the discrete positioning points in the absolute spatial coordinate data are parameterized by chord length, and the parameter values ​​corresponding to each point are calculated to construct a uniform node vector. A cubic B-spline basis function is set, and the basis function values ​​on each node interval are obtained by recursive calculation. With the discrete positioning points as constraints, the least squares method is used to solve for the coordinates of the control points that minimize the sum of the squared distances from the fitted curve to each point. By iteratively adjusting the node distribution density, the fitting error is gradually converged to a preset threshold range, and finally a continuous three-dimensional spatial trajectory curve that passes through all discrete positioning points and has a smooth curvature change is generated, which truly reflects the actual spatial direction of the buried cable.

[0033] In the real-world 3D point cloud model, the buried cable is rendered as a 3D tubular geometry, color-coded according to the burial depth range and line-type coded according to the voltage level. Using the continuous 3D spatial trajectory curve of the buried cable as the central axis, a circular ring with an equal cross-section is generated according to the actual outer diameter parameters of the cable and swept along the trajectory curve to construct a 3D tubular geometry model with actual pipe diameter dimensions. The cable burial depth is divided into several intervals, with warm colors used for shallow burial areas and cool colors for deep burial areas, making different burial depth sections visually distinguishable. Simultaneously, different line-type codes are set according to the cable voltage level: 10kV cables use solid lines, 35kV cables use dashed lines, 110kV cables use dotted lines, and 220kV cables use double-dotted lines, achieving multi-dimensional visual differentiation of cable type and burial depth information. The aforementioned 3D tubular geometry is embedded into the real-world 3D point cloud model and rendered in fusion with above-ground landmarks within the same coordinate frame, generating a complete 3D spatial model expressing the spatial location of the underground cable and its relationship with the above-ground environment.

[0034] The system generates a vertical projection line of the buried cable on the ground and marks the horizontal distance, azimuth, and cable burial depth between the projection line and surrounding landmarks, achieving a multi-dimensional visualization of the cable's spatial location. The system projects the three-dimensional spatial trajectory curve of the buried cable vertically onto the ground plane, generating a vertical projection line that coincides with the ground. Using equidistant sampling points on the projection line as a reference, it automatically retrieves landmarks within a preset range and calculates the horizontal distance between the sampling points and the geometric centers of each landmark. Using true north as a reference, it calculates the azimuth angle from the sampling points to each landmark. It extracts the corresponding cable burial depth value at each sampling point. The system attaches and displays the above horizontal distance, azimuth, and burial depth values ​​as annotations next to the projection line and landmarks, generating a ground projection map of the cable containing spatial measurement information. It supports switching between multiple view modes, including top-view plan, side-view section, and 3D perspective view from any angle, achieving a multi-dimensional and multi-scale visualization of the cable's spatial location and its relationship with the surrounding environment, providing an intuitive spatial reference for on-site inspections and precise excavation.

[0035] S5: Load the 3D spatial model into the mobile terminal device. On-site personnel can use the mobile terminal device to view the spatial location information of the cable on the ground and its relative relationship with surrounding landmarks, so as to achieve accurate positioning of the cable.

[0036] Mobile terminal devices include tablets, smartphones, or augmented reality (AR) glasses. On-site personnel use tablets or smartphones to load a 3D spatial model and view the ground projection position of the cable and its relative distance to surrounding landmarks such as streetlights and manhole covers from a first-person perspective within a realistic 3D interface. When using AR glasses, the virtual cable trajectory line can be directly overlaid onto the corresponding position in the real ground view, achieving more intuitive on-site positioning guidance.

[0037] Specifically, a lightweight compression algorithm can be used to simplify and convert the data format of the 3D spatial model, generating a low-storage model file suitable for tablets or smartphones. On-site personnel can load the lightweight model through mobile terminals and use touch operations to zoom, rotate, and browse in a realistic 3D interface, intuitively viewing the ground projection position, burial depth changes, and relative spatial relationships with surrounding landmarks such as streetlights, manhole covers, and trees of the directly buried cable. It supports interactive distance measurement between two points on the touch screen, displaying the horizontal distance and spatial straight-line distance between the measurement points in real time, assisting in determining the excavation boundary and work area. When a cable fault occurs and emergency repair is required, on-site personnel can use the landmark reference information displayed on the mobile terminal as a reference to measure and locate the cable in the field, quickly determining the accurate location of the cable and achieving precise excavation and efficient handling of the fault point.

[0038] It also includes a model dynamic update step, specifically including: When the direct-buried cable is relocated, a new joint is added, or the wiring is changed due to fault repair, or when the surface environment changes due to road construction or greening renovation, S1 to S4 shall be re-executed. The updated detection data is collected and spatially registered and modeled again to generate an updated three-dimensional spatial model; It archives and manages historical versions, supporting version rollback and change comparison analysis.

[0039] Example 2: To verify the effectiveness of the method proposed in this application, comparative tests were conducted using the traditional GIS map method, the single electromagnetic induction detection method, and the method proposed in this application, respectively, under the same experimental scenario. The experimental results comparing cable positioning accuracy and fault tracing efficiency are shown in Tables 1 and 2 below: Table 1 Comparison of Cable Positioning Accuracy

[0040] Table 2 Comparison of Fault Tracking Efficiency

[0041] It is evident that the horizontal positioning accuracy of traditional GIS mapping methods is only ±5m to ±15m, with a deviation of up to 8.6m from the actual urban location. While the relative positioning accuracy of a single electromagnetic induction detection method is ±0.5m, the transfer deviation still reaches 1.2m. The method proposed in this application, through multi-source fusion and absolute coordinate registration, improves both horizontal and vertical positioning accuracy to ±5cm, reducing the actual location deviation to 0.05m, achieving a leap from "meter-level" to "centimeter-level" accuracy. Traditional manual surveying methods have an average positioning time of 4.2 hours, require 3.5 trial excavations, and involve approximately 18 cubic meters of earthwork. 3 The single-detection-device positioning method takes 1.5 hours, involves two excavations, and involves approximately 8 cubic meters of earthwork. 3 The method described in this application reduces the positioning time to 12 minutes, allows for precise excavation in one operation, and involves approximately 2.5 cubic meters of earthwork. 3 The efficiency is improved by about 95% compared to the traditional method and by about 87% compared to the single-equipment method.

[0042] Therefore, this application adopts the above-mentioned method for automatic detection and trajectory mapping of underground cables. Through the spatiotemporal alignment and fusion of multi-source detection equipment collaborative detection and RTK high-precision positioning, semantic segmentation and spatial registration of real-scene 3D point cloud models, and 3D visualization and interactive measurement of mobile terminals, it realizes centimeter-level accurate positioning of the spatial location of directly buried cables, integrated underground-surface full-scene 3D visualization modeling, and rapid on-site fault tracing and precise excavation. It effectively solves the industry problems of low positioning accuracy, slow fault tracing, poor visualization and high risk of blind excavation in the existing technology.

[0043] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application and not to limit them. Although this application has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of this application, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of this application.

Claims

1. A method for automatic detection and trajectory mapping of underground cables, characterized in that, Includes the following steps: S1: Multi-source detection equipment is used to conduct underground detection of directly buried cables in the city to obtain the relative position information of the cables; at the same time, the absolute spatial coordinates of the detection equipment are obtained through RTK real-time dynamic differential positioning technology; the relative position information and the absolute spatial coordinates are spatiotemporally aligned and fused to obtain the absolute spatial coordinate data of the directly buried cables in the national geodetic coordinate system. S2: Collect street view images of the ground environment in the area where the direct-buried cable is located, obtain street view image data and laser point cloud data, and generate a real-scene three-dimensional point cloud model based on the laser point cloud data and street view image data. S3: Extract the features of ground features from the real-world 3D point cloud model, and automatically identify and classify the ground features based on semantic segmentation technology; perform spatial registration between the absolute spatial coordinate data and the real-world 3D point cloud model to generate a 3D spatial model; S4: Based on the three-dimensional spatial model, draw the underground spatial trajectory curve of the direct-buried cable, and present the burial depth, direction and relative positional relationship with ground features in the real-scene three-dimensional point cloud model in a visual way. S5: Load the 3D spatial model into the mobile terminal device. On-site personnel can use the mobile terminal device to view the spatial location information of the cable on the ground and its relative relationship with surrounding landmarks, so as to achieve accurate positioning of the cable.

2. The method for automatic detection and trajectory mapping of underground cables according to claim 1, characterized in that, Multi-source detection equipment includes electromagnetic induction detection equipment, ground penetrating radar, and acoustic positioning equipment, used to obtain information on the burial depth, direction, and horizontal offset of the cable relative to the detection equipment.

3. The method for automatic detection and trajectory mapping of underground cables according to claim 2, characterized in that, The spatiotemporal alignment and fusion calculations in S1 specifically include: Based on a unified time reference, the relative position information is matched with the absolute spatial coordinates at the corresponding time for time synchronization. The Kalman filter algorithm is used to fuse and optimize the relative position information and absolute spatial coordinates to obtain the absolute spatial coordinate data of the buried cable in the national geodetic coordinate system.

4. The method for automatic detection and trajectory mapping of underground cables according to claim 3, characterized in that, S2 specifically includes: Using a vehicle-mounted mobile measurement system or a portable 3D laser scanning device, mobile street view data is collected along the path of the buried cable to obtain panoramic images and laser point cloud data; The laser point cloud data is denoised, filtered, downsampled, and registered and stitched to generate a three-dimensional point cloud model. The panoramic image is enhanced and geometrically corrected, and the texture information is extracted and mapped to a 3D point cloud model to generate a realistic 3D point cloud model.

5. The method for automatic detection and trajectory mapping of underground cables according to claim 4, characterized in that, S3 specifically includes: Based on a deep learning semantic segmentation network, this method automatically identifies, classifies, and segments at least one long-term fixed landmark in a real-world 3D point cloud model, including trees, lampposts, manhole covers, traffic signs, and building facades, and extracts the geometric center coordinates and feature descriptors of each landmark. Using the geometric center coordinates of ground features as control points, a point cloud registration algorithm based on feature descriptors is used to perform high-precision spatial registration between absolute spatial coordinate data and the real-world 3D point cloud model. Establish a mapping relationship between the spatial location of underground cables and above-ground landmarks to generate a three-dimensional spatial model.

6. The method for automatic detection and trajectory mapping of underground cables according to claim 5, characterized in that, S4 specifically includes: The B-spline curve fitting algorithm is used to smooth the absolute spatial coordinate data and generate a continuous three-dimensional spatial trajectory curve for the direct-buried cable. In the real-world 3D point cloud model, the buried cable is rendered and displayed in the form of a 3D tubular geometry, and is color-coded according to the burial depth range and line-type coded according to the voltage level. Generate a vertical projection line of the buried cable on the ground, and mark the horizontal distance, azimuth angle and cable burial depth between the projection line and surrounding landmarks, so as to realize a multi-dimensional visualization of the cable's spatial location.

7. The method for automatic detection and trajectory mapping of underground cables according to claim 6, characterized in that, Mobile devices in S5 can be tablets, smartphones, or augmented reality (AR) glasses.

8. The method for automatic detection and trajectory mapping of underground cables according to claim 7, characterized in that, It also includes a model dynamic update step, specifically including: When the direct-buried cable is relocated, a new joint is added, or the wiring is changed due to fault repair, or when the surface environment changes due to road construction or greening renovation, S1 to S4 shall be re-executed. The updated detection data is collected and spatially registered and modeled again to generate an updated three-dimensional spatial model; It archives and manages historical versions, supporting version rollback and change comparison analysis.

Images