Underground cable positioning method and device and electronic equipment

Abstract

The invention discloses an underground cable positioning method, an underground cable positioning device and electronic equipment. The method comprises the following steps: acquiring an inspection image and equipment positioning data corresponding to inspection equipment; determining a cable three-dimensional model according to the equipment positioning data; determining space correlation parameters between the inspection image and the cable three-dimensional model; determining depth data corresponding to the inspection image; according to the space correlation parameters and the depth data, fusing the inspection image with the cable three-dimensional model to obtain a cable fusion model; and positioning the underground cable according to the cable fusion model. According to the invention, the technical problem of inaccurate positioning of the underground cable during positioning of the underground cable in the prior art is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of cable inspection, in particular to a positioning method and device for underground cable and electronic equipment. BACKGROUND

[0002] In the related art, as an important carrier for energy transmission, the accurate positioning of underground cable during inspection is the key to avoiding construction damage, quickly troubleshooting, ensuring operation safety and efficiency, and is directly related to power supply reliability and on-site operation safety. However, in the related art, when positioning the underground cable, the problem of inaccurate positioning of the underground cable exists.

[0003] In view of the above problems, no effective solution has been proposed so far. SUMMARY

[0004] The embodiments of the present application provide a positioning method and device for underground cable and electronic equipment to at least solve the problem of inaccurate positioning of underground cable in the related art.

[0005] According to an aspect of the embodiments of the present application, a positioning method for underground cable is provided, comprising: obtaining an inspection image corresponding to an inspection device and device positioning data; determining a cable three-dimensional model according to the device positioning data, wherein the cable three-dimensional model is a three-dimensional data model of the underground cable corresponding to an inspection area, and the inspection area is an area in which the inspection device inspects the underground cable; determining a spatial correlation parameter between the inspection image and the cable three-dimensional model, wherein the spatial correlation parameter is a parameter used to represent the spatial correlation relationship between the inspection image and the cable three-dimensional model; determining depth data corresponding to the inspection image; fusing the inspection image and the cable three-dimensional model according to the spatial correlation parameter and the depth data to obtain a cable fusion model; and positioning the underground cable according to the cable fusion model.

[0006] Optionally, the fusing the inspection image with the cable three-dimensional model according to the spatial correlation parameter, the depth data, and the inspection image to obtain a cable fusion model comprises: in a case where the spatial correlation parameter comprises a first correlation parameter and a second correlation parameter, the first correlation parameter representing a coordinate conversion relationship between a first spatial coordinate system and a third spatial coordinate system, and the second correlation parameter representing a coordinate conversion relationship between a second spatial coordinate system and the third spatial coordinate system, the first spatial coordinate system being a spatial coordinate system corresponding to the cable three-dimensional model, the second spatial coordinate system being a spatial coordinate system corresponding to the inspection image, and the third spatial coordinate being a spatial coordinate corresponding to the inspection device; fusing the inspection image with the cable three-dimensional model according to the first correlation parameter, the second correlation parameter, and the depth data to obtain the cable fusion model.

[0007] Optionally, the fusing the inspection image with the cable three-dimensional model according to the first correlation parameter, the second correlation parameter, and the depth data to obtain a cable fusion model comprises: fusing the inspection image with the cable three-dimensional model according to the first correlation parameter and the second correlation parameter to obtain an initial fusion model; determining an object distribution parameter according to the depth data and the inspection image, the object distribution parameter being used to represent a spatial distribution feature of an above-ground object, the depth data comprising depth parameters corresponding to a plurality of pixel points, and the inspection image comprising the plurality of pixel points; determining a cable distribution parameter according to the initial fusion model, the cable distribution parameter being used to represent a spatial distribution feature of an underground cable; and adjusting the initial fusion model according to the object distribution parameter and the cable distribution parameter to obtain the cable fusion model.

[0008] Optionally, the adjusting the initial fusion model according to the object distribution parameter and the cable distribution parameter to obtain the cable fusion model comprises: determining a spatial correspondence relationship according to the object distribution parameter and the cable distribution parameter, the spatial correspondence relationship representing a spatial correspondence relationship between the above-ground object and the underground cable; determining a shielding part corresponding to the underground cable and a shielding object corresponding to the shielding part according to the spatial correspondence relationship, the shielding object being an above-ground object whose shielding index of the shielding part is greater than a shielding threshold; determining a visual parameter corresponding to the shielding object according to a part feature corresponding to the shielding part; and adjusting the initial fusion model according to the visual parameter to obtain the cable fusion model, wherein the adjusting comprises shielding adjustment.

[0009] Optionally, the fusing the inspection image with the cable three-dimensional model according to the first correlation parameter and the second correlation parameter to obtain an initial fusion model comprises: determining video data corresponding to the inspection image; determining spatial variation features corresponding to a plurality of target feature points respectively according to the video data, the inspection image comprising the plurality of target feature points; and fusing the inspection image with the cable three-dimensional model according to the spatial variation features corresponding to the plurality of target feature points respectively, the first correlation parameter and the second correlation parameter to obtain the initial fusion model.

[0010] Optionally, before the determining spatial variation features corresponding to a plurality of target feature points respectively according to the video data, the method further comprises: in a case where the video data comprises a plurality of object feature points, determining position variation indexes corresponding to the plurality of object feature points respectively according to the video data, wherein the corresponding position variation indexes represent degrees of variation of spatial positions of corresponding object feature points, and the plurality of object feature points are feature points used to reflect object shape features of the above-ground object; and determining a plurality of target feature points from the plurality of object feature points according to the position variation indexes corresponding to the plurality of object feature points respectively, wherein the plurality of target feature points are object feature points whose position variation indexes are less than a position variation threshold in the plurality of object feature points.

[0011] Optionally, the fusing the inspection image with the cable three-dimensional model according to the spatial correlation parameter, the depth data to obtain a cable fusion model comprises: determining inspection trajectory data corresponding to the inspection device; determining a spatial pose deviation corresponding to the inspection device according to the inspection trajectory data; and fusing the inspection image with the cable three-dimensional model according to the spatial pose deviation, the spatial correlation parameter and the depth data to obtain the cable fusion model.

[0012] According to one aspect of the present invention, a device for locating underground cables is provided, comprising: an acquisition module for acquiring an inspection image and equipment positioning data corresponding to an inspection device; a first determination module for determining a three-dimensional model of the cable based on the equipment positioning data, wherein the three-dimensional model of the cable is a three-dimensional data model of the underground cable corresponding to an inspection area, and the inspection area is the area where the inspection device inspects the underground cable; a second determination module for determining the spatial association between the inspection image and the three-dimensional model of the cable; a third determination module for determining depth data corresponding to the inspection image, wherein the depth data includes depth parameters corresponding to multiple pixels, and the inspection image includes the multiple pixels; a fourth determination module for fusing the inspection image and the three-dimensional model of the cable based on the spatial association and the depth data to obtain a cable fusion model; and a fifth determination module for locating the underground cable based on the cable fusion model.

[0013] According to one aspect of the present invention, an electronic device is provided, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to execute the instructions to implement the underground cable location method described in any of the preceding claims.

[0014] According to one aspect of the present invention, a computer-readable storage medium is provided, comprising: when instructions in the computer-readable storage medium are executed by a processor of an electronic device, enabling the electronic device to perform the underground cable location method described in any of the preceding claims.

[0015] In this embodiment of the invention, inspection images and equipment positioning data corresponding to the inspection equipment are acquired; based on the equipment positioning data, a three-dimensional model of the cable is determined, wherein the three-dimensional model of the cable is a three-dimensional data model of the underground cable corresponding to the inspection area, and the inspection area is the area where the inspection equipment inspects the underground cable; spatial association parameters between the inspection image and the three-dimensional model of the cable are determined, wherein the spatial association parameters are parameters used to represent the spatial association relationship between the inspection image and the three-dimensional model of the cable; depth data corresponding to the inspection image is determined; based on the spatial association parameters and the depth data, the inspection image and the three-dimensional model of the cable are fused to obtain a fused cable model; and the underground cable is located based on the fused cable model. By acquiring inspection images and equipment positioning data from inspection equipment, the inspection area of ​​underground cables can be located and on-site real-world references can be obtained. Based on the equipment positioning data, the corresponding 3D cable model is selected to ensure accurate matching between the model and the inspection scene. By determining the spatial correlation parameters between the inspection images and the 3D cable model, the spatial mapping logic between the two is established. Combined with the depth data corresponding to the inspection images to supplement the scene's 3D information, the inspection images and the 3D cable model are fused to obtain a cable fusion model. This achieves accurate superposition of the real scene and the virtual model. Furthermore, through the cable fusion model, accurate positioning of underground cables can be achieved, thus solving the technical problem of inaccurate positioning of underground cables in related technologies. Attached Figure Description

[0016] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0017] Figure 1 This is a flowchart of a method for locating underground cables according to an embodiment of the present invention;

[0018] Figure 2 This is a schematic diagram of the cable fusion model generation process in an optional embodiment of the present invention;

[0019] Figure 3 This is a structural block diagram of an underground cable positioning device according to an embodiment of the present invention. Detailed Implementation

[0020] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0021] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0022] First, some nouns or terms that appear in the description of the embodiments of this application shall be interpreted as follows:

[0023] PPP-RTK technology: PPP-RTK (Precise Point Positioning Real-Time Kinematic) is a high-precision positioning technology that can achieve centimeter-level real-time positioning of a single device without relying on a local base station.

[0024] 6-DOF MEMS-IMU: A 6-DOF MEMS-IMU is a miniature inertial measurement unit that can measure the motion of an object in three-dimensional space. An object can have six degrees of freedom in three-dimensional space, namely translational motion along the x, y, and z coordinate axes and rotational motion around these three coordinate axes. The 6-DOF MEMS-IMU is used to measure these motion parameters.

[0025] Dead Reckoning: Dead Reckoning is a navigation method that estimates the current position of an object based on known initial position and motion information (such as velocity, direction, and time).

[0026] PnP Algorithm: The PnP (Perspective-n-Point) algorithm is a computer vision technique used to estimate the camera pose (position and orientation) from known 3D points and their corresponding 2D image points.

[0027] ORB: ORB (Oriented Fast and Rotated BRIEF) is a feature point detection and descriptor extraction algorithm used in the field of computer vision.

[0028] FAST: FAST (Features from Accelerated Segment Test) is a fast corner detection algorithm that quickly determines whether a pixel is a corner (easily identifiable key point such as object edge or inflection point) by comparing the grayscale difference between a pixel in an image and its 16 neighboring pixels.

[0029] ORB-SLAM: ORB-SLAM is a real-time localization and mapping technology based on ORB feature points, used for simultaneous localization and mapping.

[0030] B-spline curve: A B-spline curve is a parametric curve fitting technique used to fit a series of discrete data points to a smooth curve.

[0031] RANSAC: RANSAC (Random Sample Consensus) is a robust outlier filtering and model fitting algorithm used to identify useful data and remove noise data.

[0032] SLAM Algorithm: SLAM (Simultaneous Localization and Mapping) is a technique for robots or autonomous intelligent agents to build environmental maps and estimate their own positions in unknown environments using their own sensor data.

[0033] ToF: ToF (Time-of-Flight) is a depth sensing technology that calculates the distance to an object by measuring the time it takes for a light signal to travel from its source to the target object and back to the sensor.

[0034] RGB: RGB (Red-Green-Blue) represents the red, green, and blue color channel information, used to generate color images.

[0035] Example 1

[0036] According to an embodiment of the present invention, an embodiment of a method for locating underground cables is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0037] Figure 1 This is a flowchart of a method for locating underground cables according to an embodiment of the present invention, such as... Figure 1 As shown, the method includes the following steps:

[0038] S102, acquire the inspection images and equipment positioning data corresponding to the inspection equipment.

[0039] In step S102 of this application, inspection images and equipment positioning data corresponding to the inspection equipment are obtained.

[0040] This includes inspection equipment, which is used for inspecting underground cables. It can be equipped with components such as cameras, inertial measurement units, satellite positioning modules, and time-of-flight sensors, and can move in the inspection area to collect environmental data.

[0041] This includes inspection images, which are real-world images (including color images) of the inspection area taken by the camera of the inspection equipment (such as a binocular camera). These images contain the texture and shape information of objects on the ground (manhole covers, curbs, road markings, etc.) and serve as the real-world benchmark for subsequent fusion with the cable 3D model.

[0042] This includes equipment positioning data, which is the spatial position and attitude data of the inspection equipment during the inspection process. It is obtained by fusing multiple source technologies such as satellite positioning, inertial navigation, and visual positioning (e.g., centimeter-level satellite positioning data + high-frequency inertial measurement data + visual trajectory data), and is used to lock the real-time position of the inspection equipment, providing a basis for selecting the corresponding cable 3D model.

[0043] By acquiring inspection images corresponding to the inspection equipment, a real-world benchmark is provided for subsequent fusion with the cable 3D model. At the same time, acquiring equipment positioning data can pinpoint the real-time location of the inspection equipment, thus providing a data foundation for subsequent positioning.

[0044] S104. Based on the equipment positioning data, determine the three-dimensional model of the cable. The three-dimensional model of the cable is the three-dimensional data model of the underground cable corresponding to the inspection area. The inspection area is the area where the inspection equipment inspects the underground cable.

[0045] In step S104 of this application, a three-dimensional model of the cable is determined based on the equipment positioning data.

[0046] This involves a 3D model of the cable, which is a digital 3D data model of the underground cable corresponding to the current inspection area. It includes spatial geometric information such as the cable's direction, burial depth, pipe diameter, and joint location, and serves as a virtual cable reference for subsequent fusion with the inspection images.

[0047] This includes the inspection area, which is the area where the inspection equipment is currently performing underground cable inspection tasks.

[0048] This involves underground cables, which are power transmission cables laid below ground level.

[0049] This includes underground cable inspection, which involves moving inspection equipment within a designated area to detect the operational status of underground cables by collecting images, positioning data, and other methods.

[0050] Based on equipment positioning data, a 3D model of the cable is determined, which can accurately select the underground cable model corresponding to the current inspection area, ensuring that the reference model for subsequent fusion processing is consistent with the actual inspection scenario. At the same time, the 3D cable model contains key information such as cable route and burial depth, which can provide an accurate virtual reference for subsequent fusion with inspection images, avoiding positioning deviations caused by model mismatch, thereby achieving accurate positioning and status detection of underground cables.

[0051] S106, determine the spatial association parameters between the inspection image and the cable 3D model, wherein the spatial association parameters are parameters used to represent the spatial association relationship between the inspection image and the cable 3D model.

[0052] In step S106 of this application, the spatial correlation parameters between the inspection image and the three-dimensional model of the cable are determined.

[0053] This involves spatial correlation parameters, which are used to establish a spatial mapping between the inspection image and the cable 3D model. These parameters include a first correlation parameter (external parameter matrix, which describes the transformation relationship between the cable 3D model coordinate system and the inspection equipment coordinate system) and a second correlation parameter (internal parameter matrix, which describes the transformation relationship between the inspection equipment coordinate system and the inspection image coordinate system), in order to eliminate the spatial pose deviation between the two.

[0054] This involves spatial relationships, which are the spatial correspondences between inspection images and cable 3D models, such as position matching relationships and viewpoint adaptation relationships.

[0055] By determining spatial correlation parameters, including extrinsic matrix (describing the transformation relationship between the coordinate system of the cable 3D model and the coordinate system of the inspection equipment) and intrinsic matrix (describing the transformation relationship between the coordinate system of the inspection equipment and the coordinate system of the inspection image), the spatial correspondence between the inspection image and the cable 3D model, such as position matching and viewpoint adaptation, can be accurately captured. This effectively eliminates the spatial pose deviation between the two, providing a core spatial mapping basis for the subsequent accurate fusion of the inspection image and the cable 3D model with depth data, ensuring that the fused model can truly reflect the corresponding position of the underground cable and the above-ground scene.

[0056] S108, determine the depth data corresponding to the inspection image.

[0057] In step S108 provided in this application, depth data corresponding to the inspection image is determined.

[0058] This involves depth data, which is spatial distance data corresponding to each pixel in the inspection image. For example, the actual distance from the ground object corresponding to the pixel to the inspection equipment includes depth parameters corresponding to multiple pixels, which can supplement the three-dimensional information of the inspection image and clarify the spatial distribution layers of ground objects.

[0059] By determining the depth data corresponding to the inspection images, the three-dimensional information of the images can be supplemented, clearly defining the spatial distribution and relative positions of objects on the ground. This provides crucial spatial distance support for the subsequent accurate fusion of the inspection images and the cable 3D model based on spatial correlation parameters, ensuring precise depth matching between the virtual cable model and the real ground scene during the fusion process and avoiding spatial misalignment. Furthermore, it ensures accurate identification of occlusion areas and objects when fusion of inspection images and the cable 3D model, avoiding misjudgments of occlusion due to missing depth information, and achieving precise adaptation between the virtual cable model and the real scene.

[0060] S110, based on spatial correlation parameters and depth data, fuses the inspection images with the cable 3D model to obtain the cable fusion model.

[0061] In step S110 of this application, the inspection image is fused with the cable 3D model based on spatial correlation parameters and depth data to obtain a cable fusion model.

[0062] This involves fusion, which is a process that uses spatial correlation parameters as the basis for spatial mapping and combines the distance and layer information of ground objects provided by depth data to spatially align the real ground scene of the inspection image with the virtual underground cable information of the cable 3D model, determine occlusion, and visually integrate them. The core is to accurately match the virtual and real spatial positions, while identifying and processing the occlusion of underground cables by ground objects to ensure the visualization of the occluded parts.

[0063] This involves a cable fusion model, which is a three-dimensional data model formed after fusion processing. It not only fully preserves the real-world details of objects on the ground in the inspection images, but also accurately overlays the three-dimensional information of underground cables aligned with the real-world space. Furthermore, it has been optimized for occlusion situations (such as highlighting key parts of obscured cables), achieving an intuitive correspondence between underground cables and the ground scene. It is the core basis for the location and visualization of underground cables.

[0064] Based on spatial correlation parameters and combined with the distance and layer information of ground objects provided by depth data, the system spatially aligns, determines occlusion, and visually integrates the real ground scene of the inspection image with the virtual underground cable information of the cable 3D model. This enables precise matching of virtual and real spatial positions, effectively identifies and handles the occlusion of underground cables by ground objects, ensures the visualization of key parts of the occluded cable, fully preserves the real scene details of ground objects, and accurately overlays the aligned 3D information of the underground cable. This achieves an intuitive correspondence between the underground cable and the ground scene, providing core support for subsequent precise positioning and visualization of underground cables based on the cable fusion model.

[0065] S112, based on the cable fusion model, locate the underground cable.

[0066] In step S112 provided in this application, the underground cable was located based on the cable fusion model.

[0067] This involves locating underground cables, which means determining the specific location and distribution of underground cables in space, so as to accurately identify abnormal underground cables during inspections.

[0068] Based on the cable fusion model that integrates above-ground real-world details with precisely aligned 3D information of underground cables and optimizes the presentation of obstructions, the specific location and distribution of underground cables in space can be clearly identified. This ensures that the actual spatial coordinates of underground cables can be quickly located during inspections, and that abnormal underground cables can be accurately identified, providing a targeted analytical basis for subsequent inspection operations.

[0069] Through the above steps S102-S112, inspection images and equipment positioning data corresponding to the inspection equipment are obtained; based on the equipment positioning data, a three-dimensional model of the cable is determined, wherein the three-dimensional model of the cable is a three-dimensional data model of the underground cable corresponding to the inspection area, and the inspection area is the area where the inspection equipment inspects the underground cable; spatial association parameters between the inspection image and the cable three-dimensional model are determined, wherein the spatial association parameters are parameters used to represent the spatial relationship between the inspection image and the cable three-dimensional model; depth data corresponding to the inspection image is determined; based on the spatial association parameters and the depth data, the inspection image and the cable three-dimensional model are fused to obtain a cable fusion model; and the underground cable is located based on the cable fusion model. By acquiring inspection images and equipment positioning data from inspection equipment, the inspection area of ​​underground cables can be located and on-site real-world references can be obtained. Based on the equipment positioning data, the corresponding 3D cable model is selected to ensure accurate matching between the model and the inspection scene. By determining the spatial correlation parameters between the inspection images and the 3D cable model, the spatial mapping logic between the two is established. Combined with the depth data corresponding to the inspection images to supplement the scene's 3D information, the inspection images and the 3D cable model are fused to obtain a cable fusion model. This achieves accurate superposition of the real scene and the virtual model. Furthermore, through the cable fusion model, accurate positioning of underground cables can be achieved, thus solving the technical problem of inaccurate positioning of underground cables in related technologies.

[0070] As an optional embodiment, based on spatial correlation parameters and depth data, the inspection image is fused with the cable 3D model to obtain a cable fusion model. This includes: when the spatial correlation parameters include a first correlation parameter and a second correlation parameter, the inspection image is fused with the cable 3D model based on the first correlation parameter, where the first correlation parameter represents the coordinate transformation relationship between a first spatial coordinate system and a third spatial coordinate system, and the second correlation parameter represents the coordinate transformation relationship between a second spatial coordinate system and a third spatial coordinate system. The first spatial coordinate system is the spatial coordinate system corresponding to the cable 3D model, the second spatial coordinate system is the spatial coordinate system corresponding to the inspection image, and the third spatial coordinate system is the spatial coordinate system corresponding to the inspection equipment.

[0071] This embodiment describes the specific steps for fusing inspection images with a 3D cable model based on spatial correlation parameters and depth data to obtain a fused cable model.

[0072] This involves a first associated parameter, which describes the position and attitude transformation between the first spatial coordinate system (the three-dimensional model coordinate system of the cable) and the third spatial coordinate system (the coordinate system of the inspection equipment). It can be an extrinsic parameter matrix, used to achieve rigid body transformation (spatial alignment without shape and size changes) between the two coordinate systems through rotation matrix and translation vector.

[0073] This involves a second correlation parameter, which describes the projection transformation between the second spatial coordinate system (inspection image coordinate system) and the third spatial coordinate system (inspection equipment coordinate system). It can be an intrinsic parameter matrix used to integrate hardware characteristics such as camera focal length and principal point offset to achieve the mapping from three-dimensional spatial coordinates to two-dimensional image pixel coordinates.

[0074] This involves a first spatial coordinate system, which is a three-dimensional spatial coordinate system corresponding to the three-dimensional model of the cable.

[0075] This involves a third spatial coordinate system, which is a three-dimensional spatial coordinate system corresponding to the inspection equipment.

[0076] This involves coordinate transformation relationships, which are the transformation relationships between the coordinates of points in different spatial coordinate systems, ensuring that the position description of the same spatial point in different coordinate systems can be accurately matched.

[0077] This involves a second spatial coordinate system, which is a two-dimensional planar coordinate system corresponding to the inspection image.

[0078] This involves a spatial coordinate system, which is a coordinate system for describing the spatial position of an object. By setting the origin, coordinate axes, and unit length, it is possible to quantitatively describe the spatial position of an object, including two-dimensional (planar) coordinate systems and three-dimensional (solid) coordinate systems.

[0079] Using the third spatial coordinate system (inspection equipment coordinate system) as an intermediate reference, the rigid body transformation between the first spatial coordinate system (cable 3D model coordinate system) and the third spatial coordinate system is achieved through the first correlation parameter (external parameter matrix). The projection transformation between the second spatial coordinate system (inspection image coordinate system) and the third spatial coordinate system is completed with the help of the second correlation parameter (internal parameter matrix). Combined with the 3D stereo information supplemented by depth data, the inspection image and the cable 3D model can be accurately aligned under a unified spatial reference, eliminating the spatial pose deviation between the two and ensuring the accurate fusion of virtual and real information. Finally, a cable fusion model that can intuitively reflect the correspondence between underground cables and above-ground scenes is obtained, providing a reliable basis for the accurate positioning of underground cables.

[0080] As an optional embodiment, the inspection image is fused with the cable 3D model based on the first association parameter, the second association parameter, and depth data to obtain a cable fusion model. This includes: fusing the inspection image with the cable 3D model based on the first association parameter and the second association parameter to obtain an initial fusion model; determining object distribution parameters based on the depth parameters corresponding to multiple pixels and the inspection image, wherein the object distribution parameters represent the spatial distribution characteristics of objects on the ground, the depth data includes the depth parameters corresponding to multiple pixels, and the inspection image includes multiple pixels; determining cable distribution parameters based on the initial fusion model, wherein the cable distribution parameters represent the spatial distribution characteristics of underground cables; and adjusting the initial fusion model based on the object distribution parameters and the cable distribution parameters to obtain the cable fusion model.

[0081] In this embodiment, the specific steps for fusing the inspection image with the cable 3D model based on the first correlation parameter, the second correlation parameter, and depth data to obtain the cable fusion model are described.

[0082] This involves an initial fusion model, which is a 3D data model that has been spatially aligned based solely on the first correlation parameter (external parameter matrix) and the second correlation parameter (internal parameter matrix). It retains the basic spatial correspondence between the real scene of the inspection image and the 3D model of the cable, but does not optimize the occlusion adaptation problem, that is, it does not consider details such as occlusion by objects on the ground.

[0083] This involves multiple pixels, which form the basic unit set that constitutes the inspection image. Each pixel corresponds to a tiny area in the inspection scene, and its position and depth parameters can jointly reflect the spatial information of objects on the ground.

[0084] This involves object distribution parameters, which are determined based on the depth parameters of multiple pixels in the inspection image. These parameters are used to quantitatively describe the position, range, layer, and spacing of ground objects (such as manhole covers, vegetation, and buildings) in three-dimensional space, that is, to reflect the spatial distribution characteristics of ground objects.

[0085] This involves a depth parameter, which corresponds one-to-one with the pixels in the inspection image and represents the actual distance from the ground object corresponding to that pixel to the inspection equipment.

[0086] This involves cable distribution parameters, which are extracted from the initial fusion model and used to quantitatively describe the spatial characteristics of underground cables in three-dimensional space, such as their direction, burial depth, joint location, and extension range. In other words, they are used to represent the spatial distribution characteristics of underground cables.

[0087] This involves spatial distribution characteristics, which are the distribution characteristics related to the position, shape, layer, and spacing of objects (ground objects or underground cables) in three-dimensional space.

[0088] This involves adjustment processing, which is a process of optimizing and correcting the spatial correspondence between above-ground objects and underground cables that are not adapted in the initial fusion model based on object distribution parameters and cable distribution parameters (such as occlusion and spatial misalignment). The core is to make the fusion model more in line with the real scene.

[0089] An initial fusion model is obtained by fusing the inspection image with the cable 3D model based on the first and second correlation parameters. This establishes the basic spatial correspondence between the two. Then, the object distribution parameters that characterize the spatial distribution features of objects on the ground are determined by the depth parameters corresponding to multiple pixels in the inspection image. Cable distribution parameters that describe the spatial distribution features of underground cables are extracted from the initial fusion model. Finally, the initial fusion model is optimized and corrected based on these two types of parameters. This can accurately solve problems such as occlusion and spatial misalignment in the initial model, ensuring that the fusion model is highly consistent with the real scene. Ultimately, a cable fusion model that clearly presents the spatial correspondence between underground cables and objects on the ground is obtained, providing reliable support for the accurate positioning of underground cables.

[0090] As an optional embodiment, the initial fusion model is adjusted based on object distribution parameters and cable distribution parameters to obtain a cable fusion model. This includes: determining spatial correspondences based on object distribution parameters and cable distribution parameters, where the spatial correspondences represent the spatial correspondences between above-ground objects and underground cables; determining the obstructing parts corresponding to the underground cables and the obstructing objects corresponding to the obstructing parts based on the spatial correspondences, where the obstructing objects are above-ground objects whose obstruction index for the obstructing parts is greater than an obstruction threshold; determining the visual parameters corresponding to the obstructing objects based on the location features corresponding to the obstructing parts; and performing obstruction adjustment processing on the initial fusion model based on the visual parameters to obtain the cable fusion model, where the adjustment processing includes obstruction adjustment processing.

[0091] This embodiment describes the specific steps for adjusting the initial fusion model based on object distribution parameters and cable distribution parameters to obtain the cable fusion model.

[0092] This involves spatial correspondence, which is determined based on object distribution parameters and cable distribution parameters. Clarifying the relative positional relationship between above-ground objects and underground cables in three-dimensional space (such as a section of cable directly under a manhole cover, or the overlap between vegetation cover area and cable direction) is the basis for judging obstruction.

[0093] This involves ground objects, which are various entities above ground that appear in the inspection images, such as manhole covers, road markings, vegetation, buildings, and utility poles, to determine whether they obstruct the visualization of underground cables.

[0094] This includes obstructed sections, which are parts of underground cables that overlap with above-ground objects in space and are obscured by these objects, making them invisible in the real-world environment. Examples include cable joints covered by manhole covers and cable sections obscured by vegetation.

[0095] This involves obstructing objects, which are above-ground objects that effectively obstruct underground cables. Specifically, they refer to above-ground objects whose obstruction index for a certain cable section is greater than a preset obstruction threshold, and are the direct cause of the cable section being invisible.

[0096] This includes the shading index, which quantifies the degree to which above-ground objects obstruct underground cables. It can be calculated by combining factors such as the spatial range of the above-ground object, the area of ​​overlap with the cable, and the depth distance. The higher the value, the more severe the shading.

[0097] This involves an occlusion threshold, which is a pre-set critical value used to distinguish whether an object on the ground constitutes effective occlusion.

[0098] This involves location features, which are the attribute characteristics of the parts of the underground cable that are blocked, such as location, size, material, and importance. These are the key factors in determining the optimization method for blocking (e.g., joints should be highlighted first).

[0099] This involves visual parameters, which are image display parameters set to visualize occluded parts. These include highlight color values, transparency, edge glow intensity, rendering priority, etc., to ensure that the occluded cable parts are clearly presented in the model.

[0100] This involves occlusion adjustment processing, which is a targeted optimization process for the initial fusion model based on visual parameters. The core is to highlight the key parts of the occluded cable through special visual effects (such as semi-transparent highlighting and edge enhancement), eliminate the influence of ground objects on the cable visualization, and ensure that the cable is presented in its entirety.

[0101] Determining the spatial correspondence between above-ground objects and underground cables based on object distribution parameters and cable distribution parameters provides a foundation for occlusion judgment. This spatial correspondence then identifies overlapping occlusion areas within the underground cable and occlusion objects with occlusion indices exceeding the occlusion threshold. Combining the location and importance of the occluded areas, visual parameters such as highlight color values ​​and transparency are determined. Finally, based on these visual parameters, the initial fusion model undergoes occlusion adjustment processing. This allows for the highlighting of critical cable components through special visual effects such as semi-transparent highlighting and edge enhancement, eliminating the impact of above-ground objects on cable visualization, ensuring complete cable presentation, and making the final cable fusion model more closely resemble real-world inspection scenarios.

[0102] As an optional embodiment, based on a first association parameter and a second association parameter, the inspection image is fused with the cable 3D model to obtain an initial fusion model, including: determining video data corresponding to the inspection image; determining spatial variation features corresponding to multiple target feature points based on the video data, wherein the inspection image includes multiple target feature points; and fusing the inspection image with the cable 3D model based on the spatial variation features corresponding to the multiple target feature points, the first association parameter, and the second association parameter to obtain the initial fusion model.

[0103] In this embodiment, the specific steps for fusing the inspection image with the cable 3D model based on the first correlation parameter and the second correlation parameter are described to obtain the initial fused model.

[0104] This includes video data, which is dynamic image data used to reflect the inspection equipment during the inspection process. It contains a series of multiple consecutive inspection images that can reflect the dynamic changes of ground objects (such as manhole covers and road markings) or the changes in perspective during the movement of the equipment in the inspection scene.

[0105] This involves multiple target feature points, which are sets of feature pixels with significant recognizability and stability. These feature points correspond to key parts of objects on the ground (such as the corner points of manhole covers and the outline points of road markings). Their position, shape, and other features are not easily affected by the environment and are the core reference points for tracking spatial changes and achieving accurate alignment between images and models.

[0106] Determining the video data corresponding to the inspection images (including a sequence of multiple consecutive inspection images that reflect dynamic changes in ground objects or changes in equipment perspective) provides continuous and dynamic data support for extracting spatial change features of target feature points. Then, based on this video data, the spatial change features corresponding to multiple target feature points (stable and recognizable pixels of key parts of ground objects) in the inspection images are captured. By combining the first and second correlation parameters, the inspection images are fused with the cable 3D model, which can offset the spatial deviation caused by equipment movement or scene changes. This ensures the spatial alignment accuracy between the inspection images and the cable 3D model in the initial fusion model, laying a reliable foundation for further optimization of occlusion issues and construction of an accurate cable fusion model.

[0107] As an optional embodiment, before determining the spatial change features corresponding to multiple target feature points based on video data, the method further includes: when the video data includes multiple object feature points, determining the position change index corresponding to each of the multiple object feature points based on the video data, wherein the corresponding position change index represents the degree of change in the spatial position of the corresponding object feature point, and the multiple object feature points are feature points used to reflect the shape features of objects on the ground; and determining multiple target feature points from the multiple object feature points based on the position change index corresponding to each of the multiple object feature points, wherein the multiple target feature points are object feature points whose position change index is less than a position change threshold among the multiple object feature points.

[0108] This embodiment describes the specific steps before determining the spatial variation features corresponding to multiple target feature points based on video data.

[0109] This involves multiple object feature points, which are a set of all feature points extracted from video data that can reflect the shape features of objects on the ground. These feature points cover the key shape parts of objects on the ground (such as manhole covers and road markings), such as outlines, corners, and textures, and serve as the basic data source for selecting target feature points.

[0110] This includes the position change index, which is used to quantify the degree of spatial position change (such as the magnitude of change) of object feature points in multiple consecutive frames of video data. It can be obtained by calculating the coordinate offset of the feature point between different frames, the length of the movement trajectory, etc. The smaller the value, the more stable the position of the feature point.

[0111] This involves spatial location, which is the specific location information of the object's feature points in three-dimensional space, and can be represented by spatial coordinates.

[0112] This involves the object's external shape features, which are the object's own outline shape, structural details, and other external attributes (such as the circular outline of a manhole cover or the rectangular border of a road marking). These features are the core basis for extracting object feature points, ensuring that the feature points can be accurately associated with the object on the ground.

[0113] This involves feature points, which are pixels or pixel regions with unique recognizability that can characterize specific parts of an object (such as corners or edges). Their position, grayscale value, texture, and other features can be used to track objects or achieve spatial alignment.

[0114] This involves a change threshold, which is a preset critical value used to determine the stability of the position of object feature points.

[0115] When video data contains multiple object feature points reflecting the shape characteristics of objects on the ground, the position change index of each object feature point (quantifying its spatial position change degree) is calculated based on the video data. This allows for the precise selection of feature points with position change indices less than a preset change threshold as target feature points. This effectively eliminates feature points whose positions are unstable due to moving objects, equipment vibration, or scene interference, ensuring that the target feature points have high stability and recognizability. This provides a reliable reference for subsequently determining spatial change characteristics based on these feature points and achieving accurate alignment between inspection images and cable 3D models, thus avoiding the impact of unstable feature points on the accuracy of the initial fusion model.

[0116] As an optional embodiment, based on spatial correlation parameters and depth data, the inspection image is fused with the cable 3D model to obtain a cable fusion model, including: determining the inspection trajectory data corresponding to the inspection equipment; determining the spatial pose deviation corresponding to the inspection equipment based on the inspection trajectory data; and fusing the inspection image with the cable 3D model based on the spatial pose deviation, spatial correlation parameters, and depth data to obtain the cable fusion model.

[0117] This embodiment describes the specific steps for fusing inspection images with a 3D cable model based on spatial correlation parameters and depth data to obtain a fused cable model.

[0118] This includes inspection trajectory data, which is spatiotemporal data used to reflect the movement path of the inspection equipment during the inspection process. It includes information such as three-dimensional coordinates, movement direction, and speed at different time points, and fully reflects the inspection movement trajectory of the equipment.

[0119] This involves spatial attitude deviation, which is the deviation between the actual spatial state of the inspection equipment and the theoretical preset state. For example, position deviation (the offset between the actual coordinates and the preset coordinates) and attitude deviation (the difference between the actual pitch angle, heading angle, roll angle and the preset angle).

[0120] By identifying the inspection trajectory data (including 3D coordinates and movement direction at different times) that reflects the movement path of the inspection equipment during inspection, the actual movement state of the equipment can be accurately captured. Based on this trajectory data, the spatial pose deviation (including position offset and attitude angle difference) between the actual spatial state and the theoretical preset state of the equipment can be analyzed. By combining spatial correlation parameters and depth data, the inspection image is fused with the 3D model of the cable, which can specifically correct the spatial alignment error caused by the equipment pose deviation. This ensures that the position and attitude of the real scene and the virtual cable model are accurately matched during the fusion process, avoiding fusion misalignment caused by equipment movement deviation, and finally obtaining a high-precision cable fusion model.

[0121] Based on the above embodiments and optional embodiments, an optional implementation method is provided, which is described in detail below.

[0122] In related technologies, underground cables serve as crucial carriers of energy transmission. Accurate positioning during inspections is key to preventing construction damage, quickly diagnosing faults, and ensuring operational safety and efficiency, directly impacting power supply reliability and on-site safety. However, existing technologies suffer from inaccurate underground cable location.

[0123] There is currently no effective solution to the above problems.

[0124] In view of this, an optional embodiment of the present invention provides a method for locating underground cables, which can effectively solve the above-mentioned technical problems.

[0125] Figure 2 This is a schematic diagram of the cable fusion model generation process in an optional embodiment of the present invention, such as... Figure 2 As shown below, a detailed description will be provided.

[0126] S1, acquire inspection images and equipment positioning data corresponding to the inspection equipment;

[0127] Before acquiring equipment positioning data, the process includes: acquiring satellite positioning (GNSS) data, inertial navigation (INS) data, and visual positioning (VSLAM) data corresponding to the inspection equipment; and performing real-time dynamic coupling based on the satellite positioning data, inertial navigation data, and visual positioning data to obtain the equipment positioning data. Specifically, GNSS is used as the global reference to acquire equipment positioning information; INS is used as a short-term stable positioning reference to correct the equipment's spatial positioning information; and VSLAM visual positioning methods are used to perform local correction of the equipment. This can be determined in the following ways:

[0128] A1 uses GNSS as a global reference, providing meter- to centimeter-level positioning in the geographic coordinate system, and can employ PPP-RTK technology. PPP-RTK technology can achieve a horizontal accuracy of 2cm, but it is easily affected by obstructions from buildings and other structures.

[0129] A2 uses INS as a short-term stabilizer and employs a 6-DOF MEMS-IMU to output pose data at frequencies above 100Hz. Position continuity is compensated for when GNSS signals are interrupted through trajectory extrapolation. However, inertial navigation has cumulative errors and requires periodic calibration.

[0130] Furthermore, A2 may also include:

[0131] A21: Employs a 6-DOF MEMS-IMU, integrating a three-axis accelerometer (range ±16g) and a gyroscope (zero bias instability <0.3° / h), outputting attitude data at a high frequency of 100Hz to achieve trajectory calculation.

[0132] A22: Utilizes GNSS-assisted correction, and uses Kalman filtering to fuse GNSS absolute position (accuracy 1.5m) with IMU relative displacement to suppress cumulative errors.

[0133] A3 uses VSLAM for local correction, implements feature point tracking based on monocular / binocular cameras, adjusts the model's pose using the PnP algorithm, and effectively corrects inertial navigation system (INS) drift by utilizing visual feature-based real-time localization and mapping technology (ORB-SLAM).

[0134] Furthermore, A3 may also include:

[0135] A31: Employs a feature point detection and descriptor extraction algorithm (Oriented FAST and Rotated BRIEF, ORB), which balances computational efficiency with rotation / lighting robustness, thereby generating 3D map points through triangulation and constructing a local map (accuracy 0.1-0.5m).

[0136] A32: The PnP algorithm is used to match 2D-3D feature points, eliminate outliers through nonlinear parameter estimation, solve the camera pose (6 degrees of freedom), and perform local auxiliary optimization. The camera pose and map points are jointly optimized within a sliding window to suppress cumulative errors.

[0137] A33: An INS correction mechanism is constructed. Specifically, the VSLAM output pose and the pre-integrated results of the inertial measurement unit (IMU) are fused using an extended Kalman filter (EKF) to correct the loose coupling of gyro drift (heading error <0.5° / min). Robustness is improved by directly optimizing the visual reprojection error and the IMU error term using methods such as vision-based simultaneous localization and mapping.

[0138] A4. An adaptive weight allocation algorithm is set to dynamically allocate weights to open areas and weak signal areas to meet the positioning requirements under different conditions.

[0139] Furthermore, A4 may also include:

[0140] A41: Open area: GNSS weight 70%, INS weight 25%, VSLAM weight 5%;

[0141] A42: Weak signal area: VSLAM weight increased to 60%, GNSS weight decreased to 15%, INS weight 25%.

[0142] S2. Based on the equipment positioning data, determine the three-dimensional model of the cable. The three-dimensional model of the cable is the three-dimensional data model of the underground cable corresponding to the inspection area. The inspection area is the area where the inspection equipment inspects the underground cable.

[0143] S3, determine the spatial association parameters between the inspection image and the cable 3D model, where the spatial association parameters are parameters used to represent the spatial association relationship between the inspection image and the cable 3D model;

[0144] S4, determine the depth data corresponding to the inspection image;

[0145] S5. Based on spatial correlation parameters and depth data, the inspection image is fused with the cable 3D model to obtain the cable fusion model.

[0146] Furthermore, S5 may also include:

[0147] When the spatial correlation parameters include a first correlation parameter and a second correlation parameter, the first correlation parameter represents the coordinate transformation relationship between the first spatial coordinate system and the third spatial coordinate system, and the second correlation parameter represents the coordinate transformation relationship between the second spatial coordinate system and the third spatial coordinate system. The first spatial coordinate system is the spatial coordinate system corresponding to the cable 3D model, the second spatial coordinate system is the spatial coordinate system corresponding to the inspection image, and the third spatial coordinate system is the spatial coordinate system corresponding to the inspection equipment. Based on the first correlation parameter, the second correlation parameter, and the depth data, the inspection image and the cable 3D model are fused to obtain a cable fused model.

[0148] Furthermore, based on the first correlation parameter, the second correlation parameter, and the depth data, the inspection image is fused with the cable 3D model to obtain a cable fusion model, including:

[0149] Based on the first and second correlation parameters, the inspection image is fused with the cable 3D model to obtain an initial fusion model; the video data corresponding to the inspection image is determined; when the video data includes multiple object feature points, the position change index corresponding to each of the multiple object feature points is determined based on the video data, where the corresponding position change index represents the degree of change in the spatial position of the corresponding object feature point, and the multiple object feature points are feature points used to reflect the shape characteristics of objects on the ground; based on the position change index corresponding to each of the multiple object feature points, multiple target feature points are determined from the multiple object feature points, where the multiple target feature points are object feature points whose position change index is less than the position change threshold; based on the video data, the spatial change features corresponding to each of the multiple target feature points are determined, and the inspection image includes multiple target feature points; based on the spatial change features corresponding to each of the multiple target feature points, the first and second correlation parameters, the inspection image is fused with the cable 3D model to obtain the initial fusion model.

[0150] Specifically, based on the first and second correlation parameters, the inspection images are fused with the cable's 3D model to obtain an initial fusion model. That is, based on precise positioning and visual calibration technology, depth matching between the 3D model and the terminal equipment's 2D image is achieved, eliminating spatial pose deviations between the equipment model and the actual scene, and solving the problem of accurately determining the location of underground cables during traditional construction briefings. Specifically, this includes:

[0151] B1, Coordinate System Mapping and Geometric Transformation: Through camera calibration, the mapping relationship between the world coordinate system (three-dimensional space, i.e., the first spatial coordinate system), the camera coordinate system (viewing angle, i.e., the third spatial coordinate system), and the image coordinate system (two-dimensional projection, i.e., the second spatial coordinate system) is established. Rigid body transformation (translation / rotation) is achieved using the extrinsic parameter matrix, and projection transformation is completed using the intrinsic parameter matrix. Distortion correction is used to improve accuracy.

[0152] Furthermore, B1 may also include:

[0153] B11: Use rigid body transformation to form the extrinsic parameter matrix (i.e., the first associated parameter), and then use the rotation matrix. Translation vector To achieve the transformation from the world coordinate system to the camera coordinate system:

[0154]

[0155] in, The coordinates of a 3D point in the camera coordinate system. The coordinates of a three-dimensional point in the world coordinate system. It is a translation vector. The rotation matrix can be represented by Euler angles, quaternions, or axis angles. Furthermore, the above equation transforms the 3D cable model in the first spatial coordinate system to the third spatial coordinate system.

[0156] B12: The perspective projection from the camera coordinate system to the image coordinate system is formed by using projection transformation to create the intrinsic parameter matrix (i.e., the second correlation parameter).

[0157]

[0158] in, This represents the pixel position in the horizontal direction (horizontal axis) of the image coordinate system. This represents the pixel position in the vertical direction (vertical axis) of the image coordinate system. For camera lens parameters (such as focal length). The coordinates are in the X-axis direction of the camera coordinate system. The Y-axis coordinate of the camera coordinate system. The coordinates are along the Z-axis of the camera coordinate system.

[0159] The intrinsic parameter matrix K can be represented as:

[0160]

[0161] in, Indicates the principal point offset; This refers to the axis tilt parameter; It is the focal length.

[0162] The final pixel coordinates are obtained through homogeneous coordinate transformation.

[0163] B2, Establish a six-degree-of-freedom error compensation model:

[0164]

[0165] in, This is the error compensation amount. This is either a scaling factor or a transformation matrix. For three-dimensional position error components, This represents the three-dimensional attitude error components.

[0166] The above formula describes With six variables The linear relationship between them Indicates the change or difference. In representing minute changes or variations, specifically, Represents minute changes in spatial coordinates. This represents a small change in angle or direction. The above formula can be used for rigid body motion analysis to describe the relationship between changes in position and attitude and a certain physical quantity (such as momentum or force), or to describe the relationship between the pose changes of an end effector and joint variables, or to represent the relationship between changes in momentum and spatial displacement and rotation.

[0167] B3 features real-time dynamic 3D reconstruction and collaborative acquisition of target area video. It achieves dynamic tracking by combining feature extraction (edge / corner detection), triangulation to calculate 3D coordinates, and Kalman filtering. The positioning error can be controlled within the centimeter level.

[0168] Furthermore, B3 may also include:

[0169] B31: Extract feature points (using ORB or FAST algorithms) and perform rotation / lighting robust matching, filtering out moving object interference through semantic segmentation. ORB, in particular, can extract stable feature points of ground objects from the inspection image, providing reliable visual markers for subsequent 3D coordinate calculation and pose tracking.

[0170] B32: Solving the 3D coordinates of feature points based on the epipolar-corrected image. :

[0171]

[0172] Where b is the baseline distance and d is the parallax.

[0173] Based on the above, a sparse model (ORB-SLAM) can be constructed first, and then a dense model can be generated through deep learning.

[0174] B33: Employs Kalman filtering for dynamic tracking optimization. The state vector includes position, velocity, and acceleration, and noise is suppressed through a prediction-correction mechanism. An IMU compensates for camera motion blur, and GNSS provides absolute position constraints.

[0175] B4, Pose Correction and Error Optimization, addresses pose deviations caused by device movement by using trajectory data acquired by a binocular camera to correct the pose of laser point clouds, and verifies it through evaluation standards such as straightness and flatness.

[0176] Furthermore, B4 may also include:

[0177] B41: Extract the six-degree-of-freedom (6DOF) pose for trajectory calculation.

[0178] B42: Motion modeling is performed by fitting a continuous trajectory using B-spline curves.

[0179] B43: Construct an objective function that includes reprojection error, iterative nearest point (ICP) error, and IMU constraints, and solve the objective function to achieve joint optimization.

[0180] B44: Error compensation is achieved by using RANSAC to remove outliers and sliding window optimization.

[0181] Further, based on the depth parameters corresponding to multiple pixels and the inspection image, object distribution parameters are determined. These object distribution parameters represent the spatial distribution characteristics of objects on the ground. The depth data includes the depth parameters corresponding to multiple pixels, and the inspection image includes multiple pixels. Based on the initial fusion model, cable distribution parameters are determined. These cable distribution parameters represent the spatial distribution characteristics of underground cables. Based on the object distribution parameters and cable distribution parameters, spatial correspondences are determined. These spatial correspondences represent the spatial correspondence between objects on the ground and underground cables. Based on these spatial correspondences, the occlusion parts corresponding to the underground cables and the occluding objects corresponding to the occlusion parts are determined. The occluding objects are ground objects whose occlusion index is greater than the occlusion threshold. Based on the feature characteristics corresponding to the occluding parts, visual parameters corresponding to the occluding objects are determined. Based on the visual parameters, the initial fusion model is subjected to occlusion adjustment processing to obtain the cable fusion model. This adjustment processing includes occlusion adjustment processing, which can be implemented in the following ways:

[0182] C1 utilizes a ToF sensor to generate millimeter-level precision depth maps in real time (±1mm error), and synchronizes with an RGB camera via hardware-level timestamps (<1ms error).

[0183] Furthermore, C1 may also include:

[0184] C11: Jointly calibrate the ToF and RGB camera extrinsics to ensure spatial coordinate alignment.

[0185] C12: Synchronization error is controlled within 1ms by timestamp interpolation or sliding window matching.

[0186] C13: Employs a zero-copy memory sharing mechanism to reduce data transfer latency.

[0187] C14: Separate threads process the ToF depth map and RGB image respectively, and timing consistency is ensured through a lockstep mechanism.

[0188] C2, combining IMU (1000Hz sampling rate) and SLAM algorithm, achieves centimeter-level spatial positioning in dynamic scenes.

[0189] Furthermore, C2 may also include:

[0190] C21: Real-time identification of dynamic objects such as pedestrians and vehicles in inspection images and removal of their corresponding feature points to avoid trajectory drift caused by dynamic interference, ensuring the accuracy of six-degree-of-freedom pose calculation and 3D mapping.

[0191] C22: Introducing motion consistency test: Judging dynamic objects by the residual between the IMU predicted trajectory and the visual matching result.

[0192] C23: Using maximum a posteriori probability (MAP) estimation, the initialization time is reduced to within 2 seconds, and the accuracy is improved by 3 times.

[0193] C24: Automatically creates a sub-map when tracking is lost, and merges it with the main map after loop closure detection to reduce cumulative error.

[0194] C3 implements layered rendering. The foreground layer uses high priority to render critical parts of the occluded cable (such as connectors), and enhances visibility through alpha blending. The background layer uses low resolution to render the static environment, saving computing power.

[0195] Furthermore, C3 may also include:

[0196] C31: Implements the highlight shader module to achieve a semi-transparent highlight effect for cable connectors, enhancing the visibility of occluded areas through edge detection algorithms. Includes a rendering queue to ensure highest priority and avoid depth conflicts.

[0197] C32: Set up a layered rendering controller, create a camera system, with the foreground camera handling the foreground layer (critical objects such as cable connectors), and the background camera rendering the static environment at 1 / 4 resolution. Dynamic resolution scaling of the background layer is achieved through render textures.

[0198] C33: Employs a dynamic occlusion detection system, using spherical projection to detect obstructed cable connectors, and achieves a smooth transition of the highlight effect through coroutines. When occlusion is detected, a highlight material is superimposed on the original material.

[0199] Based on depth perception-driven augmented reality (AR) layered rendering technology, the device uses its onboard ToF sensor to build an environmental depth map in real time, dynamically identify surface obstructions (such as manhole covers and vegetation), and automatically adjust the visualization effect of the cable model to ensure that key parts of the obscured equipment remain visible in the AR view.

[0200] Furthermore, S5 may also include:

[0201] Determine the inspection trajectory data corresponding to the inspection equipment; based on the inspection trajectory data, determine the spatial pose deviation corresponding to the inspection equipment; based on the spatial pose deviation, spatial correlation parameters, and depth data, fuse the inspection image with the cable 3D model to obtain the cable fusion model.

[0202] S6, based on the cable fusion model, locates the underground cable.

[0203] The above optional implementation methods can achieve at least the following beneficial effects:

[0204] (1) Compared with related technologies, this invention can lock the inspection area of ​​underground cables and obtain on-site real-world reference by acquiring inspection images and equipment positioning data of inspection equipment. Based on the equipment positioning data, the three-dimensional cable model corresponding to the area is selected to ensure that the model and the inspection scene are accurately matched. By determining the spatial correlation parameters between the inspection image and the three-dimensional cable model, the spatial mapping logic between the two is established. The scene stereo information is supplemented by the depth data corresponding to the inspection image. The inspection image and the three-dimensional cable model are fused to obtain the cable fusion model, realizing the accurate superposition of the real scene and the virtual model. Then, through the cable fusion model, the accurate positioning of underground cables can be achieved, thus solving the technical problem of inaccurate positioning of underground cables in related technologies.

[0205] (2) Compared with related technologies, the present invention obtains an initial fusion model by fusing the inspection image with the cable three-dimensional model based on the first correlation parameter and the second correlation parameter. This establishes the basic spatial correspondence between the two. Then, the object distribution parameters that characterize the spatial distribution features of objects on the ground are determined by the depth parameters corresponding to multiple pixels in the inspection image. The cable distribution parameters that describe the spatial distribution features of underground cables are extracted from the initial fusion model. Finally, the initial fusion model is optimized and corrected based on these two types of parameters. This can accurately solve problems such as unfitted occlusion and spatial misalignment in the initial model, ensuring that the fusion model is highly consistent with the real scene. Finally, a cable fusion model that can clearly present the spatial correspondence between underground cables and objects on the ground is obtained, providing reliable support for the accurate positioning of underground cables.

[0206] (3) Compared with related technologies, the present invention determines the spatial correspondence between above-ground objects and underground cables based on object distribution parameters and cable distribution parameters, which can provide a basis for occlusion judgment. Then, based on the spatial correspondence, the occlusion parts of the underground cable that overlap with the above-ground objects and the occlusion objects with an occlusion index greater than the occlusion threshold are locked. Then, the location, importance and other part characteristics of the occlusion parts are combined to determine the high brightness color value, transparency and other visual parameters. Finally, the initial fusion model is adjusted for occlusion based on these visual parameters. The key parts of the occluded cable can be highlighted through special visual effects such as semi-transparent highlighting and edge enhancement, eliminating the influence of above-ground objects on cable visualization, ensuring the complete presentation of the cable, and making the final cable fusion model more in line with the real inspection scenario.

[0207] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0208] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0209] Example 2

[0210] According to an embodiment of the present invention, an apparatus for implementing the above-described method for locating underground cables is also provided. Figure 3 This is a structural block diagram of an underground cable positioning device according to an embodiment of the present invention, such as... Figure 3 As shown, the device includes: an acquisition module 302, a first determination module 304, a second determination module 306, a third determination module 308, a fourth determination module 310, and a fifth determination module 312. The device will be described in detail below.

[0211] The system comprises: an acquisition module 302 for acquiring inspection images and equipment positioning data corresponding to the inspection equipment; a first determination module 304 connected to the acquisition module 302 for determining a three-dimensional cable model based on the equipment positioning data, wherein the three-dimensional cable model is a three-dimensional data model of the underground cable corresponding to the inspection area, and the inspection area is the area where the inspection equipment inspects the underground cable; a second determination module 306 connected to the first determination module 304 for determining the spatial relationship between the inspection image and the three-dimensional cable model; a third determination module 308 connected to the second determination module 306 for determining the depth data corresponding to the inspection image, wherein the depth data includes depth parameters corresponding to multiple pixels, and the inspection image includes multiple pixels; a fourth determination module 310 connected to the third determination module 308 for fusing the inspection image and the three-dimensional cable model based on the spatial relationship and the depth data to obtain a fused cable model; and a fifth determination module 312 connected to the fourth determination module 310 for locating the underground cable based on the fused cable model.

[0212] It should be noted that the above-mentioned acquisition module 302, first determination module 304, second determination module 306, third determination module 308, fourth determination module 310 and fifth determination module 312 correspond to steps S102 to S112 in the method for locating underground cables. The multiple modules and the corresponding steps are the same in terms of implementation examples and application scenarios, but are not limited to the content disclosed in the above embodiment 1.

[0213] Example 3

[0214] According to another aspect of the present invention, an electronic device is also provided, comprising: a processor; and a memory for storing processor-executable instructions, wherein the processor is configured to execute the instructions to implement the underground cable location method of any of the above.

[0215] Example 4

[0216] According to another aspect of the present invention, a computer-readable storage medium is also provided, which, when the instructions in the computer-readable storage medium are executed by a processor of an electronic device, enables the electronic device to perform the underground cable location method described above.

[0217] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0218] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0219] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0220] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0221] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0222] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0223] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for locating underground cables, characterized in that, include: Acquire inspection images and equipment positioning data corresponding to the inspection equipment; Based on the equipment positioning data, a three-dimensional model of the cable is determined, wherein the three-dimensional model of the cable is a three-dimensional data model of the underground cable corresponding to the inspection area, and the inspection area is the area where the inspection equipment inspects the underground cable; Determine the spatial association parameters between the inspection image and the cable 3D model, wherein the spatial association parameters are parameters used to represent the spatial association relationship between the inspection image and the cable 3D model; Determine the depth data corresponding to the inspection image; Based on the spatial correlation parameters and the depth data, the inspection image is fused with the cable 3D model to obtain a cable fusion model; Based on the cable fusion model, the underground cables are located.

2. The method according to claim 1, characterized in that, The process of fusing the inspection image with the cable 3D model based on the spatial correlation parameters and the depth data to obtain a cable fusion model includes: When the spatial association parameters include a first association parameter and a second association parameter, the first association parameter represents the coordinate transformation relationship between a first spatial coordinate system and a third spatial coordinate system, and the second association parameter represents the coordinate transformation relationship between a second spatial coordinate system and the third spatial coordinate system. The first spatial coordinate system is the spatial coordinate system corresponding to the three-dimensional model of the cable, the second spatial coordinate system is the spatial coordinate system corresponding to the inspection image, and the third spatial coordinate system is the spatial coordinate system corresponding to the inspection equipment. Based on the first correlation parameter, the second correlation parameter, and the depth data, the inspection image is fused with the cable 3D model to obtain a cable fusion model.

3. The method according to claim 2, characterized in that, The step of fusing the inspection image with the cable 3D model based on the first correlation parameter, the second correlation parameter, and the depth data to obtain a cable fusion model includes: Based on the first correlation parameter and the second correlation parameter, the inspection image is fused with the cable 3D model to obtain an initial fused model; Based on the depth parameters corresponding to multiple pixels and the inspection image, object distribution parameters are determined, wherein the object distribution parameters are used to represent the spatial distribution characteristics of objects on the ground, the depth data includes the depth parameters corresponding to multiple pixels, and the inspection image includes the multiple pixels. Based on the initial fusion model, cable distribution parameters are determined, wherein the cable distribution parameters are used to represent the spatial distribution characteristics of underground cables; Based on the object distribution parameters and the cable distribution parameters, the initial fusion model is adjusted to obtain the cable fusion model.

4. The method according to claim 3, characterized in that, The process of adjusting the initial fusion model based on the object distribution parameters and the cable distribution parameters to obtain the cable fusion model includes: Based on the object distribution parameters and the cable distribution parameters, a spatial correspondence is determined, wherein the spatial correspondence represents the spatial correspondence between the above-ground object and the underground cable; Based on the spatial correspondence, the obstruction part corresponding to the underground cable and the obstruction object corresponding to the obstruction part are determined, wherein the obstruction object is a ground object whose obstruction index of the obstruction part is greater than the obstruction threshold. Based on the location features corresponding to the occluded part, determine the visual parameters corresponding to the occluded object; Based on the visual parameters, the initial fusion model is subjected to occlusion adjustment processing to obtain the cable fusion model, wherein the adjustment processing includes occlusion adjustment processing.

5. The method according to claim 3, characterized in that, The step of fusing the inspection image with the cable 3D model based on the first association parameter and the second association parameter to obtain an initial fusion model includes: Determine the video data corresponding to the inspection image; Based on the video data, spatial variation features corresponding to multiple target feature points are determined, and the inspection image includes the multiple target feature points; Based on the spatial variation characteristics corresponding to the multiple target feature points, the first association parameter, and the second association parameter, the inspection image is fused with the cable 3D model to obtain an initial fusion model.

6. The method according to claim 5, characterized in that, Before determining the spatial variation features corresponding to multiple target feature points based on the video data, the method further includes: When the video data includes multiple object feature points, a position change index corresponding to each of the multiple object feature points is determined based on the video data. The corresponding position change index represents the degree of change in the spatial position of the corresponding object feature point. The multiple object feature points are feature points used to reflect the shape characteristics of the objects on the ground. Based on the position change index corresponding to the multiple object feature points, multiple target feature points are determined from the multiple object feature points, wherein the multiple target feature points are the object feature points whose position change index is less than the position change threshold among the multiple object feature points.

7. The method according to any one of claims 1 to 6, characterized in that, The process of fusing the inspection image with the cable 3D model based on the spatial correlation parameters and the depth data to obtain a cable fusion model includes: Determine the inspection trajectory data corresponding to the inspection equipment; Based on the inspection trajectory data, determine the spatial pose deviation corresponding to the inspection equipment; Based on the spatial pose deviation, the spatial correlation parameters, and the depth data, the inspection image is fused with the cable 3D model to obtain a cable fusion model.

8. A positioning device for underground cables, characterized in that, include: The acquisition module is used to acquire inspection images and equipment positioning data corresponding to the inspection equipment; The first determining module is used to determine a three-dimensional model of the cable based on the equipment positioning data, wherein the three-dimensional model of the cable is a three-dimensional data model of the underground cable corresponding to the inspection area, and the inspection area is the area where the inspection equipment inspects the underground cable. The second determining module is used to determine the spatial relationship between the inspection image and the three-dimensional model of the cable; The third determining module is used to determine the depth data corresponding to the inspection image, wherein the depth data includes depth parameters corresponding to multiple pixels, and the inspection image includes the multiple pixels; The fourth determining module is used to fuse the inspection image with the cable 3D model based on the spatial correlation and the depth data to obtain a cable fusion model. The fifth determining module is used to locate the underground cable based on the cable fusion model.

9. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the method for locating underground cables as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, When the instructions in the computer-readable storage medium are executed by the processor of the electronic device, the electronic device is able to perform the method for locating underground cables as described in any one of claims 1 to 7.

Images