Railway facility visualized browsing publishing method, system, device and medium

Abstract

A railway facility visual browsing publishing method, system, device and medium. In the method, laser point cloud, panoramic image and exposure point pose data are extracted; the point cloud is mapped to generate a structured depth map, which is indexed and associated with the panoramic image and then loaded in an interactive window; interactive click coordinates for the panoramic image are collected, and the depth map corresponding pixel row and column coordinates are obtained through spherical texture mapping; point depth values are extracted, combined with a space transformation matrix and pose data to calculate three-dimensional absolute space coordinates; then facility size data is obtained by performing geometric space calculation according to instructions, and is superimposed on the panoramic image together with a railway vector layer for joint rendering, and a space linkage visual picture is output, thereby effectively simplifying the detection operation process, reducing the time-consuming of repeated area searching, and helping to improve the operation efficiency of railway facility daily inspection and management.

Description

Technical Field

[0001] This application relates to the field of railway facility inspection technology, specifically to a method, system, equipment, and medium for visual browsing and publishing of railway facilities. Background Technology

[0002] Current railway line inspection operations involve using track inspection vehicles equipped with panoramic cameras and laser scanning equipment to travel along the railway line and collect panoramic image data and laser point cloud data. After the panoramic image data and laser point cloud data are collected, they are stored on separate image browsing platforms and point cloud processing software, respectively. After completing the appearance inspection of the facilities on the image platform, the operators must switch to the point cloud processing software to re-search for the corresponding areas before they can perform spatial geometric measurement operations.

[0003] Under the above-mentioned operating mode, the railway facility inspection process involves multiple platform operations, the operation process is cumbersome, the overall inspection efficiency is low, and it is difficult to meet the actual operation needs of daily management of railway line facilities. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides a method, system, equipment, and medium for visual browsing and publishing of railway facilities.

[0005] The first aspect of this application provides a method for visually browsing and publishing railway facilities, employing the following technical solution: Receive multi-source initial spatial data of railway facilities, and extract laser point cloud data, panoramic image data, and exposure point position and attitude data associated with the panoramic image data contained in the multi-source initial spatial data; Based on the preset laser angle resolution, the laser point cloud data is mapped onto a planar mesh model that matches the position and attitude data of the exposure point, thereby generating a structured depth map file; Extract the common image code between the panoramic image data and the structured depth map file, and establish an index association relationship between the panoramic image data and the structured depth map file based on the common image code; Based on the index association, the panoramic image data and the structured depth map file are loaded into the interactive window of the preset display control interface; Collect the coordinates of interactive click actions targeting the panoramic image data within the interactive window; Extract the spherical texture coordinates of the interactive click action coordinates onto the preset panoramic spherical model, and convert the spherical texture coordinates into the pixel row and column coordinates corresponding to the structured depth map file based on the preset spherical texture mapping conversion formula; Extract the depth value of the point mapped to the region pointed to by the pixel row and column coordinates, construct and call the spatial pose fusion transformation matrix to perform spatial transformation calculation on the depth value of the point in combination with the exposure point position and pose data, and obtain the three-dimensional absolute spatial coordinates; According to the preset measurement instructions, the three-dimensional absolute spatial coordinates are captured and geometric spatial measurement is performed to obtain facility size measurement data. The preset railway vector layer data is extracted, and the railway vector layer data and the facility size measurement data are superimposed on the panoramic image data to perform joint rendering. The spatial linkage visualization of railway facilities is output in the interactive window.

[0006] By adopting the above technical solution, the laser point cloud is converted into a structured depth map associated with the panoramic image index. By using spherical mapping and spatial pose fusion, operators can directly click on the panoramic image to obtain three-dimensional coordinates and facility measurement data. This eliminates the need to switch repeatedly between the image platform and the point cloud software, effectively simplifies the inspection operation process, reduces the time spent on repeated area searches, and helps improve the efficiency of daily inspection and management of railway facilities.

[0007] Optionally, the step of mapping the laser point cloud data onto a planar mesh model matching the exposure point position and pose data based on a preset laser angle resolution to generate a structured depth map file includes: Using the origin of the spatial coordinates in the exposure point position and attitude data as the reference point, the straight-line spatial distance from each laser reflection point in the laser point cloud data to the reference point is calculated point by point to obtain the initial depth value corresponding to each laser reflection point. Extract the horizontal and vertical angular resolution parameters from the laser angular resolution. Based on the horizontal angular resolution parameters, map each laser reflection point to the column position of the planar mesh model. Based on the vertical angular resolution parameters, map each laser reflection point to the row position of the planar mesh model to obtain the row index value and column index value corresponding to each laser reflection point. The row index value, the column index value, and the corresponding initial depth value are written into the structured depth map file according to a preset depth matrix format.

[0008] By adopting the above technical solution, discrete laser point clouds are transformed into regular structured depth maps with the exposure point as the center and according to the angular resolution. This makes the depth information correspond one-to-one with the pixels of the panoramic image. When the operator clicks on the target position on the panoramic image, the corresponding depth value can be directly extracted without switching to the point cloud software to find the corresponding area again. This helps to simplify the spatial measurement operation steps, reduce the complexity of the operation, and improve the efficiency and convenience of railway facility inspection.

[0009] Optionally, the step of extracting the spherical texture coordinates mapped from the interactive click action coordinates onto a preset panoramic spherical model, and converting the spherical texture coordinates into pixel row and column coordinates corresponding to the structured depth map file based on a preset spherical texture mapping conversion formula, includes: Using the center of the panoramic spherical model as a virtual observation base point, and based on the spherical texture mapping conversion formula, a gaze vector is constructed pointing from the virtual observation base point to the coordinates of the interactive click action. The intersection of the gaze vector and the inner surface of the panoramic spherical model is calculated as the target projection point. Read the normalized horizontal texture component and normalized vertical texture component of the target projection point in the preset panoramic spherical model texture coordinate system, and combine the normalized horizontal texture component and the normalized vertical texture component to establish the spherical texture coordinates. Extract the total number of pixels in the width direction and the total number of pixels in the height direction from the structured depth map file. Multiply the normalized horizontal texture component in the spherical texture coordinates by the total number of pixels in the width direction and round it to obtain the column coordinate value. Multiply the normalized vertical texture component in the spherical texture coordinates by the total number of pixels in the height direction and round it to obtain the row coordinate value. Combine the column coordinate value and the row coordinate value to obtain the pixel row and column coordinates.

[0010] By adopting the above technical solution, and utilizing the center-of-sphere projection and texture normalization conversion, the click position on the panoramic image is directly mapped to the pixel row and column coordinates of the structured depth map, establishing a fast correspondence channel between interactive operation and depth information reading. This reduces the time spent searching for the corresponding area again in the point cloud software, making the spatial positioning response more direct and helping to improve the continuity and detection efficiency of railway facility measurement operations.

[0011] Optionally, the step of extracting the point depth value of the region mapped and matched by the pixel row and column coordinates, constructing and calling the spatial pose fusion transformation matrix in combination with the exposure point position and pose data to perform spatial transformation calculation on the point depth value to obtain three-dimensional absolute spatial coordinates includes: Based on the pixel row and column coordinates, the corresponding storage location is retrieved in the structured depth map file, the depth value recorded at the storage location is read, and the depth value is confirmed as the depth value of the point. Extract the attitude angle parameters from the exposure point position attitude data, extract the device error compensation parameters from the preset device error compensation parameter table, and construct the spatial pose fusion transformation matrix based on the attitude angle parameters and the device error compensation parameters; By combining the pixel row and column coordinates with the point depth value, an intermediate space vector is constructed based on the camera coordinate system corresponding to the exposure point position and pose data. The intermediate spatial vector is input into the spatial pose fusion transformation matrix to perform a rotation transformation operation to obtain the relative coordinates. The exposure point world coordinates are extracted from the exposure point position and pose data. The exposure point world coordinates and the relative coordinates are then translated and superimposed to obtain the three-dimensional absolute spatial coordinates.

[0012] By adopting the above technical solution, a spatial pose fusion transformation matrix is ​​constructed using depth map retrieval and attitude error compensation, restoring two-dimensional click coordinates to three-dimensional absolute spatial coordinates. This achieves accurate calculation of the real spatial position from panoramic image interactive operation, reduces the repeated switching and manual measurement between panoramic images and point cloud software by operators, and helps to improve the continuity and detection efficiency of geometric measurement of railway facilities.

[0013] Optionally, the step of extracting the attitude angle parameters from the exposure point position attitude data, extracting the device error compensation parameters from a preset device error compensation parameter table, and constructing the spatial pose fusion transformation matrix based on the attitude angle parameters and the device error compensation parameters includes: Extract the attitude angle parameters from the exposure point position attitude data, construct multiple axial rotation matrices according to a preset rotation order, and multiply the multiple axial rotation matrices in sequence to obtain the global attitude rotation matrix. Lens deformation correction parameters and sensor shake correction parameters are extracted from the equipment error compensation parameter table as the equipment error compensation parameters, and a local error rotation matrix is ​​constructed based on the lens deformation correction parameters and sensor shake correction parameters; Multiplying the local error rotation matrix by the global attitude rotation matrix yields the spatial pose fusion transformation matrix.

[0014] By adopting the above technical solution, the global attitude rotation matrix is ​​multiplied and fused with the local error matrix that includes lens deformation and sensor jitter correction, and a spatial pose fusion transformation matrix that can synchronously compensate for the device's own attitude and acquisition error is constructed. This helps to improve the accuracy of the conversion from two-dimensional click coordinates to three-dimensional spatial coordinates, reduce the registration deviation when fusing multi-source data, and enhance the reliability of directly measuring facilities in panoramic images.

[0015] Optionally, the step of capturing the three-dimensional absolute spatial coordinates according to preset measurement instructions and performing geometric spatial measurement to obtain facility size measurement data includes: The measurement type parameter in the measurement instruction is parsed, and the corresponding geometric calculation rule is matched in the preset measurement rule library according to the measurement type parameter; The number of three-dimensional absolute space coordinates to be captured is determined according to the geometric calculation rules, and the corresponding number of three-dimensional absolute space coordinates are captured in sequence to form a measurement endpoint set. Substitute the set of measurement endpoints into the geometric calculation rules to perform geometric spatial measurement, and encapsulate the measurement results with the measurement type parameters to obtain the facility size measurement data.

[0016] By adopting the above technical solution, the measurement type parameters are analyzed and matched with the corresponding geometric calculation rules, and the corresponding number of three-dimensional spatial coordinates are automatically captured to complete the measurement calculation. This transforms various facility measurement needs into a standardized automatic calculation process, reducing the time spent by operators manually picking up coordinates and performing point-by-point conversions. This helps to improve the execution efficiency and consistency of results in the measurement of geometric parameters of railway facilities.

[0017] Optionally, the step of substituting the set of measurement endpoints into the geometric calculation rules to perform geometric spatial measurement, and associating and encapsulating the measurement results with the measurement type parameters to obtain the facility size measurement data includes: Extract the horizontal and vertical coordinate components of each of the three-dimensional absolute spatial coordinates from the set of measurement endpoints; According to the geometric calculation rules, the horizontal coordinate component values ​​are subjected to planar projection distance calculation to obtain the horizontal measurement result; and the vertical coordinate component values ​​are subjected to elevation difference calculation to obtain the elevation measurement result. The horizontal measurement results, the elevation measurement results, and the measurement type parameters are associated with each other according to a preset data encapsulation format to obtain the facility size measurement data.

[0018] By adopting the above technical solution, the three-dimensional spatial coordinates are independently calculated for planar distance and elevation difference in the horizontal and vertical directions, and the results are associated with the measurement type and encapsulated. This allows the output of geometric measurement to directly correspond to the two-dimensional span and elevation difference information of the facility, reducing the manual conversion steps for operators in panoramic images and helping to improve the intuitiveness of railway facility size measurement and data reading efficiency.

[0019] A second aspect of this application provides a railway facility visualization browsing and publishing system, comprising: The data receiving module is configured to receive multi-source initial spatial data of railway facilities and extract laser point cloud data, panoramic image data and exposure point position and attitude data associated with the panoramic image data contained in the multi-source initial spatial data. The depth map generation module is configured to map the laser point cloud data onto a planar mesh model that matches the position and attitude data of the exposure point based on a preset laser angle resolution, thereby generating a structured depth map file. The index association module is configured to extract the common image code between the panoramic image data and the structured depth map file, and establish an index association relationship between the panoramic image data and the structured depth map file based on the common image code; The data loading module is configured to load the panoramic image data and the structured depth map file in the interactive window of the preset display control interface according to the index association relationship. The interactive acquisition module is configured to acquire the coordinates of interactive click actions in response to the panoramic image data within the interactive window; The coordinate transformation module is configured to extract the spherical texture coordinates mapped from the interactive click action coordinates onto a preset panoramic spherical model, and convert the spherical texture coordinates into pixel row and column coordinates corresponding to the structured depth map file based on a preset spherical texture mapping conversion formula. The spatial coordinate calculation module is configured to extract the point depth value of the region mapped and matched by the pixel row and column coordinates, construct and call the spatial pose fusion transformation matrix to perform spatial transformation calculation on the point depth value in combination with the exposure point position and pose data, and obtain three-dimensional absolute spatial coordinates; The measurement and rendering module is configured to capture the three-dimensional absolute spatial coordinates according to the preset measurement instructions, perform geometric spatial measurement calculations to obtain facility size measurement data, extract preset railway vector layer data, overlay the railway vector layer data and the facility size measurement data on the panoramic image data to perform joint rendering, and output a spatial linkage visualization of railway facilities in the interactive window.

[0020] A third aspect of this application provides an electronic device including a processor, a memory, a user interface, and a network interface, wherein the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described in any of the foregoing.

[0021] A fourth aspect of this application provides a computer-readable storage medium storing instructions that, when executed, perform the method described in any of the preceding descriptions.

[0022] In summary, one or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: By converting laser point cloud data into a structured depth map that is depth-correlated with panoramic images, and using spherical texture mapping and spatial pose transformation with error compensation from fusion equipment, multi-source spatial data are deeply fused within a unified interactive window. This allows operators to directly click on the panoramic image to obtain accurate 3D spatial coordinates and facility measurement data. Simultaneously, the railway vector layer is overlaid with the measurement results for rendering, realizing integrated operation of image browsing, spatial measurement, and facility information visualization. This reduces the time spent repeatedly switching between multiple platforms and repeatedly searching for areas, simplifies the operation process of railway facility inspection, and helps improve the overall efficiency of daily management of line facilities. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the system architecture of an embodiment of a railway facility visualization browsing and publishing method or a railway facility visualization browsing and publishing system applying this application; Figure 2 This is a flowchart illustrating a method for visually browsing and publishing railway facilities as disclosed in an embodiment of this application; Figure 3 This is a schematic diagram of a module of a railway facility visualization browsing and publishing system disclosed in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of an electronic device disclosed in an embodiment of this application.

[0024] Explanation of reference numerals in the attached figures: 100, System architecture; 101, First terminal device; 102, Second terminal device; 103, Third terminal device; 104, Network; 105, Server; 401, Processor; 402, Communication bus; 403, User interface; 404, Network interface; 405, Memory. Detailed Implementation

[0025] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0026] In the description of the embodiments of this application, the words "for example" or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design that is described as "for example" or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design options. Rather, the use of the words "for example" or "for instance" is intended to present the relevant concepts in a specific manner.

[0027] In the description of the embodiments of this application, the term "multiple" means two or more. For example, multiple systems means two or more systems, and multiple screen terminals means two or more screen terminals. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0028] like Figure 1 As shown, the system architecture used in this application embodiment can adopt a bottom-up layered organization form, including a data layer, a core function layer, an interface layer, and an application layer. Each layer communicates with the other through a preset data channel and calling interface to exchange data and commands, so as to support the overall operation process of visual browsing and publishing of railway facilities.

[0029] The data layer is used for unified storage and management of multi-source basic data involved in the visualization and publishing of railway facilities. It includes a panoramic database, a vector database, a point cloud database, a POI database, and a configuration database. Specifically, the panoramic database stores panoramic image data collected along the railway line and exposure point position and attitude data associated with the panoramic image data; the vector database stores railway vector layer data such as the line centerline, track edge lines, mileage markers, catenary poles, signal lights, turnouts, and platform boundaries or clearance contours; the point cloud database stores laser point cloud data collected along the line by laser scanning equipment, and structured depth map files generated from the laser point cloud data; the POI database stores the location information, attribute information, and associated annotation data of points of interest (POIs) for railway facilities; and the configuration database stores laser angular resolution, spherical texture mapping conversion formulas, equipment error compensation parameter tables, measurement rule bases, coordinate system identifiers, and other configuration parameters required for system operation. The specific data storage medium for the data layer can be one or more combinations of relational databases, key-value databases, file systems, or object storage; this application does not limit this.

[0030] The core functional layer carries the main business logic of the railway facility visualization browsing and publishing method, including a panoramic roaming module, a POI annotation module, a layer loading module, a measurement tool module, and a tool setting module. Specifically, the panoramic roaming module, based on panoramic image data from the panoramic database, enables viewpoint rotation, zooming, panning, and exposure point switching along the railway line direction within an interactive window of a preset display control interface; the POI annotation module, combining point-of-interest data from the POI database, overlays railway facility annotation objects onto the panoramic image data; the layer loading module, based on index relationships, loads panoramic image data and structured depth map files, and overlays railway vector layer data from the vector database onto the interactive window as needed; the measurement tool module, responding to measurement commands, performs 3D absolute spatial coordinate calculation and geometric spatial measurement based on the structured depth map file and spatial pose fusion transformation matrix to obtain facility size measurement data; and the tool setting module loads operating parameters from the configuration database, configuring and updating measurement accuracy, display units, coordinate system identifiers, and rendering styles.

[0031] The interface layer provides standardized data and command channels between the core functional layer and the application layer, as well as with external systems. It includes API interfaces and communication adaptation units. The API interfaces provide services such as data querying, measurement calls, layer subscriptions, and rendering control in the form of RESTful interfaces, RPC interfaces, or WebSocket interfaces. The communication adaptation unit adapts and converts the communication protocols, data formats, and coordinate system identifiers of different application terminals and external platforms to ensure consistent data interaction across terminals and platforms.

[0032] The application layer serves as the entry point for interaction and invocation between operators and external systems. It includes user terminals and external platforms. The user terminals display a spatially linked visualization of railway facilities within an interactive window and receive interactive operations from operators. The external platforms can be third-party systems such as railway facility management platforms, engineering surveying platforms, geographic information platforms, or dispatching and command platforms. These systems interact with the core functional layer through the interface layer to obtain business data such as panoramic image data, structured depth map files, facility dimension measurement data, and railway vector layer data.

[0033] The user terminal and external platform can be hardware or software. When they are hardware, they can be various electronic devices with displays, including but not limited to smartphones, tablets, laptops, desktop computers, and embedded operating terminals. When they are software, they can be installed in the aforementioned electronic devices and can be implemented as a single software module or multiple software modules for providing distributed services; this application does not limit this.

[0034] This embodiment discloses a method for visually browsing and publishing railway facilities. Figure 2 This is a flowchart illustrating a method for visually browsing and publishing railway facilities as disclosed in an embodiment of this application. Figure 2 As shown, the method includes the following steps: S201. Receive multi-source spatial initial data of railway facilities, and extract laser point cloud data, panoramic image data, and exposure point position and attitude data associated with the panoramic image data contained in the multi-source spatial initial data. Specifically, the initial spatial data from multiple sources can be acquired by track inspection vehicles, railway inspection vehicles, mobile measurement systems, or other acquisition devices. The system can obtain this data through file import, database reading, or real-time reception via network interface. For example, laser point cloud data can be in LAS, LAZ, PCD, or PLY format; panoramic image data can be in JPG, PNG, or TIFF format; and exposure point position and attitude data can be in CSV, JSON, TXT, or database records. After receiving the data, the system parses and matches various types of data according to image encoding, file name, acquisition timestamp, or mileage information, extracting the point cloud data and exposure point position and attitude data corresponding to each panoramic image. The laser point cloud data includes at least the three-dimensional coordinates of multiple laser reflection points, and may further include information such as reflection intensity and scanning time. The panoramic image data provides the real-world texture of railway facilities. The exposure point position and attitude data includes at least the image encoding of the panoramic image, the world coordinates of the exposure point, and attitude angle parameters, which may include yaw angle, pitch angle, and roll angle. To ensure subsequent processing is feasible, the system can also perform format verification, integrity verification, and coordinate system consistency checks on the received data, and generate exception prompts when data is missing or mismatched. Thus, the system can obtain the basic data required for subsequent generation of structured depth maps, image indexing, and 3D coordinate calculation.

[0035] S202. Based on the preset laser angle resolution, the laser point cloud data is mapped onto a planar mesh model that matches the position and attitude data of the exposure point to generate a structured depth map file. Specifically, the system uses the exposure point position and pose data corresponding to the current panoramic image as a reference to determine the world coordinates of the exposure point in the panoramic image, and uses these world coordinates as the local calculation origin. Then, it filters laser reflection points located within a preset spatial range, preset time range, or preset mileage range from the laser point cloud data, calculates the straight-line spatial distance from each laser reflection point to the local calculation origin, and obtains the corresponding initial depth value. The system further reads a preset laser angle resolution, which includes horizontal and vertical angular resolution parameters. Based on the horizontal angular resolution parameters, it determines the column division rules of the planar mesh model, and based on the vertical angular resolution parameters, it determines the row division rules of the planar mesh model. For any laser reflection point, the system calculates its column index and row index value in the planar mesh model based on the horizontal and vertical angles of the laser reflection point relative to the exposure point. For example, the horizontal angle can be divided by the horizontal angular resolution and rounded to obtain the column index value, and the vertical angle can be divided by the vertical angular resolution and rounded to obtain the row index value. Then, the system writes the row index value, column index value, and corresponding initial depth value into a structured depth map file according to a preset depth matrix format, so that each valid pixel position in the structured depth map file records the depth value in the corresponding spatial direction. When multiple laser reflection points are mapped to the same grid position, the depth value closest to the exposure point, the average of multiple depth values, or the depth value corresponding to the point with the highest reflection intensity can be selected as the depth value of that grid position. When there is no valid laser reflection point at a certain grid position, a preset invalid depth identifier can be written. The structured depth map file can be stored in the form of a single-channel floating-point matrix, a binary matrix file, a GeoTIFF file, or a custom depth file. Its file header can record image encoding, grid width, grid height, horizontal angular resolution, vertical angular resolution, invalid depth identifier, and corresponding exposure point position and pose data.

[0036] Optionally, the step of mapping the laser point cloud data to a planar mesh model matching the exposure point position and attitude data based on a preset laser angular resolution to generate a structured depth map file includes: using the origin of the spatial coordinates in the exposure point position and attitude data as a reference point, calculating the straight-line spatial distance from each laser reflection point in the laser point cloud data to the reference point point to obtain the initial depth value corresponding to each laser reflection point; extracting the horizontal angular resolution parameter and the vertical angular resolution parameter from the laser angular resolution, mapping each laser reflection point to the column position of the planar mesh model according to the horizontal angular resolution parameter, and mapping each laser reflection point to the row position of the planar mesh model according to the vertical angular resolution parameter to obtain the row index value and column index value corresponding to each laser reflection point; and writing the row index value, the column index value, and the corresponding initial depth value into the structured depth map file according to a preset depth matrix format.

[0037] As an optional implementation, if the coordinates of the reference point are O(X0, Y0, Z0) and the coordinates of any laser reflection point are P(Xi, Yi, Zi), then the system can calculate the initial depth value of the laser reflection point according to the three-dimensional Euclidean distance formula, i.e., Di = sqrt((Xi - X0)). 2 +(Yi-Y0) 2 +(Zi-Z0) 2 In this process, the origin of the spatial coordinates in the exposure point position and attitude data can be the exposure center of the panoramic camera, the origin of the laser scanning coordinate system, or the origin of a unified coordinate system after external parameter calibration. The coordinates of the laser reflection point can be pre-converted to a coordinate reference system consistent with the reference point. Thus, the true spatial distance of each laser reflection point relative to the current panoramic image exposure position can be obtained.

[0038] Furthermore, the system calculates the horizontal and vertical angles of each laser reflection point relative to the reference point. For example, the horizontal angle can be obtained from atan2(Yi-Y0, Xi-X0), and the vertical angle can be obtained from atan2(Zi-Z0, sqrt((Xi-X0)). 2 +(Yi-Y0) 2 The system then discretizes the horizontal angle according to the horizontal angular resolution parameter to obtain the column index value of the laser reflection point in the planar grid model, and discretizes the vertical angle according to the vertical angular resolution parameter to obtain the row index value of the laser reflection point in the planar grid model. The horizontal and vertical angular resolution parameters can be pre-configured according to the lidar model, point cloud density, panoramic image resolution, or computational accuracy requirements; this application does not impose any limitations on these parameters.

[0039] Furthermore, the system can establish a two-dimensional depth matrix, where matrix rows correspond to vertical angular directions and matrix columns correspond to horizontal angular directions. Matrix elements are used to store depth values ​​in the corresponding directions. For multiple laser reflection points mapped to the same row and column position, the system can select the initial depth value closest to the reference point, the average of multiple initial depth values, or the initial depth value corresponding to the laser reflection point with the highest reflection intensity and write it into the matrix element. For positions not mapped to valid laser reflection points, the system can write preset invalid values. The structured depth map file can be stored in a binary matrix file, a single-channel floating-point image, a GeoTIFF file, or a custom file format, and can record image encoding, matrix size, angular resolution, invalid value identifier, and coordinate system information to facilitate rapid retrieval of corresponding depth values ​​based on image position.

[0040] S203. Extract the common image code between the panoramic image data and the structured depth map file, and establish an index association relationship between the panoramic image data and the structured depth map file based on the common image code; Specifically, the system can parse the filename, header information, metadata table, or acquisition record of the panoramic image data to obtain the image code that uniquely identifies the panoramic image. Simultaneously, it can parse the filename, header fields, index table, or auxiliary configuration file of the structured depth map file to obtain the corresponding depth map image code. The system then performs consistency matching on the two types of image codes, identifying the image code that exists simultaneously in both the panoramic image data and the structured depth map file as the common image code. The image code can be a frame number generated by the acquisition device, a timestamp, a mileage number, a unique image identifier, or a composite code formed by combining the line number, acquisition date, and frame sequence number; this application does not limit this. After determining the common image code, the system can establish an index association table. This index association table includes at least the common image code, the panoramic image file storage path, and the structured depth map file storage path, and may further include fields such as exposure point position and attitude data, acquisition time, line mileage, coordinate system identifier, and data version number. In one implementation, the index association relationship can be stored in a relational database, such as MySQL, PostgreSQL, or SQLite; in another implementation, it can also be stored in a key-value database, a JSON index file, an XML configuration file, or an in-memory hash table. To ensure accurate association, the system can also perform deduplication, format normalization, and integrity verification on shared image codes before establishing index associations. When the same image code corresponds to multiple panoramic image files or multiple structured depth map files, the final associated object can be determined based on the principle of the most recent acquisition time, data version priority, or file integrity verification results. Through the above processing, the system can form a fast retrieval channel from panoramic images to structured depth map files, enabling users to directly find the corresponding structured depth map file based on the image code when selecting a panoramic image or image location, thereby improving the efficiency of 3D information query and coordinate calculation for railway facilities.

[0041] S204. Based on the index association relationship, load the panoramic image data and the structured depth map file into the interactive window of the preset display control interface; Specifically, when a user selects an image code, route segment, acquisition time, or mileage location in the preset display control interface, the system queries the panoramic image file path and structured depth map file path corresponding to that image code based on the aforementioned index association relationship, and reads the corresponding panoramic image data and structured depth map file respectively. The panoramic image data can be loaded into the image rendering layer of the interactive window to display a rotatable, scalable, or pannable panoramic image. The structured depth map file can be loaded into the depth data cache layer of the interactive window to provide the corresponding depth value when the user clicks or selects an image location. The preset display control interface can be a desktop application interface, a browser webpage interface, a mobile terminal interface, or an embedded terminal interface. The interactive window can be implemented as a two-dimensional panoramic unfolding window, a spherical panoramic window, or a three-dimensional scene window. In one implementation, the system attaches the panoramic image as a texture to the inside of a spherical model and sets the user's viewpoint near the center of the sphere to achieve immersive panoramic browsing. In another implementation, the system displays the panoramic image on a two-dimensional canvas according to latitude and longitude, and controls the viewpoint movement through mouse, touch, or keyboard events. When loading structured depth map files, the system can read the depth matrix, matrix size, angular resolution, invalid value identifiers, and coordinate system information into memory. It can also read the data in chunks as needed or create a cached index to accommodate real-time queries of large depth map files. To ensure spatial correspondence between panoramic imagery and structured depth map files within the interactive window, the system can perform consistency checks based on image encoding, exposure point position and pose data, or coordinate system identifiers in the index association relationship. After successful checks, the system binds the two to the same interactive object. If no matching structured depth map file is found, the system can load only the panoramic image and mark it as unsuitable for depth queries. Through these methods, the system can simultaneously load visual images and corresponding depth data within the same interactive window, providing a data foundation for subsequent depth reading, spatial coordinate calculation, or target facility location based on user-selected locations.

[0042] S205. Collect the coordinates of interactive click actions on the panoramic image data within the interactive window; Specifically, when a user performs interactive actions such as clicking, selecting, double-clicking, long-pressing, or selecting in the interactive window on the panoramic image using a mouse, touchscreen, stylus, trackball, or other input devices, the system listens for the corresponding input event and obtains the screen coordinates of the interactive action in the interactive window coordinate system. For example, it can obtain the horizontal and vertical pixel coordinates of the click point relative to the upper left corner of the interactive window, denoted as (u, v). In one implementation, if the panoramic image is displayed in the form of a two-dimensional latitude and longitude unfolded map, the system can directly convert the screen coordinates into image pixel coordinates in the panoramic image data based on the scaling ratio, translation offset, and image display area boundary of the interactive window. In another implementation, if the panoramic image is displayed in a spherical panoramic mode, the system can back-project the screen coordinates into a spatial ray emanating from the viewpoint based on the current camera viewpoint parameters, field of view, projection matrix, and window size, calculate the intersection point of the spatial ray with the panoramic sphere, and then convert the intersection point into latitude and longitude angular coordinates or pixel coordinates in the panoramic image. The system can also determine the validity of the collected interactive click action coordinates, such as whether the click point is located within the panoramic image display area, whether it falls within a valid area with queried depth, whether it is covered by occluded interface controls, and whether the corresponding image encoding has been bound to a structured depth map file. When the click coordinates are invalid, the system can discard the click action or return an invalid message. The interactive click action coordinates may include one or more of screen coordinates, image pixel coordinates, latitude and longitude angular coordinates, and line-of-sight vectors. The specific coordinate form can be configured according to the rendering method of the interactive window, and this application does not limit it.

[0043] S206. Extract the spherical texture coordinates mapped from the interactive click action coordinates onto the preset panoramic spherical model, and convert the spherical texture coordinates into the pixel row and column coordinates corresponding to the structured depth map file based on the preset spherical texture mapping conversion formula. Specifically, the panoramic spherical model can be a unit sphere or a sphere with a preset radius centered at the panoramic image acquisition point or the camera optical center. The panoramic image data is attached as a texture to the inner or outer surface of the spherical model. After the system obtains the coordinates of the user's interactive click action in the interactive window, it can back-project the screen coordinates into a three-dimensional ray emanating from the viewpoint based on the current window size, camera intrinsic parameters or virtual camera parameters, viewing direction, projection matrix, and model view matrix. The intersection point of the three-dimensional ray and the panoramic spherical model is then obtained, and the texture coordinates corresponding to the intersection point are designated as spherical texture coordinates, for example, denoted as (s, t), where s and t can be normalized texture coordinates, with values ​​ranging from 0 to 1. For panoramic images using latitude and longitude unfolded projection, the system can convert spherical texture coordinates into longitude and latitude angles of the panoramic image according to a preset spherical texture mapping conversion formula. For example, let the longitude angle θ = 2πs - π and the latitude angle φ = π / 2 - πt, or use an equivalent angle conversion relationship based on the origin direction and vertical axis direction of the specific texture coordinates. Subsequently, the system converts the longitude and latitude angles into pixel column coordinates col and pixel row coordinates row within the structured depth map file based on the width W, height H, horizontal field of view, vertical field of view, and pixel arrangement direction of the structured depth map file. For example, in the depth map and the panoramic image... When the scene map uses the same equidistant cylindrical projection, it can be converted using col=round(s×(W-1)) and row=round(t×(H-1)), or col=floor((θ-θmin) / (θmax-θmin)×W) and row=floor((φmax-φ) / (φmax-φmin)×H), where θmin, θmax, φmin, and φmax are the angular ranges covered by the depth map. Their specific values ​​can be determined based on the calibration parameters of the acquisition device, the depth map generation rules, or the file header information. This application does not impose any limitations on this. In another implementation, if the structured depth map file uses a hexahedral texture, fisheye projection, or block spherical projection, the system can first determine the number of the face or block it belongs to based on the three-dimensional direction vector of the intersection of the spheres, and then convert the corresponding pixel row and column coordinates according to the local texture mapping relationship of the face or block. To ensure the usability of the conversion results, the system can also perform boundary clipping, integerization, out-of-bounds checks, and invalid region checks on the pixel row and column coordinates. For example, it can limit `row` to the range [0, H-1] and `col` to the range [0, W-1], or return an invalid flag when the coordinates exceed the coverage area of ​​the depth map. Through these methods, the system can accurately map the user's visible click position in the panoramic image to the depth pixel position in the structured depth map file, providing a reliable data index for subsequently reading the depth value at that position and calculating the 3D spatial coordinates of the target point.

[0044] Optionally, the step of extracting the spherical texture coordinates mapped from the interactive click action coordinates onto a preset panoramic spherical model, and converting the spherical texture coordinates into pixel row and column coordinates corresponding to the structured depth map file based on a preset spherical texture mapping conversion formula, includes: using the center of the panoramic spherical model as a virtual observation base point, constructing a gaze vector pointing from the virtual observation base point to the interactive click action coordinates based on the spherical texture mapping conversion formula, calculating the intersection of the gaze vector and the inner surface of the panoramic spherical model as the target projection point; and reading the target projection point in the preset panoramic spherical model texture coordinates. The normalized horizontal texture component and normalized vertical texture component under the standard system are combined to establish the spherical texture coordinates; the total number of pixels in the width direction and the total number of pixels in the height direction of the structured depth map file are extracted; the normalized horizontal texture component in the spherical texture coordinates is multiplied by the total number of pixels in the width direction and rounded to obtain the column coordinate value; the normalized vertical texture component in the spherical texture coordinates is multiplied by the total number of pixels in the height direction and rounded to obtain the row coordinate value; the pixel row and column coordinates are obtained by combining the column coordinate value and the row coordinate value.

[0045] Specifically, the panoramic spherical model can be a spherical model with a radius of R, centered at the panoramic image acquisition point or the camera's optical center. The virtual observation base point can be set at the center of this spherical model. After the system obtains the coordinates of the interactive click action, it can back-project the two-dimensional screen coordinates into a three-dimensional direction vector based on the width, height, current viewing direction, field of view, projection matrix, and view matrix of the interactive window. This three-dimensional direction vector is then normalized to obtain the gaze vector. If the radius of the spherical model is R and the virtual observation base point is located at the center, the system can multiply the gaze vector by the spherical radius to obtain the coordinates of the intersection point of the gaze vector on the inner surface of the spherical surface, and determine this intersection point as the target projection point. In other implementations, the target projection point can also be calculated by finding the intersection of a ray and the spherical equation. Through this processing, the user's click position in the display interface can be accurately projected onto the panoramic spherical model.

[0046] Furthermore, the panoramic spherical model texture coordinate system can adopt a normalized texture coordinate system, with its horizontal texture component denoted as s and its vertical texture component denoted as t, both ranging from [0, 1]. Here, s represents the relative position of the target projection point in the horizontal or longitude direction of the panoramic image, and t represents the relative position of the target projection point in the vertical or latitude direction of the panoramic image. The system can directly read the texture coordinates returned by the rendering engine in the ray picking results, or it can calculate the corresponding latitude and longitude angles based on the three-dimensional coordinates of the target projection point, and then convert s and t according to the mapping relationship between latitude and longitude angles and texture coordinates. For example, the system can calculate the longitude and latitude angles based on the direction vector of the target projection point relative to the center of the sphere, then normalize the longitude angle to the horizontal texture component and the latitude angle to the vertical texture component. Thus, the system can obtain the spherical texture coordinates (s, t) corresponding to the user's click position.

[0047] Furthermore, the system can read the total number of pixels W in the width direction and the total number of pixels H in the height direction from the file header, metadata fields, index table, or parsing results of the structured depth map file. Then, it multiplies the normalized horizontal texture component s in the spherical texture coordinates by the total number of pixels W in the width direction and rounds it to obtain the column coordinate value col. It then multiplies the normalized vertical texture component t by the total number of pixels H in the height direction and rounds it to obtain the row coordinate value row. Finally, it combines the column coordinate value and the row coordinate value to form the pixel row and column coordinates (row, col). The rounding method can be down-rounding, rounding to the nearest integer, or rounding to the nearest pixel. To avoid boundary position out-of-bounds errors, the system can also limit col to the range [0, W-1] and row to the range [0, H-1]. In this way, the system can determine the depth pixel position corresponding to the interactive click action coordinates in the structured depth map file, thus providing a clear index for subsequent depth value readings.

[0048] S207. Extract the point depth value of the region mapping and matching of the pixel row and column coordinates, construct and call the spatial pose fusion transformation matrix to perform spatial transformation calculation on the point depth value in combination with the exposure point position and pose data to obtain three-dimensional absolute spatial coordinates. Specifically, the system can access the corresponding pixel position or the corresponding pixel neighborhood in the structured depth map file based on the obtained pixel row and column coordinates (row, col), and read the point depth value d that is mapped and matched at that position. The point depth value can represent the distance from the target point to the exposure point along the line of sight of the panoramic acquisition device, or it can represent the radial distance of the target point relative to the radius of the spherical model. Its unit can be meters, millimeters, or depth units defined by the file metadata. When the corresponding pixel position is an invalid value, a hole value, or a noise value, the system can use methods such as neighborhood search, bilinear interpolation, median filtering, or nearest effective depth compensation to obtain the available depth value. Subsequently, the system calculates the three-dimensional relative coordinates of the target point in the local coordinate system of the exposure point based on the depth value of the point and the spherical direction information corresponding to the pixel row and column coordinates. For example, the pixel row and column coordinates can be converted into normalized texture coordinates, then converted into longitude angle θ and latitude angle φ, and a unit direction vector v = (cosφ·sinθ, sinφ, cosφ·cosθ) can be constructed. Then, the local three-dimensional coordinates P_local = d·v can be set. In another implementation, the gaze vector obtained in step S206 can be used directly as the unit direction vector, and the depth value can be multiplied by the gaze vector to obtain the local three-dimensional coordinates. The system further constructs a spatial pose fusion transformation matrix based on the exposure point position and attitude data. This data can include one or more of the following: the absolute position coordinates of the acquisition device during panoramic image acquisition, heading angle, pitch angle, roll angle, camera mounting extrinsic parameters, and coordinate system transformation parameters. For example, a rotation matrix R can be constructed based on the heading angle, pitch angle, and roll angle, and a translation vector T can be constructed based on the absolute position of the exposure point. These two can be combined into a homogeneous transformation matrix M = [RT; 0001]. If necessary, it can also be multiplied with the camera extrinsic parameter matrix, the device coordinate system to engineering coordinate system transformation matrix, and the geographic coordinate projection transformation matrix to obtain the spatial pose fusion transformation matrix. Finally, the system calls this spatial pose fusion transformation matrix to perform a spatial transformation on the local 3D coordinates, for example, using P_world = M·[P_local, 1]. T The system obtains the three-dimensional absolute spatial coordinates of the target point in the engineering coordinate system, map coordinate system, geographic coordinate system, or other preset absolute coordinate system. Through the above processing, the system can convert the user's clicked target in the panoramic image from a two-dimensional interactive position into a three-dimensional coordinate point with real spatial meaning, thereby supporting subsequent applications such as spatial measurement, point labeling, facility positioning, or 3D modeling.

[0049] Optionally, the step of extracting the point depth value of the region mapped and matched by the pixel row and column coordinates, constructing and calling the spatial pose fusion transformation matrix to perform spatial transformation calculation on the point depth value in combination with the exposure point position and attitude data to obtain the three-dimensional absolute spatial coordinates includes: retrieving the corresponding storage location in the structured depth map file according to the pixel row and column coordinates, reading the depth value recorded at the storage location, and confirming the depth value as the point depth value; extracting the attitude angle parameter in the exposure point position and attitude data, extracting the device error compensation parameter from the preset device error compensation parameter table, and constructing the spatial pose fusion transformation matrix according to the attitude angle parameter and the device error compensation parameter; combining the pixel row and column coordinates with the point depth value to construct an intermediate spatial vector based on the camera coordinate system corresponding to the exposure point position and attitude data; inputting the intermediate spatial vector into the spatial pose fusion transformation matrix to perform a rotation transformation operation to obtain relative coordinates, extracting the exposure point world coordinates in the exposure point position and attitude data, and performing a translation superposition operation on the exposure point world coordinates and the relative coordinates to obtain the three-dimensional absolute spatial coordinates.

[0050] Specifically, after obtaining the pixel row and column coordinates (row, col), the system can calculate the storage index of the depth data based on the total number of pixels W in the width direction, the total number of pixels H in the height direction, the data length of a single depth value, and the starting offset address of the depth data in the structured depth map file. For example, for a depth map stored in row-major order, the one-dimensional index can be determined by index = row × W + col, and the corresponding storage location can be determined by offset = baseOffset + index × byteLength. Subsequently, the system reads the depth value from the storage location and decodes it according to the scaling factor, unit identifier, or encoding format in the file metadata to obtain the actual distance value. If the depth value at the storage location is a null value, a zero value, an abnormally large value, or is marked as invalid, the system can search for the nearest valid depth value in its neighborhood, or obtain a replacement depth value by means of neighborhood mean, bilinear interpolation, median filtering, etc. The neighborhood range, invalid value threshold, and interpolation method can be configured according to actual application requirements, and this application does not limit them.

[0051] Furthermore, the attitude angle parameters may include one or more of the following: heading angle, pitch angle, and roll angle. The equipment error compensation parameters may include camera mounting angle, lens center offset, extrinsic parameter deviation between the IMU and the camera, calibration rotation compensation, or scale compensation. The system can construct a basic rotation matrix based on the attitude angle parameters, for example, constructing rotation matrices around the three coordinate axes separately and multiplying them in a preset order to obtain the attitude rotation matrix. Simultaneously, an error compensation matrix is ​​constructed based on the equipment error compensation parameters, and the attitude rotation matrix and the error compensation matrix are fused, for example, by multiplying matrices to obtain a spatial pose fusion transformation matrix. As an optional solution, the equipment error compensation parameters can be read from the equipment factory calibration table, field calibration file, or runtime calibration results; the attitude angles can also be represented using Euler angles, quaternions, or direction cosine matrices, as long as an equivalent spatial rotation transformation can be formed.

[0052] Furthermore, the system can first convert the pixel row and column coordinates into normalized texture coordinates or spherical orientation angles. For example, it can calculate s=col / W and t=row / H based on the width W and height H of the structured depth map file, then convert s and t into longitude angles θ and latitude angles φ, and construct a unit direction vector in the camera coordinate system based on the longitude and latitude angles. Subsequently, this unit direction vector is multiplied by the point depth value d to obtain the intermediate space vector P_cam. For example, in one implementation, the intermediate space vector can be represented as P_cam=d·(cosφ·sinθ, sinφ, cosφ·cosθ); in another implementation, the gaze vector obtained by back-projecting the coordinates of the interactive click action can be directly reused, and P_cam=d·v can be set. Here, the coordinate axis direction, the zero point position of the angle, and whether the depth value represents radial distance or gaze distance can be configured according to the generation rules of the panoramic depth map and the definition of the camera coordinate system.

[0053] Furthermore, the system can multiply the intermediate spatial vector P_cam with the rotation component R_fuse in the spatial pose fusion transformation matrix to obtain the world orientation relative coordinates of the target point with respect to the exposure point, P_rel = R_fuse × P_cam. Then, it reads the world coordinates T = (X0, Y0, Z0) of the exposure point from the exposure point position and pose data and executes P_world = T + P_rel to obtain the three-dimensional absolute spatial coordinates (X, Y, Z) of the target point. In another equivalent implementation, the system can also combine the rotation matrix and translation vector into a four-dimensional homogeneous transformation matrix, and then use [P_world, 1] to obtain the world orientation relative coordinates of the target point. T =[R_fuse T;0 0 0 1]×[P_cam,1] T Rotation and translation calculations are completed in one step. This method reliably converts pixel depths in the depth map into real spatial coordinates, improving the accuracy of panoramic interactive point selection and positioning, as well as spatial measurement. Optionally, the step of extracting the attitude angle parameters from the exposure point position attitude data, extracting the device error compensation parameters from a preset device error compensation parameter table, and constructing the spatial pose fusion transformation matrix based on the attitude angle parameters and the device error compensation parameters includes: extracting the attitude angle parameters from the exposure point position attitude data, constructing multiple axial rotation matrices according to a preset rotation order, multiplying the multiple axial rotation matrices sequentially to obtain a global attitude rotation matrix; extracting lens deformation correction parameters and sensor jitter correction parameters from the device error compensation parameter table as the device error compensation parameters, constructing a local error rotation matrix based on the lens deformation correction parameters and sensor jitter correction parameters; and multiplying the local error rotation matrix with the global attitude rotation matrix to obtain the spatial pose fusion transformation matrix.

[0054] Specifically, the attitude angle parameters may include yaw, pitch, and roll. The system can construct rotation matrices around the vertical axis, lateral axis, and forward axis, respectively, for example, Rz(yaw), Ry(pitch), and Rx(roll). Then, matrix multiplication is performed according to a preset rotation order, for example, R_global = Rz(yaw) × Ry(pitch) × Rx(roll), to obtain a global attitude rotation matrix characterizing the overall attitude of the exposure point acquisition device in the world coordinate system. The preset rotation order can be determined based on the device coordinate system definition, inertial navigation output conventions, or camera extrinsic parameter calibration rules. It can also use XYZ, ZYX, or other equivalent rotation orders, as long as they are consistent with the definition of the attitude angle parameters.

[0055] Furthermore, the equipment error compensation parameter table can be a factory calibration table, a field calibration table, or a dynamically updated compensation parameter table during operation. Lens deformation correction parameters can include lens optical axis deflection angle, lens mounting angle, and directional offset after distortion compensation. Sensor jitter correction parameters can include gyroscope drift compensation angle, vibration compensation angle at the moment of acquisition, or small attitude correction amounts output by the stabilization algorithm. The system can convert these correction parameters into small rotation angles around one or more coordinate axes and construct corresponding correction rotation matrices, such as constructing a lens deformation correction matrix R_lens and a sensor jitter correction matrix R_shake. Then, the local error rotation matrix is ​​obtained through R_local = R_lens × R_shake or R_local = R_shake × R_lens. The units of the above parameters can be angles or radians, which can be uniformly converted before calculation. Their specific values, update cycles, and compensation order can be configured according to the equipment calibration results and application accuracy requirements.

[0056] Furthermore, if the local error correction is defined in the camera coordinate system or the device local coordinate system, the system can use R_fuse = R_global × R_local to first correct the viewing direction in the local coordinate system and then transform it to the world coordinate system. If the local error correction is defined in the world coordinate system or the global coordinate system, R_fuse = R_local × R_global can also be used. This application does not limit this, as long as it is consistent with the calibration coordinate system of the error compensation parameters. The spatial pose fusion transformation matrix can contain only the rotation part, or it can be further combined with the translation vector corresponding to the world coordinates of the exposure point to form a homogeneous transformation matrix. By fusing the global pose with the local device error compensation, the influence of lens installation deviation, imaging distortion, and sensor jitter on the three-dimensional coordinate calculation results can be reduced, and the accuracy and stability of the three-dimensional absolute spatial coordinates obtained from the depth point conversion can be improved.

[0057] S208. According to the preset measurement instructions, capture the three-dimensional absolute spatial coordinates and perform geometric spatial measurement to obtain facility size measurement data. Extract the preset railway vector layer data, overlay the railway vector layer data and the facility size measurement data on the panoramic image data and perform joint rendering. Output the railway facility spatial linkage visualization screen in the interactive window.

[0058] Specifically, the preset measurement instructions may include instructions such as point coordinate query, two-point distance measurement, multi-point polyline length measurement, facility height measurement, trackside equipment width measurement, area measurement, or offset measurement from the railway centerline. The system can determine the current measurement type based on the user's click, selection, drag, or menu selection operations in the interactive window, and capture one or more sets of three-dimensional absolute spatial coordinates corresponding to the measurement type. For example, in two-point distance measurement, the system captures the three-dimensional absolute spatial coordinates P1 (X1, Y1, Z1) and P2 (X2, Y2, Z2) of two target points, and calculates the facility spacing using the Euclidean distance formula. In height measurement, the system can capture the three-dimensional absolute spatial coordinates of the bottom point and the top point of the facility, and calculate the facility height based on the elevation difference or spatial distance between the two. The geometric spatial measurement calculation can be completed by the local rendering end, the server-side spatial calculation module, or a combination of both. During the calculation process, coordinate unification, unit conversion, projected coordinate transformation, outlier removal, and result rounding can be performed as needed. Relevant thresholds, measurement accuracy, display units, and decimal places can be configured according to the needs of railway inspection, facility documentation, or engineering surveying. Subsequently, the system extracts railway vector layer data from a preset data source. The railway vector layer data may include vector elements of the line centerline, track edge, mileage markers, contact wire poles, signals, turnouts, platform boundaries, clearance outlines, or other railway facilities. Its data format can be GeoJSON, Shapefile, PostGIS spatial table, vector tiles, or custom binary vector files. The system converts the railway vector layer data to a spatial reference system consistent with the panoramic image data and three-dimensional absolute spatial coordinates, and projects the vector elements and facility size measurement data onto the screen coordinate positions corresponding to the panoramic image data based on the exposure point position and attitude data, camera model, and interactive window projection parameters. Finally, in the rendering layer, the system uses the panoramic image data as the base map, overlays the railway vector layer data in the form of lines, points, surfaces, or symbols, and overlays the facility size measurement data in the form of dimension lines, numerical labels, auxiliary vertical lines, semi-transparent areas, or measurement result panels. This joint rendering can be implemented using Canvas, WebGL, OpenGL, a GIS rendering engine, or graphic overlay layers. It can also update the overlay position and display ratio in real time based on the user's viewpoint rotation, scaling, translation, or switching of exposure points, thereby outputting a spatially linked and visualized view of railway facilities within the interactive window. Through this step, the system can spatially and consistently integrate and display the 3D coordinate measurement results, railway vector data, and panoramic imagery, allowing users to intuitively view the location, size, and line relationships of facilities within a panoramic scene, improving the efficiency of railway facility inspection, verification, and spatial management.

[0059] Optionally, the step of capturing the three-dimensional absolute space coordinates according to the preset measurement instructions and performing geometric space measurement to obtain facility size measurement data includes: parsing the measurement type parameter in the measurement instructions; matching the corresponding geometric calculation rule in the preset calculation rule library according to the measurement type parameter; determining the number of the three-dimensional absolute space coordinates to be captured according to the geometric calculation rule; sequentially capturing the corresponding number of the three-dimensional absolute space coordinates to form a measurement endpoint set; substituting the measurement endpoint set into the geometric calculation rule to perform geometric space measurement; associating and encapsulating the measurement result with the measurement type parameter to obtain the facility size measurement data.

[0060] Specifically, the measurement commands can be triggered by buttons, menus, shortcut operations, or external interfaces in the interactive window. Measurement type parameters can include point measurement, two-point distance measurement, height measurement, polyline length measurement, area measurement, slope measurement, and offset measurement. Upon receiving a measurement command, the system can read its type identifier, unit identifier, accuracy requirement, and coordinate source identifier, and search for the corresponding geometric calculation rule in the calculation rule library. For example, two-point distance measurement corresponds to a three-dimensional Euclidean distance calculation rule, height measurement corresponds to an elevation difference calculation rule, polyline length measurement corresponds to an adjacent endpoint distance accumulation rule, and area measurement corresponds to a spatial polygon area calculation rule. The calculation rule library can be stored in the form of configuration files, database tables, script rules, or built-in function mapping tables, and its rule content can be expanded according to the type of railway facility and measurement business requirements.

[0061] Furthermore, if the geometric calculation rule is a two-point distance or height measurement rule, the system determines that two three-dimensional absolute spatial coordinates need to be captured; if it is a polyline length measurement rule, the system can capture no less than two three-dimensional absolute spatial coordinates based on the number of endpoints clicked consecutively by the user; if it is an area measurement rule, the system can capture at least three three-dimensional absolute spatial coordinates for closing polygons. The three-dimensional absolute spatial coordinates can originate from the user's point selection results in the panoramic image, the automatic recognition results after selecting facility boundaries, or externally input facility location data. The system can write the captured coordinates into the measurement endpoint set according to the triggering order of the measurement command, the user's point selection order, or the endpoint role order specified by the rule base, for example, denoted as {P1, P2, ..., Pn}; when the number of coordinates is insufficient, the coordinates are invalid, or exceed the preset error threshold, the system can prompt for reselection, automatically remove abnormal points, or call the neighborhood depth value to recalculate the coordinates. The specific processing method can be configured according to actual application requirements.

[0062] Furthermore, for two-point distance measurement, the system can substitute P1(X1, Y1, Z1) and P2(X2, Y2, Z2) into the Euclidean distance formula to calculate the spatial distance; for height measurement, the system can calculate the elevation difference between the two endpoints or the projected distance along the vertical direction; for polygon length measurement, the system can accumulate the distances between adjacent endpoints; for area measurement, the system can first project the set of endpoints onto a suitable local plane, and then calculate the polygon area. After the calculation is completed, the system can associate and encapsulate the calculation results with information such as measurement type parameters, endpoint coordinates, coordinate units, result units, accuracy level, timestamps, and facility identifiers to form structured facility size measurement data. The encapsulation format can be a JSON object, a database record, a binary message, or a data object recognizable by the rendering engine; this application does not limit this. In this way, the calculation rules and the number of endpoints can be automatically selected according to different measurement types, so that the measurement of distance, height, area, and other dimensions of railway facilities has a unified data structure and a reusable processing flow.

[0063] Optionally, the step of substituting the set of measurement endpoints into the geometric calculation rules to perform geometric spatial measurement, and associating and encapsulating the measurement results with the measurement type parameters to obtain the facility size measurement data includes: extracting the horizontal and vertical coordinate components of each of the three-dimensional absolute spatial coordinates in the set of measurement endpoints; performing planar projection distance calculation on the horizontal coordinate components according to the geometric calculation rules to obtain a horizontal measurement result; performing elevation difference calculation on the vertical coordinate components according to the geometric calculation rules to obtain an elevation measurement result; and associating the horizontal measurement result, the elevation measurement result, and the measurement type parameters with the data according to a preset data encapsulation format to obtain the facility size measurement data.

[0064] Specifically, three-dimensional absolute spatial coordinates are typically composed of values ​​in three dimensions, and the system decouples these components according to the definition rules of the current coordinate system. As a first implementation, if the three-dimensional absolute spatial coordinates use a spatial rectangular coordinate system or a projected plane coordinate system (such as the Gauss-Kruger projection coordinate system), the system can directly extract the X-axis and Y-axis coordinate values ​​as horizontal coordinate components and the Z-axis elevation value as vertical coordinate components. As a second implementation, if the three-dimensional absolute spatial coordinates use a geodetic coordinate system (such as WGS84 latitude and longitude coordinates), the system can first calculate or project its latitude and longitude into a planar grid coordinate system as horizontal coordinate components and extract its geodetic height or elevation as vertical coordinate components. This extraction process can be implemented by a corresponding coordinate analysis script or a built-in matrix splitting function, providing basic data for subsequent independent calculations of the spatial planar and elevation relationships of railway facilities.

[0065] Furthermore, regarding the calculation of the horizontal component, the first implementation scheme is for the system to use the two-dimensional Euclidean distance formula to calculate the straight-line distance between two points on the horizontal projection plane; the second implementation scheme, for polyline or area measurements, is for the system to accumulate the horizontal projection distances of adjacent points segment by segment or substitute them into the polygon area formula for calculation; the third implementation scheme considers the influence of the Earth's curvature and uses a great circle route model or polar projection model to calculate the geodesic horizontal distance. Regarding the calculation of the vertical component, the first implementation scheme is for the system to directly calculate the absolute difference of the vertical coordinate components (i.e., Z values) of two endpoints as the relative elevation difference; the second implementation scheme, for multiple endpoints, can calculate the maximum elevation difference between the highest and lowest points, or calculate the relative elevation difference of each point relative to a preset reference surface (such as a reference rail top elevation). Parameters involved in the calculation process, such as unit unification, significant figure truncation, or extreme value elimination, can be flexibly configured according to railway engineering surveying specifications.

[0066] This embodiment discloses a railway facility visualization browsing and publishing system. Figure 3 This is a schematic diagram of a module of a railway facility visualization browsing and publishing system disclosed in an embodiment of this application, such as... Figure 3 As shown, the system includes the following modules: The data receiving module 301 is configured to receive multi-source spatial initial data of railway facilities and extract laser point cloud data, panoramic image data and exposure point position and attitude data associated with the panoramic image data contained in the multi-source spatial initial data. The depth map generation module 302 is configured to map the laser point cloud data onto a planar mesh model that matches the exposure point position and attitude data based on a preset laser angle resolution, thereby generating a structured depth map file. The index association module 303 is configured to extract the common image code between the panoramic image data and the structured depth map file, and establish an index association relationship between the panoramic image data and the structured depth map file based on the common image code; The data loading module 304 is configured to load the panoramic image data and the structured depth map file in the interactive window of the preset display control interface according to the index association relationship. Interactive acquisition module 305 is configured to acquire the coordinates of interactive click actions targeting the panoramic image data within the interactive window; The coordinate transformation module 306 is configured to extract the spherical texture coordinates mapped by the interactive click action onto a preset panoramic spherical model, and convert the spherical texture coordinates into pixel row and column coordinates corresponding to the structured depth map file based on a preset spherical texture mapping conversion formula. The spatial coordinate calculation module 307 is configured to extract the point depth value of the region mapped and matched by the pixel row and column coordinates, construct and call the spatial pose fusion transformation matrix to perform spatial transformation calculation on the point depth value in combination with the exposure point position and pose data, and obtain three-dimensional absolute spatial coordinates; The measurement and rendering module 308 is configured to capture the three-dimensional absolute spatial coordinates according to the preset measurement instructions, perform geometric spatial measurement calculations, obtain facility size measurement data, extract preset railway vector layer data, overlay the railway vector layer data and the facility size measurement data on the panoramic image data to perform joint rendering, and output a railway facility spatial linkage visualization screen in the interactive window.

[0067] The depth map generation module 302 is further configured to: take the origin of the spatial coordinates in the exposure point position and attitude data as a reference point, calculate the straight-line spatial distance from each laser reflection point in the laser point cloud data to the reference point point, and obtain the initial depth value corresponding to each laser reflection point; extract the horizontal angular resolution parameter and the vertical angular resolution parameter in the laser angular resolution, map each laser reflection point to the column position of the planar mesh model according to the horizontal angular resolution parameter, and map each laser reflection point to the row position of the planar mesh model according to the vertical angular resolution parameter, and obtain the row index value and column index value corresponding to each laser reflection point; and write the row index value, the column index value and the corresponding initial depth value into the structured depth map file according to a preset depth matrix format.

[0068] The coordinate transformation module 306 is further configured to: use the center of the panoramic spherical model as a virtual observation base point, construct a gaze vector pointing from the virtual observation base point to the coordinates of the interactive click action based on the spherical texture mapping conversion formula, calculate the intersection of the gaze vector and the inner surface of the panoramic spherical model as the target projection point; read the normalized horizontal texture component and normalized vertical texture component of the target projection point in the preset panoramic spherical model texture coordinate system, and combine the normalized horizontal texture component and the normalized vertical texture component to establish the spherical texture coordinates; extract the total number of pixels in the width direction and the total number of pixels in the height direction of the structured depth map file, multiply the normalized horizontal texture component in the spherical texture coordinates by the total number of pixels in the width direction and round to obtain the column coordinate value, multiply the normalized vertical texture component in the spherical texture coordinates by the total number of pixels in the height direction and round to obtain the row coordinate value, and combine the column coordinate value and the row coordinate value to obtain the pixel row and column coordinates.

[0069] The spatial coordinate calculation module 307 is further configured to: retrieve the corresponding storage location in the structured depth map file according to the pixel row and column coordinates, read the depth value recorded at the storage location, and confirm the depth value as the point depth value; extract the attitude angle parameters from the exposure point position attitude data, extract the device error compensation parameters from the preset device error compensation parameter table, and construct the spatial pose fusion transformation matrix according to the attitude angle parameters and the device error compensation parameters; combine the pixel row and column coordinates with the point depth value to construct an intermediate spatial vector based on the camera coordinate system corresponding to the exposure point position attitude data; input the intermediate spatial vector into the spatial pose fusion transformation matrix to perform a rotation transformation operation to obtain relative coordinates, extract the exposure point world coordinates from the exposure point position attitude data, and perform a translation superposition operation on the exposure point world coordinates and the relative coordinates to obtain the three-dimensional absolute spatial coordinates.

[0070] The spatial coordinate calculation module 307 is further configured to: extract the attitude angle parameters from the exposure point position attitude data, construct multiple axial rotation matrices according to a preset rotation order, multiply the multiple axial rotation matrices sequentially to obtain a global attitude rotation matrix; extract lens deformation correction parameters and sensor jitter correction parameters from the device error compensation parameter table as the device error compensation parameters, construct a local error rotation matrix based on the lens deformation correction parameters and sensor jitter correction parameters; and multiply the local error rotation matrix with the global attitude rotation matrix to obtain the spatial pose fusion transformation matrix.

[0071] The measurement rendering module 308 is further configured to: parse the measurement type parameter in the measurement instruction; match the corresponding geometric calculation rule in the preset calculation rule library according to the measurement type parameter; determine the number of the three-dimensional absolute space coordinates to be captured according to the geometric calculation rule; capture the corresponding number of the three-dimensional absolute space coordinates in sequence to form a measurement endpoint set; substitute the measurement endpoint set into the geometric calculation rule to perform geometric space measurement; and associate and encapsulate the measurement result with the measurement type parameter to obtain the facility size measurement data.

[0072] The measurement and rendering module 308 is further configured to: extract the horizontal and vertical coordinate components of each of the three-dimensional absolute spatial coordinates in the measurement endpoint set; perform planar projection distance calculation on the horizontal coordinate components according to the geometric calculation rules to obtain a horizontal measurement result; perform elevation difference calculation on the vertical coordinate components according to the geometric calculation rules to obtain an elevation measurement result; and associate the horizontal measurement result, the elevation measurement result, and the measurement type parameter with a preset data encapsulation format to obtain the facility size measurement data.

[0073] This embodiment also discloses an electronic device, as shown in the reference. Figure 4 The electronic device may include: at least one processor 401, at least one communication bus 402, a user interface 403, a network interface 404, and at least one memory 405. The communication bus 402 is used to enable communication between these components. The user interface 403 may include a display screen or a camera; optionally, the user interface 403 may also include a standard wired interface or a wireless interface. The network interface 404 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0074] The processor 401 may include one or more processing cores. The processor 401 connects to various parts of the server using various interfaces and lines, and performs various server functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in memory 405, and by calling data stored in memory 405. Optionally, the processor 401 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 401 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required for display; and the modem handles wireless communication. It is understood that the modem may also be implemented as a separate chip without being integrated into the processor 401.

[0075] The memory 405 may include random access memory (RAM) or read-only memory. Optionally, the memory 405 may include a non-transitory computer-readable storage medium. The memory 405 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 405 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 405 may also be at least one storage device located remotely from the aforementioned processor 401. Figure 4 As shown, the memory 405, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an application for a method of visually browsing and publishing railway facilities.

[0076] exist Figure 4 In the electronic device shown, the user interface 403 is mainly used to provide an input interface for the user and to obtain the user input data; while the processor 401 can be used to call an application for visual browsing of railway facilities stored in the memory 405. When executed by one or more processors 401, the electronic device performs one or more methods as described in the above embodiments.

[0077] Furthermore, the functional units in the various embodiments of this application 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. 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 device (CMD). Based on this understanding, the technical solution of this application, 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 memory 405 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 in the various embodiments of this application. The aforementioned memory 405 includes various media capable of storing program code, such as a USB flash drive, portable hard drive, magnetic disk, or optical disk.

[0078] The foregoing description is merely an exemplary embodiment of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Other embodiments of this disclosure will be readily apparent to those skilled in the art upon consideration of the disclosure in this specification. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described in this disclosure. The specification and embodiments are to be considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

Claims

1. A method for visually browsing and publishing railway facilities, characterized in that, The method includes: Receive multi-source initial spatial data of railway facilities, and extract laser point cloud data, panoramic image data, and exposure point position and attitude data associated with the panoramic image data contained in the multi-source initial spatial data; Based on the preset laser angle resolution, the laser point cloud data is mapped onto a planar mesh model that matches the position and attitude data of the exposure point, thereby generating a structured depth map file; Extract the common image code between the panoramic image data and the structured depth map file, and establish an index association relationship between the panoramic image data and the structured depth map file based on the common image code; Based on the index association, the panoramic image data and the structured depth map file are loaded into the interactive window of the preset display control interface; Collect the coordinates of interactive click actions targeting the panoramic image data within the interactive window; Extract the spherical texture coordinates of the interactive click action coordinates onto the preset panoramic spherical model, and convert the spherical texture coordinates into the pixel row and column coordinates corresponding to the structured depth map file based on the preset spherical texture mapping conversion formula; Extract the depth value of the point mapped to the region pointed to by the pixel row and column coordinates, construct and call the spatial pose fusion transformation matrix to perform spatial transformation calculation on the depth value of the point in combination with the exposure point position and pose data, and obtain the three-dimensional absolute spatial coordinates; According to the preset measurement instructions, the three-dimensional absolute spatial coordinates are captured and geometric spatial measurement is performed to obtain facility size measurement data. The preset railway vector layer data is extracted, and the railway vector layer data and the facility size measurement data are superimposed on the panoramic image data to perform joint rendering. The spatial linkage visualization of railway facilities is output in the interactive window.

2. The method according to claim 1, characterized in that, The process of mapping the laser point cloud data onto a planar mesh model matching the exposure point position and pose data, based on a preset laser angle resolution, generates a structured depth map file, including: Using the origin of the spatial coordinates in the exposure point position and attitude data as the reference point, the straight-line spatial distance from each laser reflection point in the laser point cloud data to the reference point is calculated point by point to obtain the initial depth value corresponding to each laser reflection point. Extract the horizontal and vertical angular resolution parameters from the laser angular resolution. Based on the horizontal angular resolution parameters, map each laser reflection point to the column position of the planar mesh model. Based on the vertical angular resolution parameters, map each laser reflection point to the row position of the planar mesh model to obtain the row index value and column index value corresponding to each laser reflection point. The row index value, the column index value, and the corresponding initial depth value are written into the structured depth map file according to a preset depth matrix format.

3. The method according to claim 1, characterized in that, The step of extracting the spherical texture coordinates mapped from the interactive click action coordinates onto a preset panoramic spherical model, and converting the spherical texture coordinates into pixel row and column coordinates corresponding to the structured depth map file based on a preset spherical texture mapping conversion formula, includes: Using the center of the panoramic spherical model as a virtual observation base point, and based on the spherical texture mapping conversion formula, a gaze vector is constructed pointing from the virtual observation base point to the coordinates of the interactive click action. The intersection of the gaze vector and the inner surface of the panoramic spherical model is calculated as the target projection point. Read the normalized horizontal texture component and normalized vertical texture component of the target projection point in the preset panoramic spherical model texture coordinate system, and combine the normalized horizontal texture component and the normalized vertical texture component to establish the spherical texture coordinates. Extract the total number of pixels in the width direction and the total number of pixels in the height direction from the structured depth map file. Multiply the normalized horizontal texture component in the spherical texture coordinates by the total number of pixels in the width direction and round it to obtain the column coordinate value. Multiply the normalized vertical texture component in the spherical texture coordinates by the total number of pixels in the height direction and round it to obtain the row coordinate value. Combine the column coordinate value and the row coordinate value to obtain the pixel row and column coordinates.

4. The method according to claim 1, characterized in that, The step involves extracting the depth values ​​of the pixel row and column coordinates pointing to the mapped region, constructing and calling a spatial pose fusion transformation matrix, and combining the exposure point position and pose data to perform a spatial transformation calculation on the depth values ​​to obtain three-dimensional absolute spatial coordinates, including: Based on the pixel row and column coordinates, the corresponding storage location is retrieved in the structured depth map file, the depth value recorded at the storage location is read, and the depth value is confirmed as the depth value of the point. Extract the attitude angle parameters from the exposure point position attitude data, extract the device error compensation parameters from the preset device error compensation parameter table, and construct the spatial pose fusion transformation matrix based on the attitude angle parameters and the device error compensation parameters; By combining the pixel row and column coordinates with the point depth value, an intermediate space vector is constructed based on the camera coordinate system corresponding to the exposure point position and pose data. The intermediate spatial vector is input into the spatial pose fusion transformation matrix to perform a rotation transformation operation to obtain the relative coordinates. The exposure point world coordinates are extracted from the exposure point position and pose data. The exposure point world coordinates and the relative coordinates are then translated and superimposed to obtain the three-dimensional absolute spatial coordinates.

5. The method according to claim 4, characterized in that, The step of extracting the attitude angle parameters from the exposure point position attitude data, extracting the device error compensation parameters from a preset device error compensation parameter table, and constructing the spatial pose fusion transformation matrix based on the attitude angle parameters and the device error compensation parameters includes: Extract the attitude angle parameters from the exposure point position attitude data, construct multiple axial rotation matrices according to a preset rotation order, and multiply the multiple axial rotation matrices in sequence to obtain the global attitude rotation matrix. Lens deformation correction parameters and sensor shake correction parameters are extracted from the equipment error compensation parameter table as the equipment error compensation parameters, and a local error rotation matrix is ​​constructed based on the lens deformation correction parameters and sensor shake correction parameters; Multiplying the local error rotation matrix by the global attitude rotation matrix yields the spatial pose fusion transformation matrix.

6. The method according to claim 1, characterized in that, The process of capturing the three-dimensional absolute spatial coordinates according to preset measurement instructions and performing geometric spatial measurement to obtain facility size measurement data includes: The measurement type parameter in the measurement instruction is parsed, and the corresponding geometric calculation rule is matched in the preset measurement rule library according to the measurement type parameter; The number of three-dimensional absolute space coordinates to be captured is determined according to the geometric calculation rules, and the corresponding number of three-dimensional absolute space coordinates are captured in sequence to form a measurement endpoint set. Substitute the set of measurement endpoints into the geometric calculation rules to perform geometric spatial measurement, and encapsulate the measurement results with the measurement type parameters to obtain the facility size measurement data.

7. The method according to claim 6, characterized in that, The step of substituting the set of measurement endpoints into the geometric calculation rules to perform geometric spatial measurement, and then associating and encapsulating the measurement results with the measurement type parameters to obtain the facility size measurement data includes: Extract the horizontal and vertical coordinate components of each of the three-dimensional absolute spatial coordinates from the set of measurement endpoints; According to the geometric calculation rules, the horizontal coordinate component values ​​are subjected to planar projection distance calculation to obtain the horizontal measurement result; and the vertical coordinate component values ​​are subjected to elevation difference calculation to obtain the elevation measurement result. The horizontal measurement results, the elevation measurement results, and the measurement type parameters are associated with each other according to a preset data encapsulation format to obtain the facility size measurement data.

8. A railway facility visualization browsing and publishing system, characterized in that, The system includes: The data receiving module is configured to receive multi-source initial spatial data of railway facilities and extract laser point cloud data, panoramic image data and exposure point position and attitude data associated with the panoramic image data contained in the multi-source initial spatial data. The depth map generation module is configured to map the laser point cloud data onto a planar mesh model that matches the position and attitude data of the exposure point based on a preset laser angle resolution, thereby generating a structured depth map file. The index association module is configured to extract the common image code between the panoramic image data and the structured depth map file, and establish an index association relationship between the panoramic image data and the structured depth map file based on the common image code; The data loading module is configured to load the panoramic image data and the structured depth map file in the interactive window of the preset display control interface according to the index association relationship. The interactive acquisition module is configured to acquire the coordinates of interactive click actions in response to the panoramic image data within the interactive window; The coordinate transformation module is configured to extract the spherical texture coordinates mapped from the interactive click action coordinates onto a preset panoramic spherical model, and convert the spherical texture coordinates into pixel row and column coordinates corresponding to the structured depth map file based on a preset spherical texture mapping conversion formula. The spatial coordinate calculation module is configured to extract the point depth value of the region mapped and matched by the pixel row and column coordinates, construct and call the spatial pose fusion transformation matrix to perform spatial transformation calculation on the point depth value in combination with the exposure point position and pose data, and obtain three-dimensional absolute spatial coordinates; The measurement and rendering module is configured to capture the three-dimensional absolute spatial coordinates according to the preset measurement instructions, perform geometric spatial measurement calculations to obtain facility size measurement data, extract preset railway vector layer data, overlay the railway vector layer data and the facility size measurement data on the panoramic image data to perform joint rendering, and output a spatial linkage visualization of railway facilities in the interactive window.

9. An electronic device, characterized in that, The device includes a processor, a memory, a user interface, and a network interface. The memory is used to store instructions. The user interface and the network interface are both used to communicate with other devices. The processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed, perform the method as described in any one of claims 1-7.

Images