Power transmission line gun monitoring method, system and device fusing laser radar and storage medium

Abstract

The application discloses a power transmission line gun mechanism monitoring method, system and device fusing a laser radar and a storage medium, relates to the technical field of risk monitoring, and distinguishes measured support sections and fitted compensation sections according to the real hitting conditions of point clouds on an initial conductor space model by spatially fusing gun mechanism image data and laser radar point cloud data, further determines the neighborhood extension trend of the fitted compensation section by using adjacent measured support sections, and performs consistency checking, so as to identify suspicious conductor sections that may be covered by smooth fitting; then, a local risk envelope region is generated based on the sections, and a risk result is determined in combination with the spatial position of a channel target, so that the monitoring no longer depends on a single fitted conductor position, can include local sagging, wind deviation, ice sinking and other abnormalities that may exist in a point cloud missing area in judgment, and reduces the risk of overestimating a safety distance and missing hidden dangers.

Description

Technical Field

[0001] This invention relates to the field of risk monitoring technology, specifically to a method, system, device, and storage medium for monitoring transmission line guns using integrated lidar. Background Technology

[0002] The bullet camera monitoring method for power transmission lines, which integrates lidar, typically involves simultaneously deploying bullet cameras and lidar equipment on power transmission line towers, high points in corridors, or fixed monitoring supports. The bullet cameras acquire two-dimensional images of the power transmission line and corridor environment to identify targets such as conductors, towers, insulators, trees, construction machinery, and foreign objects. The lidar equipment performs a spatial scan of the same monitoring area to obtain three-dimensional point cloud information corresponding to the target objects, determining their spatial location, height, outline, and distance relationship to the power transmission line. In practice, the bullet camera and lidar equipment are first spatially calibrated to establish a mapping relationship between the two-dimensional image coordinates and the three-dimensional point cloud coordinates. Then, the image recognition results are fused with the point cloud spatial structure. This allows for the identification of target types while simultaneously calculating the actual spatial distance between the target object and the conductor. Based on distance thresholds or risk level rules, it determines whether there are safety hazards such as tree obstructions, external mechanical damage, approaching foreign objects, or insufficient corridor clearance.

[0003] However, in the aforementioned monitoring methods, lidar may not be able to collect complete and continuous point cloud data of the transmission line conductors. Because conductors typically have small diameters, long spans, unstable surface reflections, and are susceptible to wind deflection, icing, temperature, and load variations, the actual point cloud may contain only a few discrete conductor points, or even be missing in certain spans or local sections. To continue spatial distance calculations, existing methods often use pole mounting points, historical conductor morphology, or a small number of previously collected conductor points to fit and complete the conductor's spatial position. However, this fitting and completion usually tends to create a continuous, smooth conductor model, easily smoothing out real anomalies such as increased sag, local wind deflection, icing subsidence, and deformation near clamps. This results in the distances between trees, foreign objects, or construction machinery and the conductors calculated later being based on the completed model positions rather than the actual conductor positions. Consequently, this leads to problems such as overestimating safe distances, masking local hazards, and delayed risk warnings. Summary of the Invention

[0004] The purpose of this invention is to solve the problems mentioned in the background art above, and to propose a method, system, device and storage medium for monitoring transmission line guns by integrating lidar.

[0005] A first aspect of this invention provides a method for monitoring transmission line guns using fused lidar, the method comprising: S1: Acquire bullet camera image data and lidar point cloud data of the transmission line monitoring area, and perform spatial fusion on the bullet camera image data and lidar point cloud data to obtain the initial conductor spatial model and channel target spatial location results; S2: Based on the actual hit situation of lidar point cloud data on the initial traverse space model, the initial traverse space model is divided into the measured support segment and the fitting compensation segment; S3: Determine the neighborhood extension trend corresponding to the fitted compensation segment based on the measured support segments adjacent to the fitted compensation segment; S4: Perform consistency verification between the fitted compensation segment and the corresponding neighborhood extension trend to identify suspicious conductor segments in the fitted compensation segment; S5: Based on the neighborhood extension trend corresponding to the suspicious conductor segment, generate the local risk envelope region of the suspicious conductor segment; S6: Determine the risk monitoring results of transmission line channels based on the spatial relationship between the target spatial location results and the local risk envelope area.

[0006] Optionally, based on the actual hit rate of the lidar point cloud data on the initial traverse space model, the steps to divide the initial traverse space model into a measured support segment and a fitted compensation segment are as follows: According to the direction of the transmission line extension, the initial conductor space model is divided into multiple continuous conductor model segments; Extract the centerline position and endpoint position of each conductor model segment; Using the centerline position of each traverse model segment as the center, generate the point cloud hit detection range corresponding to that traverse model segment; Spatial matching is performed between the point cloud corresponding to the traverse and the point cloud hit detection range corresponding to each traverse model segment to determine the point cloud corresponding to the traverse that falls within the hit detection range of each point cloud. The point cloud hit length of each traverse model segment is determined based on the distribution length of the point cloud corresponding to the traverse that falls within the hit detection range of each point cloud. A traverse model segment whose point cloud hit length is not less than a preset ratio of the corresponding traverse model segment length is determined as the measured support segment; The traverse model segments whose point cloud hit length is less than a preset proportion of the corresponding traverse model segment length are determined as fitting compensation segments.

[0007] Optionally, the step of determining the neighborhood extension trend corresponding to the fitted compensation segment based on the measured support segments adjacent to the fitted compensation segment is as follows: Based on the division results of the measured support section and the fitted compensation section, determine the front adjacent measured support section and the rear adjacent measured support section of each fitted compensation section along the extension direction of the transmission line. Extract the spatial center point sequence of the point cloud corresponding to the traverse in the adjacent front and rear measured support segments respectively; According to the extension direction of the transmission line, the spatial center point sequence of adjacent measured support sections on the front side is sequentially connected to obtain the front measured extension line. According to the extension direction of the transmission line, the spatial center point sequence of the adjacent measured support section on the rear side is sequentially connected to obtain the measured extension line on the rear side. The direction of extension of the front neighborhood is determined based on the height change direction and lateral offset direction of the front measured extension line near the fitting compensation section. The extension direction of the rear neighborhood is determined based on the height change direction and lateral offset direction of the measured extension line near the fitting compensation segment. The forward and backward neighborhood extension directions are used together as the neighborhood extension trend corresponding to the fitting compensation segment.

[0008] Optionally, the steps for verifying the consistency between the fitted compensation segment and the corresponding neighborhood extension trend, and determining the suspicious conductor segments in the fitted compensation segment, are as follows: The fit compensation segment is checked for consistency with the corresponding neighborhood extension trend. The neighborhood squeeze fit index and the trend release continuity index are calculated. The neighborhood squeeze fit index and the trend release continuity index are added together to obtain the consistency index. Suspicious conductor segments in the fit compensation segment are determined based on the consistency index and the preset consistency index threshold.

[0009] Optionally, the calculation steps for the neighborhood squeeze coincidence index are as follows: Along the extension direction of the fitting compensation segment, the fitting compensation segment is divided into multiple equally ordered positions, and the fitting compensation segment cross-section points corresponding to each equally ordered position are extracted. Based on the extension direction of the front neighborhood, extend from the end of the front adjacent measured support segment close to the fitting compensation segment into the interior of the fitting compensation segment to obtain the front clamping point corresponding to each equidistant position. Based on the extension direction of the rear neighborhood, extend backward from the end of the adjacent measured support segment close to the fitting compensation segment into the interior of the fitting compensation segment to obtain the rear clamping point corresponding to each equidistant position. Calculate the spatial distance between the fitting compensation section point and the front clamping point, the spatial distance between the fitting compensation section point and the rear clamping point, and the spatial distance between the front clamping point and the rear clamping point at each of the equal order positions. For each equidistant position, twice the spatial distance between the front clamping point and the rear clamping point is divided by the sum of the spatial distance between the fitting compensation section cross-section point and the front clamping point, the spatial distance between the fitting compensation section cross-section point and the rear clamping point, and the spatial distance between the front clamping point and the rear clamping point, to obtain the cross-section clamping fit value corresponding to that equidistant position. Multiply the cross-sectional clamping and matching values ​​corresponding to all the equally ordered positions to obtain the cross-sectional clamping and matching product; The neighborhood squeezing coincidence index is obtained by taking the square root of the product of cross-sectional squeezing coincidences corresponding to the number of equidistant positions.

[0010] Optionally, the calculation steps for the trend reversal continuity index are as follows: Extract the front end change vector from the end of the front measured extension line near the fitting compensation section, and extract the rear end change vector from the end of the rear measured extension line near the fitting compensation section. Divide the change vector of the front end by the length of the change vector of the front end to obtain the unit change vector of the front end; divide the change vector of the rear end by the length of the change vector of the rear end to obtain the unit change vector of the rear end. Add the unit change vector on the front side and the unit change vector on the back side, and divide the sum by the length of the sum to obtain the main direction of the release. Multiply the length of the change vector at the front end with the length of the change vector at the rear end, and then perform a square root operation on the result of the multiplication to obtain the neighborhood reference variable. Extract the cross-sectional points of two adjacent fitting compensation segments sequentially along the extension direction of the fitting compensation segment, and calculate the spatial displacement of the cross-sectional point of the latter fitting compensation segment relative to the cross-sectional point of the former fitting compensation segment. Multiply the spatial displacement by the main direction of the elution to obtain the directional projection value of the corresponding adjacent cross-sectional interval; Add the absolute values ​​of the directional projection values ​​together and divide the sum by two to obtain the positive release variables for the corresponding adjacent cross-sectional intervals; For each adjacent cross-sectional interval, the positive explanatory variable is multiplied by the neighboring baseline explanatory variable, the result of the multiplication is squared and then multiplied by two, and the result is divided by the sum of the positive explanatory variable and the neighboring baseline explanatory variable to obtain the segmental explanatory variable transfer value of the adjacent cross-sectional interval. For any two adjacent cross-sectional intervals, multiply the positive explanatory variable of the previous adjacent cross-sectional interval with the positive explanatory variable of the next adjacent cross-sectional interval, perform a square root operation on the multiplication result, multiply by two, and divide the result by the sum of the positive explanatory variables of the previous adjacent cross-sectional interval and the next adjacent cross-sectional interval to obtain the adjacent explanatory continuous value. Multiply all the segmented emanation values ​​and all the adjacent emanation values ​​together to obtain the emanation continuous product; The trend exothermic continuous index is obtained by taking the square root of the exothermic continuous product and the total number of the segmented exothermic successor values ​​and adjacent exothermic continuous values ​​involved in the multiplication.

[0011] Optionally, the step of determining the suspicious traverse segments in the fitting compensation segment based on the consistency index and a preset consistency index threshold is as follows: Compare the consistency index with a preset consistency index threshold; When the consistency index is less than the preset consistency index threshold, the corresponding fitting compensation segment is identified as a suspicious conductor segment. When the consistency index is not less than the preset consistency index threshold, the corresponding fitting compensation segment is determined as a non-suspicious conductor segment. All suspicious conductor segments are arranged in order of their position in the initial conductor space model to obtain a set of suspicious conductor segments.

[0012] In a second aspect of this invention, a transmission line gun monitoring system integrating lidar is proposed. The system includes a data fusion unit, a data processing unit, and a data monitoring unit. Data fusion unit: By acquiring bullet camera image data and lidar point cloud data of the transmission line monitoring area, and spatially fusing the bullet camera image data and lidar point cloud data, the initial conductor spatial model and channel target spatial location results are obtained; The data fusion unit divides the initial traverse space model into a measured support segment and a fitting compensation segment based on the actual hit situation of the lidar point cloud data on the initial traverse space model. It also determines the neighborhood extension trend corresponding to the fitting compensation segment, performs consistency verification between the fitting compensation segment and the corresponding neighborhood extension trend, identifies suspicious traverse segments in the fitting compensation segment, and generates the local risk envelope region of the suspicious traverse segment based on the neighborhood extension trend corresponding to the suspicious traverse segment. Data monitoring unit: Determines the risk monitoring results of transmission line channels based on the spatial relationship between the target spatial location results and the local risk envelope area.

[0013] In a third aspect of this invention, a transmission line gun monitoring device integrating lidar is proposed, the device comprising: Fusion module: Acquires bullet camera image data and lidar point cloud data of the transmission line monitoring area, and performs spatial fusion on the bullet camera image data and lidar point cloud data to obtain the initial conductor spatial model and channel target spatial location results; Module division: Based on the actual hit situation of LiDAR point cloud data on the initial traverse space model, the initial traverse space model is divided into measured support segment and fitting compensation segment; Extension Trend Module: Based on the measured support segments adjacent to the fitted compensation segment, determine the neighborhood extension trend corresponding to the fitted compensation segment; Suspicious conductor module: Performs consistency verification between the fitted compensation segment and the corresponding neighborhood extension trend to identify suspicious conductor segments in the fitted compensation segment; Risk envelope module: Based on the neighborhood extension trend corresponding to the suspicious conductor segment, generate the local risk envelope region of the suspicious conductor segment; Risk monitoring module: Determines the risk monitoring results of transmission line channels based on the spatial relationship between the target spatial location results and the local risk envelope area.

[0014] In a fourth aspect of the present invention, a computer-readable storage medium is provided, characterized in that the computer-readable storage medium stores a computer program, which, when processed, executes the steps of any of the methods described above.

[0015] The beneficial effects of this invention are: This invention proposes a method, system, device, and storage medium for monitoring transmission lines using laser radar (LiDAR). After spatially fusing laser radar image data and LiDAR point cloud data, it first distinguishes between measured support segments and fitted compensation segments based on the actual hit situation of the LiDAR point cloud data on the initial conductor spatial model. Then, it uses the local change state of the conductor reflected by adjacent measured support segments of the fitted compensation segment to determine the neighborhood extension trend. The consistency between the fitted compensation segment and the neighborhood extension trend is verified, thereby identifying suspicious conductor segments that may be masked by smooth fitting due to missing point clouds. Furthermore, it generates a local risk envelope region based on the neighborhood extension trend corresponding to the suspicious conductor segments. By determining the spatial relationship between the target spatial location of the transmission line and the local risk envelope area, the risk assessment results of the transmission line channel are determined. This means that the risk assessment no longer relies solely on the single conductor model that has been fitted and completed. Instead, it includes potential local sag increases, local wind deflection, icing subsidence, or deformation near clamps in the missing areas of the conductor point cloud. This avoids mistaking the smoothed conductor position obtained from the fitting and compensation as the actual conductor position for distance calculation, reduces the possibility of overestimating the safe distance between the conductor and tree obstacles, foreign objects, or construction machinery, reduces the problem of local hidden dangers being masked and risk warnings being delayed, and improves the authenticity and reliability of the transmission line channel monitoring results. Attached Figure Description

[0016] Figure 1 A flowchart of a transmission line gun monitoring method using fused lidar provided in an embodiment of the present invention. Detailed Implementation

[0017] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0018] This invention provides a method for monitoring power transmission lines using laser radar. See also... Figure 1 , Figure 1 A flowchart illustrating a transmission line gun-mounted monitoring method using fused lidar provided in an embodiment of the present invention. The method includes the following steps: S1: Acquire bullet camera image data and lidar point cloud data of the transmission line monitoring area, and perform spatial fusion on the bullet camera image data and lidar point cloud data to obtain the initial conductor spatial model and channel target spatial location results; S2: Based on the actual hit situation of lidar point cloud data on the initial traverse space model, the initial traverse space model is divided into the measured support segment and the fitting compensation segment; S3: Determine the neighborhood extension trend corresponding to the fitted compensation segment based on the measured support segments adjacent to the fitted compensation segment; S4: Perform consistency verification between the fitted compensation segment and the corresponding neighborhood extension trend to identify suspicious conductor segments in the fitted compensation segment; S5: Based on the neighborhood extension trend corresponding to the suspicious conductor segment, generate the local risk envelope region of the suspicious conductor segment; S6: Determine the risk monitoring results of transmission line channels based on the spatial relationship between the target spatial location results and the local risk envelope area.

[0019] The transmission line gun monitoring method based on the fused lidar provided in this invention first distinguishes between measured support segments and fitted compensation segments based on the actual hit situation of the lidar point cloud data on the initial conductor spatial model after spatial fusion of gun image data and lidar point cloud data. Then, it determines the neighborhood extension trend by utilizing the local change state of the conductor reflected by the adjacent measured support segments of the fitted compensation segment, and verifies the consistency between the fitted compensation segment and the neighborhood extension trend. This allows for the identification of suspicious conductor segments that may be masked by smooth fitting due to missing point clouds. Furthermore, it generates a local risk envelope region based on the neighborhood extension trend corresponding to the suspicious conductor segments, and uses channels... The spatial relationship between the target spatial location result and the local risk envelope area determines the risk monitoring results of the transmission line corridor. This makes risk assessment no longer solely dependent on the single conductor model after fitting and completion. Instead, it includes potential local sag increases, local wind deflection, icing subsidence, or deformation near clamps in the missing areas of the conductor point cloud. This avoids mistaking the smoothed conductor position obtained from fitting and compensation as the actual conductor position for distance calculation, reduces the possibility of overestimating the safe distance between trees, foreign objects, or construction machinery and the conductor, reduces the problem of local hidden dangers being masked and risk warnings being delayed, and improves the authenticity and reliability of the transmission line corridor monitoring results.

[0020] In one embodiment, S1: The steps of acquiring bullet camera image data and lidar point cloud data of the transmission line monitoring area, and spatially fusing the bullet camera image data and lidar point cloud data to obtain the initial conductor spatial model and channel target spatial location results are as follows: The bullet camera and lidar equipment are installed at the same power transmission line monitoring point, and the shooting field of the bullet camera and the scanning area of ​​the lidar equipment jointly cover the target power transmission line span and its corresponding channel area. The monitoring image frames collected by the bullet camera and the monitoring point cloud frames collected by the lidar device are acquired separately, and the monitoring image frames and monitoring point cloud frames within the same acquisition time period are selected as the data to be fused according to the acquisition timestamp. Conductor identification and channel target identification are performed on the monitored image frames to obtain conductor image regions and channel target image regions. The channel targets include at least one of trees, construction machinery, foreign objects, and buildings. The monitoring point cloud frames are denoised and ground feature separated to obtain effective point cloud data corresponding to the data to be fused. Based on the pre-calibrated spatial mapping relationship between the gun camera and the lidar device, each point cloud in the effective point cloud data is projected onto the monitoring image frame to obtain the point cloud projection result. Based on the overlap between the point cloud projection results and the traverse image region, the point cloud projected onto the traverse image region is determined as the point cloud corresponding to the traverse. Based on the overlap between the point cloud projection results and the channel target image region, the point cloud projected onto the channel target image region is determined as the point cloud corresponding to the target. Based on the point cloud corresponding to the traverse and the point cloud corresponding to the target, generate the fusion correspondence result between the monitoring image frame and the monitoring point cloud frame; The spatial distribution of the conductor corresponding point cloud along the transmission line extension direction is extracted from the fusion correspondence results, and an initial conductor spatial model is generated based on the spatial distribution results; The spatial occupancy results of the target corresponding point cloud are extracted from the fusion correspondence results, and the three-dimensional coordinate range of each channel target is determined based on the spatial occupancy results to obtain the spatial position results of the channel target.

[0021] It should be noted that in step S1, the bullet camera and lidar equipment are first deployed at the same power transmission line monitoring point to ensure that both types of equipment observe the same span and the same channel area. For example, the bullet camera and lidar are installed on the same tower, so that the bullet camera's view can cover objects such as power lines, trees, and construction machinery, and the lidar's scanning area also covers the same space. Subsequently, monitoring image frames and monitoring point cloud frames are acquired separately, and data from the same acquisition time period are selected as the data to be fused according to the acquisition timestamp. This is to avoid location correspondence errors caused by images and point clouds coming from different times. Next, the monitoring image frames are used for power line identification and channel target identification, such as identifying the pixel area where the power line is located, the tree canopy area, the crane arm area, or the building boundary area in the image. At the same time, the monitoring point cloud frames are denoised and separated from ground objects to remove irrelevant point clouds such as raindrops, birds, and scattered ground points. Valid point cloud data that can be used for spatial judgment is retained. Then, based on the pre-calibrated spatial mapping relationship between the bullet camera and the lidar device, the valid point cloud data is projected onto the monitoring image frame, so that each spatial point cloud has a corresponding position in the image. Based on the overlap between the projected points and the guide line image area and the channel target image area, it is determined which point clouds belong to the guide line, and which belong to trees, construction machinery, foreign objects, or buildings. For example, a guide line area and a tree canopy area have been identified in the bullet camera image. Each point cloud point acquired by the lidar has its own three-dimensional coordinates. According to the calibration relationship between the bullet camera and the lidar, these three-dimensional points are projected onto the bullet camera image. If some point cloud points fall into the guide line area after projection, these points are identified as the guide line corresponding point cloud; if other point cloud points fall into the tree canopy area after projection, these points are identified as the tree canopy corresponding point cloud. In this way, not only can we know where the guide line and the tree canopy are in the image, but we can also know the true position of the corresponding guide line and tree canopy in three-dimensional space, thus providing a basis for subsequent calculation of the spatial distance between the tree canopy and the guide line.

[0022] Based on this, the point clouds corresponding to the conductor and the point clouds corresponding to the target are combined to form a fusion correspondence result, establishing a one-to-one correspondence between the objects identified in the image and their three-dimensional positions in the point cloud. The spatial distribution of the conductor's corresponding point cloud along the transmission line's extension direction is extracted from the fusion correspondence result to form an initial conductor spatial model, such as obtaining the conductor's three-dimensional extension trajectory within the current monitoring area. Simultaneously, the spatial occupancy range of the target's corresponding point cloud is extracted from the fusion correspondence result to determine the three-dimensional coordinate range of channel targets such as tree canopies, cranes, or buildings, obtaining the channel target's spatial location result. Through the above processing, the identification result of "what kind of target" in the bullet image can be combined with the ranging result of "where the target is and how much space it occupies" in the lidar point cloud, providing an accurate data foundation for subsequent judgments on whether the conductor point cloud has truly hit the target, whether the conductor model has a fitting compensation region, and whether the channel target is close to a risk area.

[0023] In one embodiment, S2: Based on the actual hit situation of the lidar point cloud data on the initial traverse space model, the step of dividing the initial traverse space model into the measured support segment and the fitting compensation segment is as follows: According to the direction of the transmission line extension, the initial conductor space model is divided into multiple continuous conductor model segments; Extract the centerline position and endpoint position of each conductor model segment; Using the centerline position of each traverse model segment as the center, generate the point cloud hit detection range corresponding to that traverse model segment; Spatial matching is performed between the point cloud corresponding to the traverse and the point cloud hit detection range corresponding to each traverse model segment to determine the point cloud corresponding to the traverse that falls within the hit detection range of each point cloud. The point cloud hit length of each traverse model segment is determined based on the distribution length of the point cloud corresponding to the traverse that falls within the hit detection range of each point cloud. A traverse model segment whose point cloud hit length is not less than a preset ratio of the corresponding traverse model segment length is determined as the measured support segment; The traverse model segments whose point cloud hit length is less than a preset proportion of the corresponding traverse model segment length are determined as fitting compensation segments.

[0024] It should be noted that in step S2, the initial conductor spatial model is first divided into multiple continuous conductor model segments along the extension direction of the transmission line. For example, a conductor between tower 25 and tower 26 is divided into several small segments of fixed length. This is to determine whether the conductor model has real point cloud support for each segment. Then, the model centerline position and segment endpoint position of each conductor model segment are extracted to clarify the specific range of the segment in three-dimensional space. Next, a point cloud hit detection range is generated with the model centerline position as the center. For example, a narrow spatial enclosure is formed around each conductor model segment to determine whether the lidar point cloud actually falls near the conductor segment. Then, the point cloud corresponding to the conductor is matched with each point cloud. The detection range is matched. If some point clouds fall within the detection range of the third traverse model segment, then these point clouds are considered to have truly hit the third traverse model segment. Then, the point cloud hit length is determined based on the coverage of these hit point clouds on the segment. For example, if the length of the third segment is 2 meters and the hit point clouds cover 1.7 meters along the segment, then its point cloud hit length is 1.7 meters. Finally, traverse model segments with point cloud hit lengths reaching a preset ratio are determined as measured support segments, and traverse model segments that do not reach the preset ratio are determined as fitting compensation segments. For example, when the preset ratio is 60%, the third segment covering 1.7 meters is determined as a measured support segment, while the fourth segment covering only 0.4 meters is determined as a fitting compensation segment. Through the above processing, it is possible to distinguish between the "parts that are indeed supported by lidar point clouds" and the "parts that are mainly obtained by model completion" in the initial traverse spatial model, providing a clear object for subsequent neighborhood trend analysis and anomaly verification for the fitted compensation segment, and avoiding the direct use of the entire initial traverse spatial model as the actual traverse position.

[0025] In one embodiment, S3: The step of determining the neighborhood extension trend corresponding to the fitted compensation segment based on the measured support segments adjacent to the fitted compensation segment is as follows: Based on the division results of the measured support section and the fitted compensation section, determine the front adjacent measured support section and the rear adjacent measured support section of each fitted compensation section along the extension direction of the transmission line. Extract the spatial center point sequence of the point cloud corresponding to the traverse in the adjacent front and rear measured support segments respectively; According to the extension direction of the transmission line, the spatial center point sequence of adjacent measured support sections on the front side is sequentially connected to obtain the front measured extension line. According to the extension direction of the transmission line, the spatial center point sequence of the adjacent measured support section on the rear side is sequentially connected to obtain the measured extension line on the rear side. The direction of extension of the front neighborhood is determined based on the height change direction and lateral offset direction of the front measured extension line near the fitting compensation section. The extension direction of the rear neighborhood is determined based on the height change direction and lateral offset direction of the measured extension line near the fitting compensation segment. The forward and backward neighborhood extension directions are used together as the neighborhood extension trend corresponding to the fitting compensation segment.

[0026] It should be noted that in step S3, the nearest measured support segments on both sides of each fitting compensation segment are first determined from the division results of the measured support segments and fitting compensation segments obtained in step S2. For example, if the fourth conductor model sub-segment is determined as a fitting compensation segment, then the third measured support segment before it and the fifth measured support segment after it are found. Then, the corresponding point clouds of the conductor in the third and fifth measured support segments are extracted and arranged into a spatial center point sequence according to the conductor extension direction. For example, the center positions of each small segment of point cloud are arranged sequentially to form a series of spatial points that can represent the true conductor trend of the measured support segment. Then, the spatial center point sequence of the front side is connected sequentially to obtain the front measured extension line. The sequence of spatial center points on the rear side is sequentially connected to obtain the measured extension line on the rear side, reflecting the current spatial orientation of the real conductors on both sides of the fitting compensation segment. Then, the measured extension lines on the front and rear sides are observed to see if they exhibit changes such as sinking, rising, shifting to the left, or shifting to the right near the fitting compensation segment. For example, if the measured extension line on the front side gradually sinks and shifts towards the tree crown near the fourth fitting compensation segment, and the measured extension line on the rear side also shows a similar sinking trend near the fourth fitting compensation segment, then the neighborhood extension trend corresponding to this fitting compensation segment can be determined to be downward and extending towards the tree crown. Finally, the neighborhood extension directions on the front and rear sides are used together as the neighborhood extension trend of this fitting compensation segment. Through the above processing, the actual direction of change of the conductor near the missing point cloud area can be inferred by using the actual conductor point cloud trend collected on both sides of the fitting compensation segment. This avoids directly using a smoothed fitting curve and ignoring abnormal trends caused by local drooping, wind deflection, or icing, providing a basis for subsequent judgment of whether there are anomalies masked by smoothing in the fitting compensation segment.

[0027] In one embodiment, S4: The step of verifying the consistency between the fitted compensation segment and the corresponding neighborhood extension trend, and determining the suspicious conductor segments in the fitted compensation segment, is as follows: The fit compensation segment is checked for consistency with the corresponding neighborhood extension trend. The neighborhood squeeze fit index and the trend release continuity index are calculated. The neighborhood squeeze fit index and the trend release continuity index are added together to obtain the consistency index. Suspicious conductor segments in the fit compensation segment are determined based on the consistency index and the preset consistency index threshold.

[0028] In one implementation, the calculation steps for the neighborhood squeeze fit index are as follows: Along the extension direction of the fitting compensation segment, the fitting compensation segment is divided into multiple equally ordered positions, and the fitting compensation segment cross-section points corresponding to each equally ordered position are extracted. Based on the extension direction of the front neighborhood, extend from the end of the front adjacent measured support segment close to the fitting compensation segment into the interior of the fitting compensation segment to obtain the front clamping point corresponding to each equidistant position. Based on the extension direction of the rear neighborhood, extend backward from the end of the adjacent measured support segment close to the fitting compensation segment into the interior of the fitting compensation segment to obtain the rear clamping point corresponding to each equidistant position. Calculate the spatial distance between the fitting compensation section point and the front clamping point, the spatial distance between the fitting compensation section point and the rear clamping point, and the spatial distance between the front clamping point and the rear clamping point at each of the equal order positions. For each equidistant position, twice the spatial distance between the front clamping point and the rear clamping point is divided by the sum of the spatial distance between the fitting compensation section cross-section point and the front clamping point, the spatial distance between the fitting compensation section cross-section point and the rear clamping point, and the spatial distance between the front clamping point and the rear clamping point, to obtain the cross-section clamping fit value corresponding to that equidistant position. Multiply the cross-sectional clamping and matching values ​​corresponding to all the equally ordered positions to obtain the cross-sectional clamping and matching product; The neighborhood squeezing coincidence index is obtained by taking the square root of the product of cross-sectional squeezing coincidences corresponding to the number of equidistant positions.

[0029] It should be noted that the data involved in the calculation of the neighborhood squeeze fit index all come from the aforementioned spatial fusion results, traverse segmentation results, and neighborhood extension trend results. Specifically, the fitting compensation segment comes from the fitting compensation segment obtained in step S2 by segmenting the initial traverse spatial model and dividing it according to the point cloud hit length; the extension direction of the fitting compensation segment comes from the orientation of the initial traverse spatial model within the fitting compensation segment; the equiorder position and the cross-section point of the fitting compensation segment are extracted from the center line of the fitting compensation segment after dividing it at the same interval along the extension direction of the fitting compensation segment; the front adjacent measured support segment and the rear adjacent measured support segment come from the measured support segment determined in step S3 according to the positional relationship on both sides of the fitting compensation segment; the front neighborhood extension direction... The directions of extension to the front and rear neighborhoods are determined based on the spatial center point sequence of the corresponding point cloud of the traverse in the measured support segments on both sides and the measured extension lines. Based on this, the front clamping points for each equi-order position are obtained by extrapolating from the end of the adjacent front measured support segment near the fitting compensation segment, following the front neighborhood extension direction into the fitting compensation segment. Similarly, the rear clamping points for each equi-order position are obtained by extrapolating from the end of the adjacent rear measured support segment near the fitting compensation segment, following the rear neighborhood extension direction into the fitting compensation segment. The spatial distances are directly calculated from the three-dimensional coordinates of the corresponding points. Therefore, this index does not require the introduction of new sensor data; instead, it utilizes the already obtained traverse point cloud, the position of the fitting compensation segment, and the actual traverse extension trends on both sides to quantitatively verify whether the fitting compensation segment conforms to the actual traverse trends on both sides.

[0030] It should be noted that the neighborhood squeezing fit index is a quantitative indicator used to measure whether a fitted compensation segment falls within a reasonable spatial squeezing range jointly extended by its adjacent front and rear measured support segments. Essentially, it determines whether the "fitted conductor segment" can be naturally squeezed and supported by the actual point cloud trends on both sides. Since the fitted compensation segment itself lacks sufficient actual point cloud support, its position cannot be directly identified as the true position of the conductor. Therefore, it is necessary to use the actual conductor trends of the front and rear measured support segments for reverse verification. If the front squeezing point formed by the extension of the front measured support segment into the fitted compensation segment, and the rear squeezing point formed by the reverse extension of the rear measured support segment, can jointly define a reasonable spatial range, and the cross-section point on the fitted compensation segment happens to be located near this range, then it indicates that the fitted compensation segment has a good connection with the actual conductor trends on both sides, and the neighborhood squeezing fit index will be large. Conversely, if the actual traverse trends on both sides indicate that the traverse should continue to descend or shift to one side, but the fitted compensation segment remains smooth, centered, or close to the historical model position, then the cross-sectional points of the fitted compensation segment will deviate from the reasonable range formed by the preceding and following clamping points, and the neighborhood clamping fit index will decrease. For example, if the point cloud in the middle of a traverse segment is missing, and a relatively smooth traverse segment is fitted, but the measured traverse point clouds on both sides of the missing segment already show a downward sag trend, if the fitted compensation segment is also located within the spatial channel clamped by the downward trends on both sides, it indicates that the compensation segment better matches the actual traverse changes, and the index is larger; if the fitted compensation segment is significantly higher than the clamping channel, it indicates that it may have smoothed out the actual local sag increase, and the index is smaller. Therefore, the larger the neighborhood squeeze fit index, the more the fitted compensation segment conforms to the true conductor extension trend reflected by the adjacent measured support segments, and the lower the possibility that the fitted compensation segment is masked by excessive smoothing. The smaller the neighborhood squeeze fit index, the more likely the fitted compensation segment is to deviate from the true neighborhood trend, and the more it needs to be judged as a suspicious conductor segment and enter the subsequent risk envelope analysis.

[0031] In one implementation, the calculation steps for the trend change continuity index are as follows: Extract the front end change vector from the end of the front measured extension line near the fitting compensation section, and extract the rear end change vector from the end of the rear measured extension line near the fitting compensation section. Divide the change vector of the front end by the length of the change vector of the front end to obtain the unit change vector of the front end; divide the change vector of the rear end by the length of the change vector of the rear end to obtain the unit change vector of the rear end. Add the unit change vector on the front side and the unit change vector on the back side, and divide the sum by the length of the sum to obtain the main direction of the release. Multiply the length of the change vector at the front end with the length of the change vector at the rear end, and then perform a square root operation on the result of the multiplication to obtain the neighborhood reference variable. Extract the cross-sectional points of two adjacent fitting compensation segments sequentially along the extension direction of the fitting compensation segment, and calculate the spatial displacement of the cross-sectional point of the latter fitting compensation segment relative to the cross-sectional point of the former fitting compensation segment. Multiply the spatial displacement by the main direction of the elution to obtain the directional projection value of the corresponding adjacent cross-sectional interval; Add the absolute values ​​of the directional projection values ​​together and divide the result by two to obtain the positive release variables for the corresponding adjacent cross-sectional intervals. For each adjacent cross-sectional interval, the positive explanatory variable is multiplied by the neighboring baseline explanatory variable, the result of the multiplication is squared and then multiplied by two, and the result is divided by the sum of the positive explanatory variable and the neighboring baseline explanatory variable to obtain the segmental explanatory variable transfer value of the adjacent cross-sectional interval. For any two adjacent cross-sectional intervals, multiply the positive explanatory variable of the previous adjacent cross-sectional interval with the positive explanatory variable of the next adjacent cross-sectional interval, perform a square root operation on the multiplication result, multiply by two, and divide the result by the sum of the positive explanatory variables of the previous adjacent cross-sectional interval and the next adjacent cross-sectional interval to obtain the adjacent explanatory continuous value. Multiply all the segmented emanation values ​​and all the adjacent emanation continuous values ​​to obtain the emanation continuous product; The trend exothermic continuous index is obtained by taking the square root of the exothermic continuous product and the total number of the segmented exothermic successor values ​​and adjacent exothermic continuous values ​​involved in the multiplication.

[0032] It should be noted that the data involved in the calculation of the trend change continuous index all originate from the fitted compensation segment, adjacent measured support segment, measured extension line, and fitted compensation segment cross-sectional points obtained in the aforementioned steps. Specifically, the fitted compensation segment originates from the fitted compensation segment determined after segmenting the initial traverse spatial model in step S2; the front and rear measured extension lines originate from the results obtained in step S3 by sequentially connecting the spatial center point sequences of the corresponding point cloud of the traverse in the front and rear adjacent measured support segments; the front end change vector is obtained by subtracting the three-dimensional coordinates of two adjacent spatial center points near the fitted compensation segment end in the front measured extension line; the rear end change vector is obtained by subtracting the three-dimensional coordinates of the two adjacent spatial center points near the fitted compensation segment end in the rear measured extension line. The three-dimensional coordinates of two adjacent spatial center points are subtracted to obtain the fitting compensation segment cross-section points, which are derived from multiple equally ordered cross-section points obtained by dividing along the extension direction of the fitting compensation segment in the neighborhood clamping fit index calculation. The spatial displacement between adjacent cross-sections is obtained by subtracting the three-dimensional coordinates of the previous fitting compensation segment cross-section point from the three-dimensional coordinates of the subsequent fitting compensation segment cross-section point. The length of each vector, the direction projection value, the positive release variable, the segmental release variable continuity value, and the adjacent release variable continuity value are all obtained from the above three-dimensional coordinates, the end change vector, and the fitting compensation segment cross-section points according to the established calculation rules. It can be seen that this index does not require additional new data collection, but uses the actual end change trend reflected by the adjacent measured traverse point cloud, as well as the cross-sectional position change of the fitting compensation segment itself, to determine whether the fitting compensation segment continuously receives the subsidence, lateral movement, or sag change trend of the actual traverse on both sides.

[0033] It should be noted that the trend release continuity index is a quantitative indicator used to measure whether the fitting compensation segment continuously inherits the actual traverse change trend shown by the adjacent measured support segments before and after it. Its focus is not on judging whether the "position of the fitting compensation segment is close to the point clouds on both sides", but on judging whether the changes such as sinking, lateral movement, and increased sag that have already appeared on both sides of the actual traverse are naturally continued and gradually released within the fitting compensation segment. Because the fitting compensation segment itself lacks sufficient real point cloud support, it is easily pulled back to a relatively stable position by the historical traverse model or smooth fitting rules. Therefore, if the front measured support segment already shows a downward sloping trend when it approaches the fitting compensation segment, and the rear measured support segment also shows a similar downward trend when it approaches the fitting compensation segment, and the adjacent cross-section points within the fitting compensation segment also continue to change along this trend direction, it indicates that the fitting compensation segment has well inherited the changing state of the real traverse on both sides, and the trend release continuity index is large. Conversely, if the measured support segments on both sides already show that the traverse is shifting downward or to one side, but the fitting compensation segment suddenly flattens out, returns to normal, or the amount of change weakens significantly, it indicates that the real trend of change is interrupted or smoothed out in the fitting segment, and the trend release continuity index will decrease. For example, in an icing scenario, if the measured point clouds on both sides of a missing section of a conductor show that the conductor is gradually sags, and the fitted compensation segment also shows continuous sag from front to back, it indicates that the compensation segment conforms to the true continuation of icing sag, resulting in a larger index. However, if the fitted compensation segment maintains a curve close to the historical smooth curve and does not reflect this sag release process, it may mask the true sag risk caused by local icing, resulting in a smaller index. Therefore, the larger the trend change continuity index, the more continuously the fitted compensation segment can inherit and release the true conductor change trend reflected by the measured support segment in the neighborhood, and the more likely it is that the fitted compensation segment has not significantly truncated or weakened local abnormal changes. Conversely, the smaller the trend change continuity index, the more likely the fitted compensation segment may have problems with over-smoothing, trend interruption, or abnormality erasure, requiring further risk assessment as a suspicious conductor segment.

[0034] In one embodiment, the step of determining the suspicious conductor segment in the fitting compensation segment based on the consistency index and a preset consistency index threshold is as follows: Compare the consistency index with a preset consistency index threshold; When the consistency index is less than the preset consistency index threshold, the corresponding fitting compensation segment is identified as a suspicious conductor segment. When the consistency index is not less than the preset consistency index threshold, the corresponding fitting compensation segment is determined as a non-suspicious conductor segment. All suspicious conductor segments are arranged in order of their position in the initial conductor space model to obtain a set of suspicious conductor segments.

[0035] It should be noted that when determining suspicious traverse segments based on the consistency index and a preset consistency index threshold, the calculated consistency index of each fitted compensation segment is first read and compared with the preset consistency index threshold. For example, if the consistency index of a fitted compensation segment is 1.62 and the preset consistency index threshold is 1.50, it indicates that the fitted compensation segment closely matches the neighborhood extension trend of the measured support segments on both sides, and can be identified as a non-suspicious traverse segment. If the consistency index of another fitted compensation segment is 1.18, which is lower than the preset consistency index threshold, it indicates that the fitted compensation segment is insufficient in terms of clamping fit or trend release continuity, and may not truly reflect the subsidence, lateral movement, or sag change trends reflected by the point clouds of the traverse on both sides. Therefore, it is identified as a suspicious traverse segment. Subsequently, all fitted compensation segments identified as suspicious traverse segments are arranged according to their position in the initial traverse space model, for example, arranged sequentially according to the traverse extension direction from tower 25 to tower 26, forming a set of suspicious traverse segments. The advantage of this approach is that it can filter out the parts of the point cloud that truly have the risk of oversmoothing, rather than amplifying the risk of all fitted compensation segments. This avoids missing local abnormal conductor segments that may be masked by fitting, reduces misjudgment of normal point cloud missing segments, and provides clear, continuous, and locatable processing objects for the subsequent generation of local risk envelope regions.

[0036] In one embodiment, S5: The step of generating a local risk envelope region for a suspicious conductor segment based on the neighborhood extension trend corresponding to the suspicious conductor segment is as follows: Extract the start and end points of the suspected traverse segment from the initial traverse space model; Based on the forward neighbor extension direction corresponding to the suspicious conductor segment, a forward trend extension trajectory is generated from the starting point of the segment into the interior of the suspicious conductor segment. Based on the extension direction of the back neighbor corresponding to the suspicious conductor segment, a back trend extension trajectory is generated from the end point of the segment to the interior of the suspicious conductor segment. The positions of the forward trend extension trajectory and the backward trend extension trajectory are matched within the suspicious guide segment to obtain the forward trend point and the backward trend point at each corresponding position; At each corresponding position, the spatial line connecting the previous trend point and the subsequent trend point is determined as the risk cross-section of that position. Connect all the risk cross-sections at the corresponding locations in sequence according to the extension direction of the suspicious conductor section to form a trend clamping surface; Centered on the trend clamping surface, a safe distance is extended along the outward direction perpendicular to the trend clamping surface to obtain the local risk envelope area of ​​the suspicious conductor segment.

[0037] It should be noted that in step S5, the starting point and ending point of the suspected conductor segment are first extracted from the initial conductor space model. For example, if a suspected conductor segment is located in the middle between towers 25 and 26, then the front and rear positions of the segment in the model are determined. Then, based on the forward neighborhood extension direction obtained in step S3, a forward trend extension trajectory is calculated from the segment starting point towards the interior of the segment. And based on the backward neighborhood extension direction, a backward trend extension trajectory is calculated from the segment ending point towards the interior of the segment. For example, if the forward real conductor shows a downward and canopy-side shift, the backward real conductor also shows a downward shift. Extending the trend, the two trend extension trajectories will both point to the spatial range where the conductor may sink. Then, the front and rear trend extension trajectories are matched according to their corresponding positions within the segment, resulting in multiple front and rear trend points. These front and rear trend points are then connected at each corresponding position to form a risk cross-section. Subsequently, these risk cross-sections are connected sequentially along the extension direction of the suspected conductor segment to form a continuous trend clamping surface. This trend clamping surface represents the spatial zone that the actual conductor may traverse within the missing point cloud segment. Finally, the area is expanded outward from the trend clamping surface at a preset safety distance to obtain a local risk envelope region. The advantage of this approach is that it no longer simply treats the suspected conductor segment as a fixed fitted curve, but rather transforms the possible spatial range that the trends of the two actual conductors on both sides point to into a risk assessment area. This allows for the inclusion of conductor positions that may be masked by smooth fitting, such as local sag, wind deviation, and icing subsidence, into subsequent distance assessments, reducing the risk of missed reports due to a single fitted line deviating from the actual conductor.

[0038] In one embodiment, S6: The step of determining the transmission line channel risk monitoring results based on the spatial relationship between the channel target spatial location results and the local risk envelope area is as follows: Extract the three-dimensional coordinate range of each channel target from the spatial location results of the channel targets; The spatial intersection of the three-dimensional coordinate range corresponding to each channel target with the local risk envelope region is determined. When the three-dimensional coordinate range corresponding to any channel target spatially overlaps with the local risk envelope region, the channel target is identified as an intrusion risk target. When the three-dimensional coordinate range corresponding to any channel target does not spatially overlap with the local risk envelope region, calculate the minimum spatial distance between the three-dimensional coordinate range corresponding to the channel target and the local risk envelope region. Compare the minimum spatial distance with the preset safe distance threshold; When the minimum spatial distance is less than the preset safe distance threshold, the target in the channel is identified as a nearby risk target; When the minimum spatial distance is not less than the preset safe distance threshold, the target in the channel is identified as a non-risk target. Based on intrusion risk targets, adjacent risk targets, and non-risk targets, the risk monitoring results of the transmission line corridor are generated.

[0039] It should be noted that in step S6, the three-dimensional coordinate range of each channel target is first extracted from the channel target spatial location results, such as the three-dimensional enclosing range corresponding to a tree canopy, a crane arm, or a building edge; then, the three-dimensional coordinate range of each channel target is spatially intersected with the local risk envelope region generated in step S5. If the three-dimensional coordinate range of a tree canopy overlaps with the local risk envelope region, it indicates that the tree canopy may have entered the risk space that the actual conductor may pass through, and it is identified as an intrusion risk target; if a crane arm does not overlap with the local risk envelope region, the three-dimensional coordinate range of the crane arm is further calculated to... The minimum spatial distance between local risk envelope regions is calculated, for example, the closest distance from the point cloud at the end of a crane boom to the boundary of the local risk envelope region. This minimum spatial distance is then compared to a preset safety distance threshold. If the minimum spatial distance is less than the preset safety distance threshold, it indicates that although the target has not yet invaded the risk envelope region, it is already within a relatively close range and is identified as a nearby risk target. If the minimum spatial distance is not less than the preset safety distance threshold, it is identified as a non-risk target. Finally, based on invading risk targets, nearby risk targets, and non-risk targets, the transmission line corridor risk monitoring results are generated, such as outputting "tree canopy intrusion risk," "crane boom proximity risk," or "building non-risk." The advantage of this approach is that risk assessment is no longer based on a smoothed, fitted single conductor curve, but rather on the local risk envelope region. This allows the possible locations of the actual conductor within the missing point cloud segment to be included in the safety distance assessment, thereby reducing underreporting of risks such as tree obstructions, foreign objects, or construction machinery due to overly conservative conductor fitting positions, and improving the reliability of transmission line corridor monitoring results.

[0040] Based on the same inventive concept, embodiments of the present invention also provide a transmission line gun monitoring system integrating lidar. This system includes: Data fusion unit, data processing unit, and data monitoring unit, among which, Data fusion unit: By acquiring bullet camera image data and lidar point cloud data of the transmission line monitoring area, and spatially fusing the bullet camera image data and lidar point cloud data, the initial conductor spatial model and channel target spatial location results are obtained; The data fusion unit divides the initial traverse space model into a measured support segment and a fitting compensation segment based on the actual hit situation of the lidar point cloud data on the initial traverse space model. It also determines the neighborhood extension trend corresponding to the fitting compensation segment, performs consistency verification between the fitting compensation segment and the corresponding neighborhood extension trend, identifies suspicious traverse segments in the fitting compensation segment, and generates the local risk envelope region of the suspicious traverse segment based on the neighborhood extension trend corresponding to the suspicious traverse segment. Data monitoring unit: Determines the risk monitoring results of transmission line channels based on the spatial relationship between the target spatial location results and the local risk envelope area.

[0041] Based on the same inventive concept, embodiments of the present invention also provide a transmission line gun monitoring device that integrates lidar. This device includes: Fusion module: Acquires bullet camera image data and lidar point cloud data of the transmission line monitoring area, and performs spatial fusion on the bullet camera image data and lidar point cloud data to obtain the initial conductor spatial model and channel target spatial location results; Module division: Based on the actual hit situation of LiDAR point cloud data on the initial traverse space model, the initial traverse space model is divided into measured support segment and fitting compensation segment; Extension Trend Module: Based on the measured support segments adjacent to the fitted compensation segment, determine the neighborhood extension trend corresponding to the fitted compensation segment; Suspicious conductor module: Performs consistency verification between the fitted compensation segment and the corresponding neighborhood extension trend to identify suspicious conductor segments in the fitted compensation segment; Risk envelope module: Based on the neighborhood extension trend corresponding to the suspicious conductor segment, generate the local risk envelope region of the suspicious conductor segment; Risk monitoring module: Determines the risk monitoring results of transmission line channels based on the spatial relationship between the target spatial location results and the local risk envelope area.

[0042] The transmission line gun monitoring system based on the fused lidar provided in this invention, after spatially fusing gun image data and lidar point cloud data, first distinguishes between measured support segments and fitted compensation segments based on the actual hit situation of lidar point cloud data on the initial conductor spatial model. Then, it determines the neighborhood extension trend by using the local change state of the conductor reflected by the adjacent measured support segments of the fitted compensation segment, and performs consistency verification between the fitted compensation segment and the neighborhood extension trend. This enables the identification of suspicious conductor segments that may be masked by smooth fitting due to missing point clouds. Furthermore, it generates a local risk envelope region based on the neighborhood extension trend corresponding to the suspicious conductor segments, and uses channels... The spatial relationship between the target spatial location result and the local risk envelope area determines the risk monitoring results of the transmission line corridor. This makes risk assessment no longer solely dependent on the single conductor model after fitting and completion. Instead, it includes potential local sag increases, local wind deflection, icing subsidence, or deformation near clamps in the missing areas of the conductor point cloud. This avoids mistaking the smoothed conductor position obtained from fitting and compensation as the actual conductor position for distance calculation, reduces the possibility of overestimating the safe distance between trees, foreign objects, or construction machinery and the conductor, reduces the problem of local hidden dangers being masked and risk warnings being delayed, and improves the authenticity and reliability of the transmission line corridor monitoring results.

[0043] In another embodiment of the present invention, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to perform power line gun monitoring of any of the fusion lidars in the above embodiments.

[0044] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention should still fall within the scope of the claims of the present invention.

Claims

1. A transmission line gun-mounted monitoring method integrating lidar, characterized in that, Includes the following steps: S1: Acquire bullet camera image data and lidar point cloud data of the transmission line monitoring area, and perform spatial fusion on the bullet camera image data and lidar point cloud data to obtain the initial conductor spatial model and channel target spatial location results; S2: Based on the actual hit situation of lidar point cloud data on the initial traverse space model, the initial traverse space model is divided into the measured support segment and the fitting compensation segment; S3: Determine the neighborhood extension trend corresponding to the fitted compensation segment based on the measured support segments adjacent to the fitted compensation segment; S4: Perform consistency verification between the fitted compensation segment and the corresponding neighborhood extension trend to identify suspicious conductor segments in the fitted compensation segment; S5: Based on the neighborhood extension trend corresponding to the suspicious conductor segment, generate the local risk envelope region of the suspicious conductor segment; S6: Determine the risk monitoring results of transmission line channels based on the spatial relationship between the target spatial location results and the local risk envelope area.

2. The transmission line gun-mounted monitoring method based on fused lidar according to claim 1, characterized in that, Based on the actual hit data of LiDAR point cloud data on the initial traverse spatial model, the steps to divide the initial traverse spatial model into measured support segments and fitted compensation segments are as follows: According to the direction of the transmission line extension, the initial conductor space model is divided into multiple continuous conductor model segments; Extract the centerline position and endpoint position of each conductor model segment; Using the centerline position of each traverse model segment as the center, generate the point cloud hit detection range corresponding to that traverse model segment; Spatial matching is performed between the point cloud corresponding to the traverse and the point cloud hit detection range corresponding to each traverse model segment to determine the point cloud corresponding to the traverse that falls within the hit detection range of each point cloud. The point cloud hit length of each traverse model segment is determined based on the distribution length of the point cloud corresponding to the traverse that falls within the hit detection range of each point cloud. A traverse model segment whose point cloud hit length is not less than a preset ratio of the corresponding traverse model segment length is determined as the measured support segment; The traverse model segments whose point cloud hit length is less than a preset proportion of the corresponding traverse model segment length are determined as fitting compensation segments.

3. The transmission line gun-mounted monitoring method based on fused lidar according to claim 1, characterized in that, The steps to determine the neighborhood extension trend corresponding to the fitted compensation segment based on the adjacent measured support segments are as follows: Based on the division results of the measured support section and the fitted compensation section, determine the front adjacent measured support section and the rear adjacent measured support section of each fitted compensation section along the extension direction of the transmission line. Extract the spatial center point sequence of the point cloud corresponding to the traverse in the adjacent front and rear measured support segments respectively; According to the extension direction of the transmission line, the spatial center point sequence of adjacent measured support sections on the front side is sequentially connected to obtain the front measured extension line. According to the extension direction of the transmission line, the spatial center point sequence of the adjacent measured support section on the rear side is sequentially connected to obtain the measured extension line on the rear side. The direction of extension of the front neighborhood is determined based on the height change direction and lateral offset direction of the front measured extension line near the fitting compensation section. The extension direction of the rear neighborhood is determined based on the height change direction and lateral offset direction of the measured extension line near the fitting compensation segment. The forward and backward neighborhood extension directions are used together as the neighborhood extension trend corresponding to the fitting compensation segment.

4. The transmission line gun-mounted monitoring method based on fused lidar according to claim 1, characterized in that, The steps to verify the consistency between the fitted compensation segment and the corresponding neighborhood extension trend, and to identify suspicious conductor segments in the fitted compensation segment, are as follows: The fit compensation segment is checked for consistency with the corresponding neighborhood extension trend. The neighborhood squeeze fit index and the trend release continuity index are calculated. The neighborhood squeeze fit index and the trend release continuity index are added together to obtain the consistency index. Suspicious conductor segments in the fit compensation segment are determined based on the consistency index and the preset consistency index threshold.

5. The transmission line gun-mounted monitoring method based on fused lidar according to claim 4, characterized in that, The calculation steps for the neighborhood squeeze coincidence index are as follows: Along the extension direction of the fitting compensation segment, the fitting compensation segment is divided into multiple equally ordered positions, and the fitting compensation segment cross-section points corresponding to each equally ordered position are extracted. Based on the extension direction of the front neighborhood, extend from the end of the front adjacent measured support segment close to the fitting compensation segment into the interior of the fitting compensation segment to obtain the front clamping point corresponding to each equidistant position. Based on the extension direction of the rear neighborhood, extend backward from the end of the adjacent measured support segment close to the fitting compensation segment into the interior of the fitting compensation segment to obtain the rear clamping point corresponding to each equidistant position. Calculate the spatial distance between the fitting compensation section point and the front clamping point, the spatial distance between the fitting compensation section point and the rear clamping point, and the spatial distance between the front clamping point and the rear clamping point at each of the equal order positions. For each equidistant position, twice the spatial distance between the front clamping point and the rear clamping point is divided by the sum of the spatial distance between the fitting compensation section cross-section point and the front clamping point, the spatial distance between the fitting compensation section cross-section point and the rear clamping point, and the spatial distance between the front clamping point and the rear clamping point, to obtain the cross-section clamping fit value corresponding to that equidistant position. Multiply the cross-sectional clamping and matching values ​​corresponding to all the equally ordered positions to obtain the cross-sectional clamping and matching product; The neighborhood squeezing coincidence index is obtained by taking the square root of the product of cross-sectional squeezing coincidences corresponding to the number of equidistant positions.

6. The transmission line gun-mounted monitoring method based on fused lidar according to claim 4, characterized in that, The calculation steps for the trend reversal continuous index are as follows: Extract the front end change vector from the end of the front measured extension line near the fitting compensation section, and extract the rear end change vector from the end of the rear measured extension line near the fitting compensation section. Divide the change vector of the front end by the length of the change vector of the front end to obtain the unit change vector of the front end; divide the change vector of the rear end by the length of the change vector of the rear end to obtain the unit change vector of the rear end. Add the unit change vector on the front side and the unit change vector on the back side, and divide the sum by the length of the sum to obtain the main direction of the release. Multiply the length of the change vector at the front end with the length of the change vector at the rear end, and then perform a square root operation on the result of the multiplication to obtain the neighborhood reference variable. Extract the cross-sectional points of two adjacent fitting compensation segments sequentially along the extension direction of the fitting compensation segment, and calculate the spatial displacement of the cross-sectional point of the latter fitting compensation segment relative to the cross-sectional point of the former fitting compensation segment. Multiply the spatial displacement by the main direction of the elution to obtain the directional projection value of the corresponding adjacent cross-sectional interval; Add the absolute values ​​of the directional projection values ​​together and divide the sum by two to obtain the positive release variables for the corresponding adjacent cross-sectional intervals; For each adjacent cross-sectional interval, the positive explanatory variable is multiplied by the neighboring baseline explanatory variable, the result of the multiplication is squared and then multiplied by two, and the result is divided by the sum of the positive explanatory variable and the neighboring baseline explanatory variable to obtain the segmental explanatory variable transfer value of the adjacent cross-sectional interval. For any two adjacent cross-sectional intervals, multiply the positive explanatory variable of the previous adjacent cross-sectional interval with the positive explanatory variable of the next adjacent cross-sectional interval, perform a square root operation on the multiplication result, multiply by two, and divide the result by the sum of the positive explanatory variables of the previous adjacent cross-sectional interval and the next adjacent cross-sectional interval to obtain the adjacent explanatory continuous value. Multiply all the segmented emanation values ​​and all the adjacent emanation values ​​together to obtain the emanation continuous product; The trend exothermic continuous index is obtained by taking the square root of the exothermic continuous product and the total number of the segmented exothermic successor values ​​and adjacent exothermic continuous values ​​involved in the multiplication.

7. The transmission line gun-mounted monitoring method based on fused lidar according to claim 4, characterized in that, The steps for determining suspicious traverse segments in the fitted compensation segment based on the consistency index and a preset consistency index threshold are as follows: Compare the consistency index with a preset consistency index threshold; When the consistency index is less than the preset consistency index threshold, the corresponding fitting compensation segment is identified as a suspicious conductor segment. When the consistency index is not less than the preset consistency index threshold, the corresponding fitting compensation segment is determined as a non-suspicious conductor segment. All suspicious conductor segments are arranged in order of their position in the initial conductor space model to obtain a set of suspicious conductor segments.

8. A transmission line gun-mounted monitoring system integrating lidar, characterized in that, The system includes a data fusion unit, a data processing unit, and a data monitoring unit, wherein... Data fusion unit: By acquiring bullet camera image data and lidar point cloud data of the transmission line monitoring area, and spatially fusing the bullet camera image data and lidar point cloud data, the initial conductor spatial model and channel target spatial location results are obtained; The data fusion unit divides the initial traverse space model into a measured support segment and a fitting compensation segment based on the actual hit situation of the lidar point cloud data on the initial traverse space model. It also determines the neighborhood extension trend corresponding to the fitting compensation segment, performs consistency verification between the fitting compensation segment and the corresponding neighborhood extension trend, identifies suspicious traverse segments in the fitting compensation segment, and generates the local risk envelope region of the suspicious traverse segment based on the neighborhood extension trend corresponding to the suspicious traverse segment. Data monitoring unit: Determines the risk monitoring results of transmission line channels based on the spatial relationship between the target spatial location results and the local risk envelope area.

9. A transmission line gun-mounted monitoring device integrating lidar, used to implement the transmission line gun-mounted monitoring method integrating lidar as described in any one of claims 1-7, characterized in that, The device includes: Fusion module: Acquires bullet camera image data and lidar point cloud data of the transmission line monitoring area, and performs spatial fusion on the bullet camera image data and lidar point cloud data to obtain the initial conductor spatial model and channel target spatial location results; Module division: Based on the actual hit situation of LiDAR point cloud data on the initial traverse space model, the initial traverse space model is divided into measured support segment and fitting compensation segment; Extension Trend Module: Based on the measured support segments adjacent to the fitted compensation segment, determine the neighborhood extension trend corresponding to the fitted compensation segment; Suspicious conductor module: Performs consistency verification between the fitted compensation segment and the corresponding neighborhood extension trend to identify suspicious conductor segments in the fitted compensation segment; Risk envelope module: Based on the neighborhood extension trend corresponding to the suspicious conductor segment, generate the local risk envelope region of the suspicious conductor segment; Risk monitoring module: Determines the risk monitoring results of transmission line channels based on the spatial relationship between the target spatial location results and the local risk envelope area.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the transmission line gun-mounted monitoring method of any one of claims 1-7 using fused lidar.

Images