Live working safety distance real-time monitoring method, device, equipment and medium

Abstract

The application provides a live monitoring method and device for a live working safety distance, equipment and a medium, and belongs to the technical field of power safety. The method comprises the following steps: acquiring basic point cloud data of integrated marking information corresponding to a target power transmission line area; obtaining real-time point cloud data of the target working area in absolute coordinates based on real-time original point cloud data, real-time absolute coordinates and real-time attitude data; performing point cloud extraction on a worker based on the basic point cloud data of the integrated marking information and the real-time point cloud data to obtain target point cloud data corresponding to the worker; extracting point cloud edge points of the target point cloud data, calculating the spatial distance between each edge point and each marking information, and determining a target spatial distance from the multiple spatial distances; and performing real-time monitoring on the worker based on the target spatial distance. The live working safety distance live monitoring method and device, equipment and medium provided by the application can solve the problem that effective safety protection cannot be provided for workers.

Description

Technical Field

[0001] This application belongs to the field of power safety technology, and more specifically, relates to a method, device, equipment, and medium for real-time monitoring of safe distances for live-line work. Background Technology

[0002] In the field of monitoring safe distances for live-line work on power transmission lines, ensuring the safety of workers requires real-time control of the air gap between personnel and live or grounded conductors. Currently, there are two main types of mainstream technical solutions: one is the wearable positioning device solution, which collects location data by wearing multiple real-time dynamic positioning devices on the limbs and head of the workers, and calculates the safe distance by combining it with a pre-acquired laser point cloud model of the towers and lines; the other is the binocular camera vision solution, which collects images by using a fixed-position binocular camera, obtains spatial information through stereo matching, distinguishes personnel from equipment through scene recognition, and calculates the near-electric safety distance based on coordinates.

[0003] However, existing solutions have significant limitations: wearable devices increase the burden on personnel, strong electric fields can easily cause signal failure, and they can only monitor the wearing area, resulting in blind spots; binocular cameras are affected by strong light or obstructions, have high computational load and poor real-time performance, lack flexibility in fixed installation, and coordinate determination cannot accurately reflect the actual edge contour. These problems lead to insufficient accuracy and reliability in safety distance monitoring, failing to provide stable and effective safety guarantees for workers and making it difficult to meet the safety requirements of live-line work. Summary of the Invention

[0004] The purpose of this application is to provide a method, device, equipment, and medium for real-time monitoring of safety distances during live-line work, in order to solve the problem of not being able to provide effective safety guarantees for workers.

[0005] A first aspect of this application provides a method for real-time monitoring of safe distances for live-line working, applied to a drone, comprising:

[0006] Acquire basic point cloud data with integrated annotation information corresponding to the target transmission line area. The annotation information includes the electrical attributes corresponding to the basic point cloud data and the voltage level corresponding to the electrical attributes. Different voltage levels correspond to different minimum safe distance specifications.

[0007] The system acquires real-time raw point cloud data corresponding to the target work area, the real-time absolute coordinates of the UAV, and the real-time attitude data of the UAV; and performs statistical filtering and radius filtering on the real-time raw point cloud data to remove noise points and isolated points caused by interference factors, thereby obtaining real-time initial point cloud data; the target work area is a specific area in the target power transmission line area that requires maintenance work.

[0008] Based on the real-time absolute coordinates and the real-time attitude data, the relative coordinates of the real-time initial point cloud data are mapped to the absolute coordinates corresponding to the target work area to obtain the real-time point cloud data of the target work area under the absolute coordinates; the real-time point cloud data is point cloud data containing the workers.

[0009] Based on the basic point cloud data with integrated annotation information and the real-time point cloud data, the point cloud of the operator is extracted to obtain the target point cloud data corresponding to the operator; the target point cloud data is the point cloud data that contains only the operator in the target work area;

[0010] Extract the point cloud edge points of the target point cloud data, and calculate the spatial distance between each edge point and each of the labeled information based on the absolute coordinates of the point cloud edge points, and determine the target spatial distance from multiple spatial distances;

[0011] The personnel are monitored in real time based on the target spatial distance.

[0012] A second aspect of this application provides a real-time monitoring device for safe distance during live-line work, applied to a drone, comprising:

[0013] The first data acquisition module is used to acquire basic point cloud data with integrated annotation information corresponding to the target transmission line area. The annotation information includes the electrical attributes corresponding to the basic point cloud data and the voltage level corresponding to the electrical attributes, wherein different voltage levels correspond to different minimum safe distance standard values.

[0014] The second data acquisition module is used to acquire real-time raw point cloud data corresponding to the target work area, the real-time absolute coordinates of the UAV, and the real-time attitude data of the UAV; and to perform statistical filtering and radius filtering on the real-time raw point cloud data to remove noise points and isolated points caused by interference factors, thereby obtaining real-time initial point cloud data; the target work area is a specific area in the target power transmission line area that needs to be inspected and maintained.

[0015] The third data acquisition module is used to map the relative coordinates of the real-time initial point cloud data to the absolute coordinates corresponding to the target work area based on the real-time absolute coordinates and the real-time attitude data, so as to obtain the real-time point cloud data of the target work area under the absolute coordinates; the real-time point cloud data is point cloud data containing the workers.

[0016] The worker point cloud extraction module is used to extract worker point clouds based on the basic point cloud data with integrated annotation information and the real-time point cloud data to obtain the target point cloud data corresponding to the worker; the target point cloud data is the point cloud data that contains only workers in the target work area;

[0017] The spatial distance calculation module is used to extract the point cloud edge points of the target point cloud data, and calculate the spatial distance between each edge point and each of the labeled information based on the absolute coordinates of the point cloud edge points, and determine the target spatial distance from multiple spatial distances;

[0018] The monitoring module is used to monitor the workers in real time based on the target spatial distance.

[0019] A third aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of the above-described method for real-time monitoring of safe distances for live-line work.

[0020] In a fourth aspect of this application, a computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements the steps of the above-described method for real-time monitoring of safe distances for live-line work.

[0021] The beneficial effects of the real-time monitoring method, device, equipment, and medium for live-line working safety distance provided in this application are as follows: This application provides a static benchmark and clear safety standards for monitoring by acquiring basic point cloud data containing electrical attributes and minimum safety distance specifications, avoiding judgment bias caused by the lack of a benchmark; after acquiring real-time data, statistical filtering and radius filtering are used to remove noise, reducing the impact of interference on subsequent calculations and improving data purity. Then, relative coordinates are mapped to absolute coordinates using real-time absolute coordinates and attitude data, making the real-time point cloud and the basic point cloud share the same coordinate system, solving the problem of relative coordinate adaptation of point cloud acquisition equipment and ensuring comparison accuracy. Subsequently, based on the two types of point clouds, a target point cloud containing only workers is selected, eliminating static background and interference from work tools, and accurately locating personnel. Edge points are extracted to calculate the target distance, focusing on the nearest potential location of personnel, and combining the potential status to determine the key distance, avoiding monitoring blind spots. Finally, real-time comparison and early warning are provided to promptly alert to risks. Therefore, this application can achieve high precision, high real-time performance, and high reliability in live-line working safety distance monitoring, effectively ensuring the safety of workers. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1A flowchart illustrating a real-time monitoring method for safe distance during live-line working, provided in an embodiment of this application;

[0024] Figure 2 This is a structural block diagram of a live-line working safety distance real-time monitoring device provided in an embodiment of this application;

[0025] Figure 3 This is a schematic block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0026] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0027] The real-time monitoring method for safe distance during live-line work provided in this embodiment can be applied to drones and is applicable to live-line work scenarios on power transmission lines. The drone is equipped with integrated monitoring equipment, which includes a first data acquisition module, a second data acquisition module, and a third data acquisition module.

[0028] The first data acquisition module is used to acquire point cloud data of the target power transmission line area, the second data acquisition module is used to acquire the absolute coordinates of the UAV, and the third data acquisition module is used to acquire the attitude data of the UAV.

[0029] In this embodiment, a Time of Flight (ToF) ranging module can be used as the first data acquisition module, a Real Time Kinematic (RTK) module can be used as the second data acquisition module, and an Inertial Measurement Unit (IMU) module can be used as the third data acquisition module.

[0030] To make the objectives, technical solutions, and advantages of this application clearer, the following description will be provided in conjunction with the accompanying drawings and specific embodiments.

[0031] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating a real-time monitoring method for safe distances during live-line work, provided in an embodiment of this application. The method can be executed by an electronic device and may include:

[0032] S101: Obtain the basic point cloud data corresponding to the integrated annotation information of the target transmission line area. The annotation information includes the electrical attributes corresponding to the basic point cloud data and the voltage level corresponding to the electrical attributes. Different voltage levels correspond to different minimum safe distance specifications.

[0033] In this embodiment, the basic point cloud data is the point cloud data corresponding to an empty scene in the target transmission line area where there are no workers or other dynamic objects. It is used as a static reference for subsequent real-time point cloud comparison. The annotation information includes two parts: first, electrical attribute annotation, which distinguishes the point cloud clusters corresponding to live conductors, fittings, etc., and the point cloud clusters corresponding to grounding conductors, towers, overhead ground wires, etc.; second, voltage level annotation, which specifies the transmission line voltage level (such as 110kV, 220kV, etc.) corresponding to the above different electrical attributes. The minimum safe distance standard value is determined by the labeled voltage level. Different voltage levels correspond to different minimum safe distance standard values, which are set with reference to the live-line working safety regulations for the corresponding voltage level.

[0034] In this embodiment, the operator can install the integrated monitoring equipment on a drone and control the drone to fly around the target transmission line area multiple times, with the flight altitude covering the entire working area. During flight, the ToF module in the integrated monitoring equipment continuously collects point cloud data of the target transmission line area, while the RTK module outputs the absolute coordinates of the drone, and the IMU module outputs the real-time attitude data of the drone (including roll angle, pitch angle, and yaw angle). The relative coordinate point cloud output by the ToF module is converted into basic point cloud data in an absolute coordinate system. This basic point cloud data does not include the operators and only covers static objects such as conductors, towers, insulators, and fittings within the target transmission line area. Labeling information is added to this basic point cloud data. First, the electrical attributes (energized components or ground potential components) corresponding to each point cloud cluster are labeled, and then the corresponding transmission line voltage level is labeled, thereby generating basic point cloud data with integrated labeling information. Subsequently, the corresponding minimum safe distance standard value can be matched according to the labeled voltage level.

[0035] S102: Acquire the real-time raw point cloud data corresponding to the target work area, the real-time absolute coordinates of the UAV, and the real-time attitude data of the UAV; and perform statistical filtering and radius filtering on the real-time raw point cloud data to remove noise points and isolated points caused by interference factors in the real-time raw point cloud data to obtain the real-time initial point cloud data; the target work area is a specific area in the target power transmission line area that needs to be inspected and maintained.

[0036] In this embodiment, the real-time raw point cloud data is the raw 3D point cloud data collected in real time by the ToF module carried by the UAV during the operation phase, which includes all scene elements such as operators, operation tools, power facilities, and ground features.

[0037] Real-time initial point cloud data is obtained by statistical filtering and radius filtering of the raw real-time point cloud data, thus removing most of the interference noise. Real-time absolute coordinates are the real-time spatial coordinates of the UAV in the absolute coordinate system at the work site, output by the RTK module on the UAV, used to provide an absolute position reference. Real-time attitude data are parameters reflecting the real-time motion state of the UAV (including roll angle, pitch angle, and yaw angle), output by the IMU module on the UAV, used to correct attitude deviations in point cloud acquisition.

[0038] In this embodiment, the drone hovers near the work area (maintaining a safe distance from the workers to avoid obstruction) and activates the monitoring equipment. The ToF ranging module collects raw point cloud data of the target work area in real time (including workers, tools, and the environment), the RTK module outputs the drone's absolute coordinates in real time, and the IMU module outputs the drone's attitude data (including roll angle, pitch angle, and yaw angle) in real time. All data is synchronized through the motherboard's high-precision clock module to ensure data consistency.

[0039] Statistical filtering is used to calculate the distance distribution between each point and its neighbors in the real-time raw point cloud data. Points whose mean distance exceeds the standard deviation are removed (e.g., setting the number of neighbors to 50 and the standard deviation factor to 1.0). Radius filtering is then used, with each real-time raw point cloud data point as the center, a radius (e.g., 0.1 meters) is set. If the number of points within this radius is less than a threshold (e.g., 5 points), the point is considered an isolated point and removed. Finally, the filtered data yields the real-time initial point cloud data.

[0040] S103: Based on real-time absolute coordinates and real-time attitude data, the relative coordinates of the real-time initial point cloud data are mapped to the absolute coordinates corresponding to the target work area to obtain the real-time point cloud data of the target work area in absolute coordinates; the real-time point cloud data is point cloud data containing the workers.

[0041] In this embodiment, real-time attitude data is a parameter collected by the IMU module on the UAV, reflecting the real-time motion state of the UAV, and is used to correct attitude deviations in point cloud acquisition. Real-time point cloud data is point cloud data formed in absolute coordinates after coordinate mapping of the initial real-time point cloud data. It includes personnel, tools, power facilities, ground features, and the environment. It is in the same coordinate system as the basic point cloud data and can be directly used for calculation and analysis.

[0042] In this embodiment, the integrated monitoring device carried by the UAV also includes a clock module, which is used to add a unified timestamp to the data acquired by the ToF module, RTK module and IMU module; based on the 100Hz sampling frequency of the IMU module, linear interpolation is performed on the 1Hz data of the RTK module (e.g., interpolating the coordinates of the RTK data at t=0.01s and t=0.02s to generate 100 sets of data); for each frame of ToF point cloud (30fps, approximately 1 frame every 33ms), the IMU+RTK interpolated data with the closest time is matched to ensure that 1 frame of point cloud corresponds to 1 set of pose data.

[0043] The real-time raw point cloud data acquired by the ToF module takes the ToF module as the origin (relative coordinates). A rotation matrix needs to be constructed based on the attitude angle acquired by the IMU module to convert the relative coordinates into the device coordinate system with the center point of the integrated monitoring device as the origin.

[0044] Real-time attitude data (roll angle) based on IMU module Pitch angle Yaw angle ), construct the rotation matrix. Multiply the rotation matrices in the order of yaw, pitch, and roll to obtain the total rotation matrix R.

[0045] Rotation matrix around the Z-axis (yaw) :

[0046]

[0047] Rotation matrix around the Y-axis (pitch) :

[0048]

[0049] Rotation matrix around the X-axis (roll) :

[0050]

[0051] Total rotation matrix R:

[0052]

[0053] The total rotation matrix R is used to describe the attitude transformation relationship between the relative coordinates of the real-time raw point cloud data acquired by the ToF module and the coordinates of the integrated monitoring device coordinate system.

[0054] Let the relative coordinates of the real-time initial point cloud data obtained after filtering be... The installation offset of the ToF module within the integrated monitoring equipment is The coordinates of the real-time initial point cloud data are then determined in a coordinate system with the center point of the integrated monitoring equipment as the origin. The calculation formula is:

[0055]

[0056] The matrix multiplication result is the coordinates after attitude correction. After superimposing the installation offset, the coordinates in the equipment coordinate system with the center of the integrated monitoring equipment as the origin are obtained.

[0057] In this embodiment, the coordinates of the integrated monitoring equipment center as the origin are converted to absolute coordinates in the geodetic coordinate system.

[0058] Extract RTK interpolated data that matches the real-time initial point cloud data to obtain the real-time absolute coordinates of the center point of the UAV integrated monitoring device. (Where the X-axis corresponds to the east, the Y-axis to the north, and the Z-axis to the elevation).

[0059] Since the coordinate axes of the integrated monitoring equipment coordinate system are completely aligned with those of the geodetic coordinate system (absolute coordinates), no additional rotation correction is required. The absolute coordinates of the real-time initial point cloud data in the geodetic coordinate system can be obtained simply by directly superimposing the coordinates of the real-time initial point cloud data in the integrated monitoring equipment coordinate system with the absolute coordinates of the center point of the integrated monitoring equipment obtained by the RTK module. The formula is:

[0060]

[0061] It achieves accurate mapping from relative coordinates to absolute coordinates, thereby obtaining real-time point cloud data of the target work area in absolute coordinates.

[0062] In this embodiment, the IMU module suffers from cumulative drift during long-term operation, leading to a gradual increase in attitude angle errors. Meanwhile, the RTK module is prone to signal loss under conditions such as tower obstruction or strong electromagnetic interference, potentially reducing its positioning accuracy to the meter level. To overcome these limitations, extended Kalman filtering can be used to fuse the RTK and IMU data, optimizing the absolute coordinates and attitude angle output of the integrated monitoring device.

[0063] S104: Extract the point cloud of the operator based on the basic point cloud data and real-time point cloud data with integrated annotation information to obtain the target point cloud data corresponding to the operator; the target point cloud data is the point cloud data that contains only the operator in the target operation area.

[0064] In this embodiment, the target point cloud data is pure point cloud data containing only the workers after point cloud filtering, reflecting the human body contours and three-dimensional spatial positions of the workers.

[0065] In this embodiment, the real-time point cloud and the integrated labeled base point cloud are differentially processed to remove overlapping static background point clouds, thus initially obtaining dynamic point cloud clusters of workers and tools. Subsequently, based on the geometric features of the tools, combined with a lightweight deep learning model, the point clouds corresponding to the tools are accurately identified and removed, ultimately obtaining target point cloud data containing only workers.

[0066] S105: Extract the edge points of the target point cloud data, and calculate the spatial distance between each edge point and each labeled information based on the absolute coordinates of the point cloud edge points, and determine the target spatial distance from multiple spatial distances.

[0067] In this embodiment, the point cloud edge points are the outermost point cloud pixels in the target point cloud data. These points are the potential locations where the worker's body is closest to a live or grounded object. The target spatial distance is a key distance selected from multiple spatial distances between the worker's point cloud edge points and each point cloud with specific annotation information (such as the minimum distance to the grounded object under equipotential conditions, and the sum of the distances to the live and grounded objects under intermediate potential conditions).

[0068] In this embodiment, edge points of the target point cloud data are extracted; based on the absolute coordinates of the edge points, the three-dimensional spatial distance between them and each labeled electrical attribute is calculated; finally, the target spatial distance is determined according to the potential state of the operator.

[0069] S106: Real-time monitoring of workers based on target spatial distance.

[0070] In this embodiment, the calculated target spatial distance is compared with the minimum safe distance specification value in the annotation information in real time. If the target spatial distance is greater than the safety specification value, only the data is transmitted back to the ground terminal equipment; if the target spatial distance is less than or equal to the safety specification value, the drone's audible and visual alarm is triggered immediately, and an alarm signal is sent to the ground terminal equipment at the same time, providing a double reminder to the operator to adjust the attitude and control distance to ensure the safety of live-line work.

[0071] As can be seen from the above, this embodiment provides a static benchmark and clear safety standards for monitoring by acquiring basic point cloud data containing electrical attributes and minimum safe distance specifications, avoiding judgment bias caused by the lack of a benchmark. After acquiring real-time data, statistical filtering and radius filtering are used to remove noise, reducing the impact of interference on subsequent calculations and improving data purity. Then, relative coordinates are mapped to absolute coordinates using real-time absolute coordinates and attitude data, making the real-time point cloud and the basic point cloud share the same coordinate system, solving the ToF relative coordinate adaptation problem and ensuring comparison accuracy. Subsequently, based on the two types of point clouds, a target point cloud containing only workers is selected, eliminating static background and interference from work tools to accurately locate personnel. Edge points are extracted to calculate target distances, focusing on the nearest potential location of personnel, and combining this with potential status to determine key distances, avoiding monitoring blind spots. Finally, real-time comparison and early warning are provided to promptly alert to risks. Therefore, this embodiment can achieve high precision, high real-time performance, and high reliability in monitoring safe distances for live-line work, effectively ensuring the safety of workers.

[0072] In one embodiment of this application, obtaining basic point cloud data with integrated annotation information corresponding to the target transmission line area includes:

[0073] The process involves acquiring raw point cloud data of the target transmission line area, the absolute coordinates of the UAV, and the attitude data of the UAV, and then fusing the raw point cloud data, absolute coordinates, and attitude data to obtain the basic point cloud data of the target transmission line area in absolute coordinates; the raw point cloud data does not include the personnel working on the project.

[0074] The basic point cloud data is sent to the terminal device so that the terminal device can add annotation information to the basic point cloud data, thus obtaining basic point cloud data with integrated annotation information.

[0075] In this embodiment, the raw point cloud data refers to the 3D point cloud data collected by the ToF module mounted on the UAV in an empty scene where there are no workers or other dynamic objects in the target transmission line area. This data contains only static objects such as conductors, towers, and insulators within the target transmission line area and does not contain any dynamic interference elements. The integrated annotation information-based basic point cloud data refers to the data where, after being sent to the terminal device, annotation information is added (including the electrical attributes corresponding to the basic point cloud data, such as point cloud clusters of live components and ground potential components; and the electrical level corresponding to these attributes, with different voltage levels corresponding to different minimum safe distance specifications). The resulting composite data, combining static point cloud data and safety monitoring standards, serves as the benchmark data for subsequent real-time point cloud comparison and safe distance judgment.

[0076] In this embodiment, the operator first installs the integrated monitoring equipment on the drone's mounting position, ensuring the equipment is securely fixed and the sensors are unobstructed. Then, the drone is controlled to fly over the target transmission line area, following a preset flight path (flying around the target tower 3-5 times, covering the entire operational range from 5 meters on the ground to 2 meters at the top of the tower). During flight, the target transmission line area is kept in an empty field, and the ToF module continuously collects raw point cloud data of static objects such as conductors, towers, insulators, and fittings within the area, with a sampling frequency set to 30fps (approximately one frame of data collected every 33 milliseconds).

[0077] While collecting raw point cloud data, the RTK module on the drone outputs the drone's absolute coordinates at a sampling frequency of 1Hz; the IMU module outputs the drone's attitude data at a sampling frequency of 100Hz.

[0078] Based on the acquired raw point cloud data, the absolute coordinates of the UAV, and the attitude data of the UAV, a fusion processing is performed. First, a unified timestamp is added to the three types of data through a clock module to achieve time synchronization. Second, a rotation matrix is ​​constructed based on the attitude data of the UAV to convert the relative coordinates of the raw point cloud data into device coordinates with the center point of the integrated monitoring device as the origin. Finally, the absolute coordinates of the UAV are superimposed to map the point cloud data in the device coordinate system to the absolute coordinate system corresponding to the target transmission line area, thus obtaining the basic point cloud data of the target transmission line area in absolute coordinates.

[0079] In this embodiment, after the operator controls the drone to land, the basic point cloud data stored in the embedded motherboard of the integrated monitoring device can be exported via USB data cable, or the basic point cloud data can be sent to the terminal device in real time via the wireless transmission module of the integrated monitoring device.

[0080] After receiving the basic point cloud data, the operator loads and displays the data using point cloud processing software, manually adding two types of annotation information: First, electrical attribute annotation: using the software's point cloud selection tool, the operator selects and marks the point cloud clusters corresponding to live components such as conductors and fittings (labeled "live body"), and the point cloud clusters corresponding to ground potential components such as poles and overhead ground wires (labeled "grounding body"). Second, voltage level annotation: referring to the "State Grid Corporation Power Safety Work Regulations (Line Section)," the operator enters the corresponding transmission line voltage level (e.g., 110kV, 220kV, etc.) for each point cloud cluster with annotated electrical attributes. Since different voltage levels correspond to different minimum safety distance specifications, the terminal equipment will automatically match the corresponding minimum safety distance specification value based on the entered voltage level (e.g., the minimum safety distance for 110kV voltage level is 1.0 meter, and the minimum safety distance for 220kV voltage level is 1.8 meters). After the annotation is completed, the terminal device automatically binds the electrical attribute annotation, voltage level annotation, and associated minimum safe distance specification value to the basic point cloud data, generating basic point cloud data with integrated annotation information.

[0081] After the terminal device completes the annotation, the operator transmits the basic point cloud data with integrated annotation information back to the embedded motherboard of the integrated monitoring device via USB data cable, or sends the data to the integrated monitoring device via wireless transmission module. After receiving the data, the integrated monitoring device stores it in a dedicated storage unit and verifies the data integrity through a data verification algorithm to ensure that the basic point cloud data with integrated annotation information is not lost or damaged, providing an accurate benchmark for real-time safe distance monitoring in subsequent operation stages.

[0082] As can be seen from the above, this embodiment obtains basic point cloud data in absolute coordinates by fusing the original point cloud data of the target transmission line area with the absolute coordinates and attitude data of the UAV, ensuring the spatial positioning accuracy of the static benchmark; then, the terminal device adds annotation information of electrical attributes and minimum safe distance specifications to the basic point cloud data to clarify the monitoring and judgment criteria; finally, the integrated and annotated data is received to provide a reliable benchmark for subsequent real-time point cloud comparison and safe distance calculation.

[0083] In one embodiment of this application, point cloud extraction of workers is performed based on basic point cloud data with integrated annotation information and real-time point cloud data to obtain target point cloud data corresponding to the workers, including:

[0084] The base point cloud data with integrated annotation information and the real-time point cloud data are subjected to three-dimensional spatial difference calculation. Static point cloud data that overlaps with the base point cloud data with integrated annotation information in the real-time point cloud data are removed to obtain dynamic point cloud data in the real-time point cloud data. The dynamic point cloud data includes point cloud data corresponding to workers and tools.

[0085] The point cloud of the operator is extracted from the dynamic point cloud data to obtain the target point cloud data corresponding to the operator.

[0086] In this embodiment, static point cloud data refers to the point cloud data corresponding to static objects (such as conductors, towers, insulators, and fittings) within the target transmission line area, included in the basic point cloud data with integrated annotation information. Their spatial positions remain fixed during the operation phase and are the portion that needs to be removed from the real-time point cloud data during the three-dimensional spatial difference calculation. Dynamic point cloud data refers to the point cloud data that does not overlap with the static point cloud data, retained from the real-time point cloud data after three-dimensional spatial difference calculation. It mainly includes the point cloud data corresponding to the operators and the point cloud data corresponding to the operating tools (such as insulating rods, insulating ropes, and potential transfer rods), whose spatial positions change dynamically with the operation process.

[0087] In this embodiment, a preset point cloud difference calculation algorithm can be invoked to preprocess the basic point cloud data and real-time point cloud data with integrated annotation information. By using an appropriate point cloud downsampling method, the amount of data of the two types of data is reduced to improve the efficiency of subsequent calculations. Then, using the basic point cloud data with integrated annotation information as the reference static set, for each point cloud pixel in the real-time point cloud data, a nearest neighbor search algorithm is used with a reasonable number of nearest neighbors and a search radius to determine whether the point cloud pixel has a spatial overlap with the point cloud in the reference static set. If the preset overlap determination condition is met, the point is determined to be static point cloud data. Finally, all static point cloud data in the real-time point cloud data is removed, and the subset of point clouds that does not overlap with the reference static set is retained, which is the dynamic point cloud data containing the point cloud data corresponding to the workers and the point cloud data corresponding to the work tools.

[0088] In this embodiment, feature extraction can be performed using a lightweight deep learning method. First, the dynamic point cloud data is converted into a point cloud feature map conforming to the model input format and input into a pre-trained lightweight point cloud segmentation model. This model has been trained on point cloud samples of personnel and tools in a live-line working scenario and has learned the geometric morphological differences between the two types of targets. Second, the model extracts the three-dimensional geometric and morphological features of each point cloud cluster through shallow convolution and attention mechanisms, and distinguishes point cloud clusters that conform to the shape and size rules of the working tools based on feature matching results; that is, the point cloud data corresponding to the working tools. Finally, the point cloud data corresponding to the working tools in the dynamic point cloud data is removed, and the remaining point cloud clusters are the target point cloud data corresponding to the workers. This lightweight model balances feature extraction accuracy and inference speed, and can run efficiently on the embedded motherboard of the UAV integrated monitoring equipment, avoiding additional computing power consumption.

[0089] As can be seen from the above, this embodiment uses three-dimensional spatial difference calculation to eliminate static point cloud data that overlaps with the basic point cloud data in the real-time point cloud data, and accurately separates the dynamic point cloud data containing workers and tools, avoiding interference from static background; then, the dynamic point cloud data is filtered to initially locate the worker point cloud.

[0090] In one embodiment of this application, the point cloud of the operator is extracted from the dynamic point cloud data to obtain the target point cloud data corresponding to the operator, including:

[0091] Bilateral filtering is applied to the dynamic point cloud data to obtain the target dynamic point cloud data;

[0092] Feature extraction is performed on the target dynamic point cloud data to obtain point cloud data corresponding to the workers and point cloud data corresponding to the work tools;

[0093] The point cloud data corresponding to the work tools and equipment is removed to obtain the target point cloud data corresponding to the workers.

[0094] During live-line work, the tools and equipment used are carried, held, or operated directly by the worker, and are considered auxiliary tools attached to the body. The two are in close spatial proximity and physical contact. This actual working relationship is reflected in point cloud data as the point cloud data corresponding to the worker and the tool / equipment are interconnected and partially overlapped, making them indistinguishable through simple spatial division. If the point cloud data corresponding to the tool / equipment is not separated and removed from the associated point cloud cluster, the subsequently extracted point cloud data will include non-human parts, affecting the accurate positioning of the worker's body edge points and the calculation of safe distances. Therefore, a specific processing procedure is needed to effectively separate the two.

[0095] In this embodiment, the target dynamic point cloud data refers to the point cloud data that has been processed by bilateral filtering to remove noise points and retain edge details. Feature extraction refers to extracting the unique geometric and morphological features of the point cloud data corresponding to the workers and the point cloud data corresponding to the work tools from the target dynamic point cloud data, so as to distinguish between the two types of point cloud data.

[0096] In this embodiment, the bilateral filtering parameters can be set to "spatial neighborhood radius of 0.08 meters, spatial distance weight standard deviation of 1.2, and intensity similarity weight standard deviation of 0.8". The bilateral filtering first searches for all neighboring points within a 0.08-meter neighborhood of each point in the dynamic point cloud data. Then, it calculates the spatial weight based on the spatial distance between the neighboring points and the target point, and calculates the intensity weight based on the difference in intensity values ​​(reflection intensity output by the ToF module) between the neighboring points and the target point (the closer the intensity, the greater the weight). Finally, it calculates the new coordinates of the target point by weighted averaging, and removes noise points with abnormal intensity and discrete spatial positions to obtain target dynamic point cloud data with clear edge contours and no noise interference.

[0097] In this embodiment, feature extraction is performed on the target dynamic point cloud data. Based on the differences in geometric shape and spatial distribution between the point cloud of workers and the point cloud of tools, the point cloud data corresponding to the workers and the point cloud data corresponding to the tools are identified and separated respectively. The point cloud of workers usually has specific human contour features, while the point cloud of tools shows the unique geometric shape of the tools, such as the slender cylindrical features of the insulating rod and the chain-like distribution features of the insulating rope.

[0098] For example, a lightweight deep learning model can be used for feature extraction. This model was trained on 1000 sets of point cloud samples of personnel and tools in live-line working scenarios and is adapted to the embedded computing power of drones. First, the target dynamic point cloud data is encoded into feature maps using 0.03-meter voxels. After inputting into the model, local geometric features are extracted through three layers of shallow convolution, and key areas are focused on using an attention mechanism: features such as the torso-limb topology and gentle curvature changes of the personnel point cloud, and features such as slender cylinders (insulating rods) and continuous chains (insulating ropes) of the tool point cloud are accurately captured. The model outputs the classification probability of each point cloud cluster (0 for personnel, 1 for tools). Points with a probability ≥0.8 are classified as tool point clouds, and those with a probability ≥0 are classified as personnel point clouds. Finally, the target dynamic point cloud data is divided into two subsets of point cloud data.

[0099] Based on the feature extraction results, the point cloud data corresponding to the work tools are removed, and the point cloud data corresponding to the workers is retained as the target point cloud data. This removal process ensures that subsequent safe distance calculations are based solely on the actual position and posture of the workers, avoiding interference from the work tools on the monitoring results.

[0100] As can be seen from the above, this embodiment first performs bilateral filtering on the dynamic point cloud data, removing noise points while preserving point cloud edge details, thus improving data purity. Then, feature extraction is used to distinguish the point cloud data of workers from those of work tools, and finally, tool point clouds are removed. This process can accurately remove interference from tools such as insulating rods and ropes, preventing tool point clouds from being misidentified as personnel point clouds, resulting in pure worker point cloud data, improving the accuracy of subsequent safe distance calculations, and reducing monitoring errors caused by tool interference.

[0101] In one embodiment of this application, after obtaining the target point cloud data corresponding to the operator, the method further includes:

[0102] The target point cloud data is filled in using a neighborhood point interpolation algorithm to obtain the filled target point cloud data.

[0103] This includes extracting the point cloud edge points from the target point cloud data and calculating the spatial distance between each edge point and each labeled information based on the absolute coordinates of the point cloud edge points, including:

[0104] Extract the edge points of the filled target point cloud data, and calculate the spatial distance between each edge point and each labeled information based on the absolute coordinates of the point cloud edge points.

[0105] In this embodiment, a voxel grid (voxel size 0.03 meters) can be used to divide the data. The number of point clouds within each voxel is counted. If the number of point clouds within a voxel is 0, and all voxels within a 3×3×3 radius around it contain valid point clouds, then the voxel is determined to be a local void region. Subsequently, a neighborhood point interpolation algorithm is called to fill the void. For each void voxel, all valid neighboring points within a 5×5×5 radius around it are searched, and the distance from each neighboring point to the center of the void voxel is calculated. Weights are set according to the reciprocal of the distance. The coordinates of the center of the void voxel are calculated by weighted average. The calculated void point coordinates are added to the target point cloud data to obtain the filled target point cloud data.

[0106] The edge points of the filled target point cloud data are extracted, and the spatial distance between each edge point and each labeled information is calculated based on the absolute coordinates of these edge points. This ensures the integrity of the point cloud data and avoids distance calculation errors caused by missing point clouds.

[0107] As can be seen from the above, this embodiment fills in local holes in the target point cloud data using a neighborhood point interpolation algorithm, solving the problem of missing personnel point clouds caused by the removal of tool point clouds, thus making the target point cloud outline more complete. Then, based on the filled point cloud, edge points are extracted and distances are calculated to ensure that the edge points accurately reflect the outermost position of the worker's body. This avoids the omission of edge points caused by holes, improves the completeness of edge point extraction, and thus ensures the accuracy of spatial distance calculation, reducing monitoring deviations caused by missing point clouds.

[0108] In one embodiment of this application, determining the target spatial distance from a plurality of spatial distances includes:

[0109] The potential status of the workers is obtained, including: equipotential status, ground potential status, and intermediate potential status.

[0110] In response to the fact that the potential state of the operator is equipotential, the minimum spatial distance between each edge point corresponding to the operator and the point cloud data corresponding to the grounding body is taken as the target spatial distance.

[0111] In response to the ground potential state of the operator, the minimum spatial distance between each edge point corresponding to the operator and the point cloud data corresponding to the charged body is taken as the target spatial distance.

[0112] In response to the operator's current potential being at an intermediate potential state, the first minimum spatial distance between each edge point corresponding to the operator and the point cloud data corresponding to the grounding body is calculated, and the second minimum spatial distance between each edge point corresponding to the operator and the point cloud data corresponding to the energized body is calculated. The sum of the first minimum spatial distance and the second minimum spatial distance is taken as the target spatial distance.

[0113] In this embodiment, the potential state refers to the potential level of the worker's body during live-line work, and is divided into three categories: equipotential state, ground potential state, and intermediate potential state. The equipotential state refers to the working state where the worker's body potential is consistent with the potential of the live conductor after establishing an electrical connection with it using tools such as a potential transfer rod. In this case, it is crucial to monitor the distance between the worker and the grounding conductor (such as a tower or overhead ground wire). The ground potential state refers to the working state where the worker stands on the ground or the grounded part of a tower (such as the tower itself), and their body potential is consistent with the ground potential. In this case, it is crucial to monitor the distance between the worker and the live conductor (such as a wire or fittings). The intermediate potential state refers to the working state where the worker is between the ground potential and the equipotential state, neither fully connected to the grounding conductor nor electrically connected to the live conductor (such as the process of entering or leaving an electric field). In this case, it is necessary to monitor the distance between the worker and the live conductor, and the distance between the worker and the grounding conductor, and the sum of the two must meet the combined clearance requirements. The first minimum spatial distance refers to the smallest distance among all spatial distances between the point cloud edge point corresponding to the worker and the point cloud data corresponding to the grounding body. It is calculated only when the worker is in an equipotential or intermediate potential state. The second minimum spatial distance refers to the smallest distance among all spatial distances between the point cloud edge point corresponding to the worker and the point cloud data corresponding to the energized body. It is calculated only when the worker is in a ground potential or intermediate potential state.

[0114] In this embodiment, the ground supervisor can manually select and send the operator's potential status to the integrated monitoring device carried by the UAV according to the operation process; the integrated monitoring device receives the potential status signal and stores it in a temporary data cache.

[0115] If the worker is in an equipotential state, the minimum distance between the worker's edge point and the grounding electrode is selected from the set of spatial distances and directly determined as the target spatial distance. For example, in 220kV equipotential work, the set of spatial distances between the worker's edge point and the grounding electrode is [1.9m, 1.85m, 1.75m, ...]. The minimum distance is 1.75m, so the target spatial distance is 1.75m, which needs to be compared with the minimum safe distance standard value of 1.8m corresponding to the grounding electrode in the integrated labeling information.

[0116] If the worker is at ground potential, the minimum distance between the worker's edge point and the live conductor is selected from the set of spatial distances and directly determined as the target spatial distance. For example, in 110kV ground potential work, the set of spatial distances between the worker's edge point and the live conductor is [1.1m, 1.05m, 0.98m, ...], with a minimum distance of 0.98m. The target spatial distance is 0.98m, which needs to be compared with the minimum safe distance standard value of 1.0m corresponding to the live conductor in the integrated labeling information.

[0117] As can be seen from the above, this embodiment determines the target spatial distance specifically based on the different potential states of the operators: for equipotential operations, the minimum distance between the operator and the grounding electrode is considered; for ground potential operations, the minimum distance between the operator and the energized electrode is considered; and for intermediate potential operations, the sum of these two distances is considered. This approach aligns with the safety monitoring needs of different operational stages, avoids ignoring the differences in potential states through uniform calculations, ensures that the determination of target spatial distances in each operational scenario complies with safety regulations, accurately covers monitoring blind spots for each potential state, and improves the professionalism and compliance of safety distance judgment.

[0118] In one embodiment of this application, real-time monitoring of workers based on target spatial distance includes:

[0119] In response to the target spatial distance being greater than the minimum safe distance specification value, the target spatial distance is sent to the terminal device;

[0120] In response to the target spatial distance being less than or equal to the minimum safe distance specification value, an alarm signal is issued and alarm information is sent to the terminal device so that the terminal device issues an alarm signal.

[0121] In this embodiment, the minimum safe distance standard value matching the potential state of the worker is extracted from the basic point cloud data with integrated annotation information, and this value is determined as the safe distance threshold. For example, in 110kV operations, if the worker is at ground potential, the minimum safe distance standard value of 1.0m corresponding to the live conductor is extracted as the safe distance threshold; in 220kV operations, if the worker is at equipotential, the minimum safe distance standard value of 1.8m corresponding to the grounding conductor is extracted as the safe distance threshold.

[0122] If the target spatial distance is greater than the safe distance threshold, the target spatial distance value, the worker's electrical potential status, and the current time are packaged and sent to the terminal device. After receiving the data, the terminal device updates the current safe distance in real time on the display interface without triggering an alarm.

[0123] If the target spatial distance is less than or equal to the safe distance threshold, the integrated monitoring device will emit a "beep" warning sound at a frequency of 2Hz and control the red flashing light to flash at a frequency of 5Hz. At the same time, an alarm message will be generated and sent to the terminal device. After receiving the alarm message, the terminal device will immediately pop up a red alarm window to display the alarm content and trigger the buzzer of the terminal device to sound an alarm. The alarm signal and alarm message will automatically stop outputting only when the operator adjusts his posture so that the target spatial distance is greater than the safe distance threshold.

[0124] As can be seen from the above, this embodiment can respond in stages according to the relationship between the target spatial distance and the minimum safe distance specification value. When it is safe, distance data is sent to the terminal equipment for real-time ground monitoring; when it is dangerous, not only does the equipment issue an alarm signal, but it also sends alarm information to the terminal equipment to trigger the terminal alarm. The dual alarm mechanism can promptly remind the operators and the ground supervisor, avoiding risk delays. Data transmission in a safe state also ensures the continuity of monitoring, effectively improving the timeliness and reliability of live-line work safety assurance.

[0125] Corresponding to the real-time monitoring method for safe distance during live-line work in the above embodiments, Figure 2 This is a structural block diagram of a real-time monitoring device for safe distances while working on live lines, provided as an embodiment of this application. For ease of explanation, only the parts relevant to the embodiment of this application are shown. References Figure 2 The live-line working safety distance real-time monitoring device 20 includes: a first data acquisition module 21, a second data acquisition module 22, a third data acquisition module 23, a worker point cloud extraction module 24, a spatial distance calculation module 25, and a monitoring module 26.

[0126] The first data acquisition module 21 is used to acquire basic point cloud data of integrated annotation information corresponding to the target transmission line area. The annotation information includes the electrical attributes corresponding to the basic point cloud data and the voltage level corresponding to the electrical attributes. Different voltage levels correspond to different minimum safe distance standard values.

[0127] The second data acquisition module 22 is used to acquire real-time raw point cloud data, real-time absolute coordinates of the UAV, and real-time attitude data of the UAV corresponding to the target operation area; and to perform statistical filtering and radius filtering on the real-time raw point cloud data to remove noise points and isolated points caused by interference factors in the real-time raw point cloud data to obtain real-time initial point cloud data; the target operation area is a specific area in the target power transmission line area that needs to be inspected and maintained.

[0128] The third data acquisition module 23 is used to map the relative coordinates of the real-time initial point cloud data to the absolute coordinates corresponding to the target work area based on the real-time absolute coordinates and real-time attitude data, so as to obtain the real-time point cloud data of the target work area in the absolute coordinates; the real-time point cloud data is point cloud data containing the workers.

[0129] The worker point cloud extraction module 24 is used to extract worker point clouds based on basic point cloud data with integrated annotation information and real-time point cloud data to obtain target point cloud data corresponding to the workers; the target point cloud data is the point cloud data that contains only workers in the target work area;

[0130] The spatial distance calculation module 25 is used to extract the edge points of the target point cloud data, and calculate the spatial distance between each edge point and each labeled information based on the absolute coordinates of the edge points, and determine the target spatial distance from multiple spatial distances.

[0131] Monitoring module 26 is used to monitor workers in real time based on target spatial distance.

[0132] In one embodiment of this application, the first data acquisition module 21 is specifically used for:

[0133] The process involves acquiring raw point cloud data of the target transmission line area, the absolute coordinates of the UAV, and the attitude data of the UAV, and then fusing the raw point cloud data, absolute coordinates, and attitude data to obtain the basic point cloud data of the target transmission line area in absolute coordinates; the raw point cloud data does not include the personnel working on the project.

[0134] The basic point cloud data is sent to the terminal device so that the terminal device can add annotation information to the basic point cloud data, thus obtaining basic point cloud data with integrated annotation information.

[0135] In one embodiment of this application, the worker point cloud extraction module 24 is specifically used for:

[0136] The base point cloud data with integrated annotation information and the real-time point cloud data are subjected to three-dimensional spatial difference calculation. Static point cloud data that overlaps with the base point cloud data with integrated annotation information in the real-time point cloud data are removed to obtain dynamic point cloud data in the real-time point cloud data. The dynamic point cloud data includes point cloud data corresponding to workers and tools.

[0137] The point cloud of the operator is extracted from the dynamic point cloud data to obtain the target point cloud data corresponding to the operator.

[0138] In one embodiment of this application, the worker point cloud extraction module 24 is further used for:

[0139] Bilateral filtering is applied to the dynamic point cloud data to obtain the target dynamic point cloud data;

[0140] Feature extraction is performed on the target dynamic point cloud data to obtain point cloud data corresponding to the workers and point cloud data corresponding to the work tools;

[0141] The point cloud data corresponding to the work tools and equipment is removed to obtain the target point cloud data corresponding to the workers.

[0142] In one embodiment of this application, the live-line working safety distance real-time monitoring device 20 further includes: a data filling module, specifically used for:

[0143] The target point cloud data is filled in using a neighborhood point interpolation algorithm to obtain the filled target point cloud data.

[0144] This includes extracting the point cloud edge points from the target point cloud data and calculating the spatial distance between each edge point and each labeled information based on the absolute coordinates of the point cloud edge points, including:

[0145] Extract the edge points of the filled target point cloud data, and calculate the spatial distance between each edge point and each labeled information based on the absolute coordinates of the point cloud edge points.

[0146] In one embodiment of this application, the spatial distance calculation module 25 is specifically used for:

[0147] The potential status of the workers is obtained, including: equipotential status, ground potential status, and intermediate potential status.

[0148] In response to the fact that the potential state of the operator is equipotential, the minimum spatial distance between each edge point corresponding to the operator and the point cloud data corresponding to the grounding body is taken as the target spatial distance.

[0149] In response to the ground potential state of the operator, the minimum spatial distance between each edge point corresponding to the operator and the point cloud data corresponding to the charged body is taken as the target spatial distance.

[0150] In response to the operator's current potential being at an intermediate potential state, the first minimum spatial distance between each edge point corresponding to the operator and the point cloud data corresponding to the grounding body is calculated, and the second minimum spatial distance between each edge point corresponding to the operator and the point cloud data corresponding to the energized body is calculated. The sum of the first minimum spatial distance and the second minimum spatial distance is taken as the target spatial distance.

[0151] In one embodiment of this application, the monitoring module 26 is specifically used for:

[0152] In response to the target spatial distance being greater than the minimum safe distance specification value, the target spatial distance is sent to the terminal device;

[0153] In response to the target spatial distance being less than or equal to the minimum safe distance specification value, an alarm signal is issued and alarm information is sent to the terminal device so that the terminal device issues an alarm signal.

[0154] See Figure 3 , Figure 3 This is a schematic block diagram of an electronic device provided according to an embodiment of this application. Figure 3 The electronic device 300 in this embodiment may include one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memories 304 store computer programs, including program instructions. The processors 301 execute the program instructions stored in the memories 304. Specifically, the processors 301 are configured to invoke the program instructions to perform the functions of the modules in the aforementioned device embodiments, for example... Figure 2 The functions of the first data acquisition module 21, the second data acquisition module 22, the third data acquisition module 23, the worker point cloud extraction module 24, the spatial distance calculation module 25, and the monitoring module 26 are shown.

[0155] It should be understood that, in the embodiments of this application, the processor 301 may be a central processing unit (CPU), but it may also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0156] Input device 302 may include a touchpad, a fingerprint sensor (for collecting the user's fingerprint information and fingerprint orientation information), a microphone, etc., and output device 303 may include a display (LCD, etc.), a speaker, etc.

[0157] The memory 304 may include read-only memory and random access memory, and provides instructions and data to the processor 301. A portion of the memory 304 may also include non-volatile random access memory. For example, the memory 304 may also store information such as basic point cloud data or real-time point cloud data with integrated annotation information.

[0158] In specific implementations, the processor 301, input device 302, and output device 303 described in the embodiments of this application can execute the implementation method described in the real-time monitoring method for safe distance of live-line work provided in the embodiments of this application, or they can execute the implementation method of the electronic device described in the embodiments of this application, which will not be repeated here.

[0159] In another embodiment of this application, a computer-readable storage medium is provided. This computer-readable storage medium stores a computer program, which includes program instructions. When executed by a processor, the program instructions implement all or part of the processes in the methods described above. Alternatively, the computer program can instruct related hardware to complete the process. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include any entity or device capable of carrying computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0160] The computer-readable storage medium can be an internal storage unit of the electronic device in any of the foregoing embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device. Furthermore, the computer-readable storage medium can include both internal and external storage units of the electronic device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0161] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0162] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the electronic devices and units described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0163] In the several embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the couplings or direct couplings or communication connections shown or discussed may be indirect couplings or communication connections through some interfaces or units, or they may be electrical, mechanical, or other forms of connection.

[0164] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of this application, depending on actual needs.

[0165] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0166] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for real-time monitoring of safe distances for live-line working, characterized in that, Applied to drones, including: Acquire basic point cloud data with integrated annotation information corresponding to the target transmission line area. The annotation information includes the electrical attributes corresponding to the basic point cloud data and the voltage level corresponding to the electrical attributes. Different voltage levels correspond to different minimum safe distance specifications. The system acquires real-time raw point cloud data corresponding to the target work area, the real-time absolute coordinates of the UAV, and the real-time attitude data of the UAV; and performs statistical filtering and radius filtering on the real-time raw point cloud data to remove noise points and isolated points caused by interference factors, thereby obtaining real-time initial point cloud data; the target work area is a specific area in the target power transmission line area that requires maintenance work; the coordinate system corresponding to the basic point cloud data is the absolute coordinate system corresponding to the target work area. Based on the real-time absolute coordinates and the real-time attitude data, the relative coordinates of the real-time initial point cloud data are mapped to the absolute coordinates corresponding to the target work area to obtain the real-time point cloud data of the target work area under the absolute coordinates; the real-time point cloud data is point cloud data containing the workers. Based on the basic point cloud data with integrated annotation information and the real-time point cloud data, the point cloud of the operator is extracted to obtain the target point cloud data corresponding to the operator; the target point cloud data is the point cloud data that contains only the operator in the target work area; Extract the point cloud edge points of the target point cloud data, and calculate the spatial distance between each edge point and each of the labeled information based on the absolute coordinates of the point cloud edge points, and determine the target spatial distance from multiple spatial distances; The operators are monitored in real time based on the target spatial distance; The step of extracting the worker's point cloud from the base point cloud data based on the integrated annotation information and the real-time point cloud data to obtain the target point cloud data corresponding to the worker includes: The basic point cloud data with integrated annotation information is subjected to three-dimensional spatial difference calculation with the real-time point cloud data. Static point cloud data that overlaps with the basic point cloud data with integrated annotation information in the real-time point cloud data is removed to obtain dynamic point cloud data in the real-time point cloud data. The dynamic point cloud data includes point cloud data corresponding to workers and work tools. The dynamic point cloud data is used to extract the point cloud data of the workers to obtain the target point cloud data corresponding to the workers.

2. The method for real-time monitoring of safe distance for live-line working as described in claim 1, characterized in that, The basic point cloud data for obtaining integrated annotation information corresponding to the target transmission line area includes: The original point cloud data of the target transmission line area, the absolute coordinates of the UAV, and the attitude data of the UAV are acquired, and the original point cloud data, the absolute coordinates, and the attitude data are fused to obtain the basic point cloud data of the target transmission line area under the absolute coordinates; the original point cloud data is the point cloud data that does not include the operators. The basic point cloud data is sent to a terminal device so that the terminal device can add annotation information to the basic point cloud data, thereby obtaining basic point cloud data with integrated annotation information.

3. The method for real-time monitoring of safe distance for live-line working as described in claim 1, characterized in that, The step of extracting the worker point cloud from the dynamic point cloud data to obtain the target point cloud data corresponding to the worker includes: The dynamic point cloud data is subjected to bilateral filtering to obtain the target dynamic point cloud data; Feature extraction is performed on the target dynamic point cloud data to obtain point cloud data corresponding to the workers and point cloud data corresponding to the work tools; the point cloud data corresponding to the work tools is removed to obtain the target point cloud data corresponding to the workers.

4. The method for real-time monitoring of safe distance for live-line working as described in claim 1 or 3, characterized in that, After obtaining the target point cloud data corresponding to the operator, the method further includes: The target point cloud data is filled in using a neighborhood point interpolation algorithm to obtain the filled target point cloud data. The step of extracting the point cloud edge points of the target point cloud data and calculating the spatial distance between each edge point and each piece of labeled information based on the absolute coordinates of the point cloud edge points includes: Extract the edge points of the filled target point cloud data, and calculate the spatial distance between each edge point and each of the labeled information based on the absolute coordinates of the edge points.

5. The method for real-time monitoring of safe distance for live-line working as described in claim 1, characterized in that, Determining the target spatial distance from multiple spatial distances includes: The potential state of the operator is obtained, including: equipotential state, ground potential state, and intermediate potential state; In response to the fact that the potential state of the operator is equipotential, the minimum spatial distance between each edge point corresponding to the operator and the point cloud data corresponding to the grounding body is taken as the target spatial distance. In response to the fact that the potential state of the operator is ground potential, the minimum spatial distance between each edge point corresponding to the operator and the point cloud data corresponding to the charged body is taken as the target spatial distance. In response to the operator being in an intermediate potential state, the first minimum spatial distance between each edge point corresponding to the operator and the point cloud data corresponding to the grounding body is calculated, and the second minimum spatial distance between each edge point corresponding to the operator and the point cloud data corresponding to the charged body is calculated. The sum of the first minimum spatial distance and the second minimum spatial distance is taken as the target spatial distance.

6. The method for real-time monitoring of safe distance for live-line working as described in claim 1, characterized in that, The real-time monitoring of the workers based on the target spatial distance includes: In response to the target spatial distance being greater than the minimum safe distance specification value, the target spatial distance is sent to the terminal device; In response to the target spatial distance being less than or equal to the minimum safe distance specification value, an alarm signal is issued and alarm information is sent to the terminal device, so that the terminal device issues the alarm signal.

7. A real-time monitoring device for safe distance during live-line working, characterized in that, Applied to drones, including: The first data acquisition module is used to acquire basic point cloud data with integrated annotation information corresponding to the target transmission line area. The annotation information includes the electrical attributes corresponding to the basic point cloud data and the voltage level corresponding to the electrical attributes, wherein different voltage levels correspond to different minimum safe distance standard values. The second data acquisition module is used to acquire real-time raw point cloud data corresponding to the target operation area, the real-time absolute coordinates of the UAV, and the real-time attitude data of the UAV; and to perform statistical filtering and radius filtering on the real-time raw point cloud data to remove noise points and isolated points caused by interference factors, thereby obtaining real-time initial point cloud data; the target operation area is the area in the target power transmission line area; the coordinate system corresponding to the basic point cloud data is the absolute coordinate system corresponding to the target operation area; The third data acquisition module is used to map the relative coordinates of the real-time initial point cloud data to the absolute coordinates corresponding to the target work area based on the real-time absolute coordinates and the real-time attitude data, so as to obtain the real-time point cloud data of the target work area under the absolute coordinates; the real-time point cloud data is point cloud data containing the workers. The worker point cloud extraction module is used to extract worker point clouds based on the basic point cloud data with integrated annotation information and the real-time point cloud data to obtain the target point cloud data corresponding to the worker; the target point cloud data is the point cloud data that contains only workers in the target work area; The spatial distance calculation module is used to extract the point cloud edge points of the target point cloud data, and calculate the spatial distance between each edge point and each of the labeled information based on the absolute coordinates of the point cloud edge points, and determine the target spatial distance from multiple spatial distances; The monitoring module is used to monitor the workers in real time based on the target spatial distance; Specifically, the worker point cloud extraction module is used to perform three-dimensional spatial difference calculation between the basic point cloud data with integrated annotation information and the real-time point cloud data, and remove static point cloud data in the real-time point cloud data that overlaps with the basic point cloud data with integrated annotation information, to obtain dynamic point cloud data in the real-time point cloud data; the dynamic point cloud data includes point cloud data corresponding to workers and work tools. The dynamic point cloud data is used to extract the point cloud data of the workers to obtain the target point cloud data corresponding to the workers.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the real-time monitoring method for safe distance during live-line work as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the real-time monitoring method for safe distance during live-line work as described in any one of claims 1 to 6.

Images