Method and System for Measuring the Distance of Hazard Sources on Transmission Lines Based on the Fusion of Point Cloud Spatial Information

Abstract

This application relates to the technical field of hazard distance measurement, and provides a method and system for measuring the distance of hazards on transmission lines based on the fusion of point cloud spatial information, including: collecting point cloud data of the target transmission line through a laser point cloud device; collecting images through a monitoring device to obtain real-time monitoring images; the backend server receives the monitoring images and frames the region of interest to locate the precise measurement start area; filtering the point cloud data according to the start area to obtain spatial point cloud analysis data; registering the point cloud analysis data with the monitoring images to generate a spatial information alignment model; measuring the distance between the target hazard and the transmission line in the model, and outputting the quantified data of the dangerous distance. This application solves the technical problem that existing transmission line monitoring devices cannot accurately locate hazards due to the lack of accurate distance measurement and real-time dynamic recognition of the external environment, and realizes the identification and distance measurement of hazards through laser point cloud devices and image monitoring devices, reducing the accident rate.

Description

Technical Field

[0001] This application relates to the technical field of hazard range measurement, and particularly to a method and system for measuring the distance of transmission line hazards based on the fusion of point cloud spatial information. Background Art

[0002] With the continuous expansion and complication of the power transmission network, the safety of transmission lines has become increasingly important. Transmission lines usually pass through complex terrains such as mountains, forests, and rivers. External environmental factors such as tree growth, building construction, and natural disasters may pose threats to the operation of the lines, resulting in faults such as line tripping and short circuits. These external hazards are often difficult to identify and measure in a timely manner, affecting the safety and stability of transmission lines. Most of the existing transmission line monitoring devices collect images and perform simple environmental monitoring through cameras or sensors installed on transmission towers. However, traditional monitoring systems usually cannot achieve accurate range measurement functions, and it is difficult to accurately locate and evaluate external hazards. Especially when facing dynamic hazards, their response capabilities are weak. Summary of the Invention

[0003] This application provides a method and system for measuring the distance of transmission line hazards based on the fusion of point cloud spatial information, aiming to solve the technical problem that existing transmission line monitoring devices cannot accurately locate and timely warn of hazards due to the lack of accurate range measurement and real-time dynamic recognition of the external environment.

[0004] In view of the above problems, this application provides a method and system for measuring the distance of transmission line hazards based on the fusion of point cloud spatial information.

[0005] In the first aspect disclosed in this application, a method for measuring the distance of transmission line hazards based on the fusion of point cloud spatial information is provided. The method includes: real-time collecting point cloud data of a target transmission line through a laser point cloud device to obtain real-time point cloud data; collecting an image of the target transmission line through a target monitoring device to obtain a real-time monitoring image, where the target monitoring device and the laser point cloud device are pre-installed on a transmission tower connected to the target transmission line; a backend server receives the real-time monitoring image transmitted back by the target monitoring device and performs region of interest (ROI) bounding based on the real-time monitoring image to locate an accurate measurement start area, where the target monitoring device is indirectly connected to the backend server through a wireless transmission relay module, and the accurate measurement start area covers the target hazard; the backend server filters the real-time point cloud data according to the accurate measurement start area to obtain spatial point cloud analysis data; the backend server registers the spatial point cloud analysis data and the real-time monitoring image to obtain a spatial information alignment model; and performs spatial distance measurement between the target hazard and the target transmission line in the spatial information alignment model to output quantified hazard distance data.

[0006] Another aspect disclosed in this application provides a transmission line hazard distance measurement system based on point cloud spatial information fusion. The system includes: a point cloud data acquisition unit: real-time point cloud data of a target transmission line is acquired through a laser point cloud device to obtain real-time point cloud data; an image acquisition unit: an image of the target transmission line is acquired through a target monitoring device to obtain a real-time monitoring image, wherein the target monitoring device and the laser point cloud device are pre-installed on a transmission tower mechanically connected to the target transmission line; a region selection unit: the backend server receives the real-time monitoring image transmitted back by the target monitoring device and performs region of interest selection based on the real-time monitoring image to locate an accurate measurement start area, wherein the target monitoring device is indirectly connected to the backend server through a wireless transmission relay module, and wherein the accurate measurement start area covers the target hazard; a spatial data filtering unit: the backend server filters the real-time point cloud data according to the accurate measurement start area to obtain spatial point cloud analysis data; a data registration unit: the backend server registers the spatial point cloud analysis data and the real-time monitoring image to obtain a spatial information alignment model; a spatial distance measurement unit: spatial distance measurement between the target hazard and the target transmission line is performed in the spatial information alignment model, and quantified hazard distance data is output.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] The above-mentioned transmission line hazard distance measurement method based on point cloud spatial information fusion acquires real-time point cloud data of a target transmission line through a laser point cloud device, and at the same time uses a target monitoring device to acquire a real-time monitoring image of the transmission line. These devices are all pre-installed on a transmission tower connected to the transmission line to form a data acquisition basis. The real-time monitoring image is transmitted to the backend server through a wireless transmission relay module. The backend server analyzes the image, selects the region of interest, and determines an accurate measurement start area covering the hazard. Subsequently, the backend server filters the real-time point cloud data according to the determined accurate measurement start area, only retaining the point cloud data related to the measurement start area to generate spatial point cloud analysis data. Then, the server registers the point cloud analysis data and the monitoring image to generate a spatial information alignment model. Through this model, the spatial distance between the hazard and the transmission line can be accurately measured, and quantified hazard distance data is output, providing scientific support for the safety assessment and risk management of the transmission line.

[0009] The above description is only an overview of the technical solution of this application. In order to understand the technical means of this application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of this application more obvious and understandable, the specific implementation manners of this application are specifically exemplified below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for use in the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0011] Figure 1 It is a schematic flowchart of a method for measuring the distance of a dangerous source of a transmission line based on the fusion of point cloud spatial information in an embodiment.

[0012] Figure 2 It is an architecture diagram of a system for measuring the distance of a dangerous source of a transmission line based on the fusion of point cloud spatial information in an embodiment.

[0013] Description of reference numerals: Point cloud data acquisition unit 1, image acquisition unit 2, region selection unit 3, spatial data filtering unit 4, data registration unit 5, spatial distance measurement unit 6. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0014] By providing a method and system for measuring the distance of a dangerous source of a transmission line based on the fusion of point cloud spatial information in the embodiments of this application, the technical problem that existing transmission line monitoring devices cannot accurately locate and timely warn of dangerous sources due to the lack of accurate distance measurement and real-time dynamic recognition of the external environment is solved.

[0015] The following will clearly and completely describe the technical solutions in the embodiments of this application with reference to the drawings in the embodiments of this application. Obviously, the described embodiments are only some embodiments of this application, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments in this application belong to the scope protected by this application.

[0016] It should be noted that the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units does not necessarily have to be limited to those clearly listed steps or units, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices.

[0017] Embodiment 1, as Figure 1As shown, this application provides a method for measuring the distance of dangerous sources of transmission lines based on the fusion of point cloud spatial information. The method includes:

[0018] The laser point cloud device is used to collect real-time point cloud data of the target transmission line.

[0019] In the embodiment of this application, the system terminal scans the environment around the target transmission line through the laser point cloud device. The laser point cloud device emits high-frequency laser beams and receives their reflected signals to capture the spatial information of the transmission line and its surrounding objects in real time. These signals are converted into point cloud data by the built-in processing unit of the device and transmitted to the system terminal to form real-time point cloud data. Each point in the real-time point cloud data contains specific three-dimensional coordinate information (such as X, Y, Z positions). These point cloud data can accurately reflect the shape, position, and spatial distribution of the transmission line and its surrounding environment, providing a basis for subsequent analysis and processing. During the acquisition process, the laser point cloud device operates continuously with high precision and high frequency to ensure that the captured spatial data is updated in real time, thereby reflecting the state of the transmission line and potential dangerous sources around it in real time.

[0020] The target monitoring device is used to collect real-time monitoring images of the target transmission line, where the target monitoring device and the laser point cloud device are pre-installed on the transmission tower mechanically connected to the target transmission line.

[0021] In one embodiment, the system terminal captures and collects image data around the transmission line in real time through the target monitoring device according to a preset frequency. The target monitoring device and the laser point cloud device are both pre-installed on the transmission tower, and the transmission tower is mechanically connected to the target transmission line. These two devices are usually in fixed positions on the transmission tower and can continuously monitor the operating conditions of the transmission line. Through the real-time monitoring images captured by the target monitoring device, the real-time environmental information around the transmission line can be obtained, and whether there are potential dangerous sources such as trees, buildings, or other objects that may threaten the safety of the line can be monitored. These images will then be sent to the backend server for processing and analysis to further determine the area that requires precise measurement.

[0022] Furthermore, before this application provides collecting images of the target transmission line through the target monitoring device, it includes:

[0023] Interactively obtain the grid basic data of the target transmission line; conduct a risk assessment of the transmission line based on the grid basic data to obtain a transmission line risk number table; preset the identification frequencies of multiple levels of hazard sources; after obtaining the target risk level of the target transmission line by calling from the transmission line risk number table, use the target risk level to traverse the identification frequencies of multiple levels of hazard sources to configure the image acquisition frequency of the target monitoring device; the target monitoring device performs intermittent image acquisition on the target transmission line with the image acquisition frequency as a constraint.

[0024] Preferably, the reserved port of the system terminal establishes a connection with the backend server to interactively obtain the grid basic data related to the target transmission line. These data include, but are not limited to, the grid topology structure, line attribute data, etc. These data are the basis for subsequent risk assessment and analysis. Subsequently, analyze the obtained grid basic data, and combine with the transmission risk evaluation function to evaluate the risk of the transmission line. The evaluation criteria include, but are not limited to, factors such as the length of the line, the surrounding environment, and the historical failure frequency. The system terminal calculates the risk index of each transmission line by comprehensively considering these factors and generates a transmission line risk number table, listing the risk levels of different lines. In addition, the system terminal also sets multiple hazard source identification frequencies according to historical experience and expert suggestions, and each hazard source identification frequency corresponds to a risk level. For example, for a first-level risk area (such as the transmission line risk number is 0.8 to 1), set a higher identification frequency (acquire 5 images per second). These frequencies are predefined and classified according to different risk levels to ensure that high-risk areas are monitored more frequently. After that, the system terminal extracts the transmission line risk numbers in the transmission line risk number table, performs matching of the corresponding risk levels, and then traverses the preset identification frequencies of multiple levels of hazard sources according to the matching results to determine the appropriate image acquisition frequency. For example, if the risk level of a certain line is the first level, the system terminal will select the corresponding acquisition frequency and configure the image acquisition parameters of the target monitoring device to ensure that it continuously monitors the transmission line at the set time interval. After configuring the image acquisition frequency, the target monitoring device performs intermittent image acquisition on the transmission line according to the set frequency interval. For example, the device is activated 5 times per second, takes images of the surrounding environment of the line, and transmits these images to the backend server for further analysis. This intermittent acquisition method can effectively save device resources while ensuring that high-risk areas are adequately monitored.

[0025] Furthermore, this application provides grid basic data, including:

[0026] The grid basic data includes grid topology data, line attribute data, system operation data, and historical maintenance data.

[0027] Optionally, the grid basic data of the target transmission line includes, but is not limited to, grid topology data, line attribute data, system operation data, and historical maintenance data. Among them, the grid topology data describes the physical layout and structure of the entire transmission line, covering the spatial position and connection relationship of the line. It includes the connection conditions of the transmission line with each node. For example, the configuration of key nodes such as substations and switch stations, and the electrical connection methods between each node. Through these data, the specific position of each transmission line in the grid and the topological relationship with adjacent lines can be clarified, facilitating the analysis of the power outage scope. The line attribute data provides the key technical parameters of each transmission line. These parameters include the physical length of the line, the type of conductor (such as ACSR aluminum-clad steel core aluminum stranded wire, AAAC all-aluminum alloy conductor, etc.), the cross-sectional area of the conductor, and the electrical impedance characteristics of the conductor. These attributes play an important role in evaluating the impedance characteristics of the transmission line. The system operation data includes the rated current value and the maximum load capacity of each transmission line. The rated current represents the current that the line can safely transmit under normal operating conditions, while the maximum load capacity refers to the highest load that the line can carry under extreme conditions. These data are used to evaluate the transmission capacity of the line and help determine whether there is an overload risk. The historical maintenance data records the relevant information of each transmission line during the past operation and maintenance process, including the frequency of historical faults, the repair time required for each fault, the measures taken during maintenance, and the remaining service life of the line. By analyzing these data, the health status and maintenance records of the line can be understood, helping to predict the possible future fault risks.

[0028] Furthermore, this application provides a method for performing risk assessment on a transmission line based on the grid basic data to obtain a risk number table of the transmission line, including:

[0029] Performing power outage scope analysis based on the grid topology data to obtain the power outage node ratios of multiple transmission lines; extracting multiple line length parameters of the multiple transmission lines from the line attribute data, and performing impedance calculation on the multiple transmission lines based on the line attribute data to obtain multiple line impedances; performing transmission capacity evaluation on the multiple transmission lines based on the system operation data to obtain multiple line transmission capacities; calling multiple historical operation and maintenance frequencies, multiple remaining service lives, and multiple line operation cycles of the multiple transmission lines from the historical maintenance data.

[0030] Optionally, in the power outage scope analysis, the system terminal first identifies the nodes directly connected to each transmission line. These nodes usually include substations, switch stations, and distribution nodes. Each node may be indirectly connected to more downstream nodes through other transmission lines. Subsequently, the directly affected power outage nodes immediately after the outage of each line are identified. These nodes are power facilities directly connected to the faulty line, such as substations or switch stations. When this line is out of service, these nodes cannot obtain power. Then, the indirectly affected power outage nodes caused by the directly affected power outage nodes are analyzed. For example, a substation may supply power to multiple downstream nodes. When this substation loses power due to the faulty line, these downstream nodes will also be indirectly powered off. The system terminal traces these indirectly affected nodes according to the topological structure, gradually expanding the power outage scope until all possible affected nodes are found. After that, the number of power outage nodes on each transmission line is counted, including directly and indirectly affected nodes, and then the ratio of the number of power outage nodes counted on each transmission line to the total number of nodes associated with this line is calculated to obtain the power outage node ratios of multiple transmission lines.

[0031] In the impedance calculation, the system terminal extracts parameters such as the conductor type, line length, and cross-sectional area of each transmission line from the line attribute data, then obtains the corresponding conductor resistivity according to the conductor type of each transmission line, and calculates the ratio of the product of the conductor resistivity and the corresponding line length of each transmission line to the cross-sectional area of the corresponding conductor to obtain the resistance value of each transmission line. Subsequently, the reactance value is calculated according to the operating frequency of each transmission line and the inductance of this line. The calculation method is ; where X is the reactance value; f is the operating frequency of the power grid, generally 50Hz or 60Hz; L is the inductance of the line, and the inductance is related to the geometric structure and distance of the conductor. After that, the system terminal multiplies the reactance value by the imaginary unit and then adds the product to the resistance value to obtain the line impedances of multiple transmission lines.

[0032] In the transmission capacity evaluation, the system terminal uses the system operation data of each transmission line to evaluate its transmission capacity. This data includes the rated current and maximum load capacity of the line. The rated current refers to the safe transmission current of the line under normal operating conditions, and the maximum load capacity is the highest current value that the line can withstand in a short period of time. Through this data, the system terminal calculates the transmission capacity according to the current service requirements using the rated current or maximum load capacity to obtain the line transmission capacities of multiple transmission lines. The calculation method is ; where P is the transmission capacity; V is the voltage level of the line, such as 110kV, 220kV, 500kV, etc.; I is the rated current or maximum load capacity. When the service requirement is for normal operation, the rated current is used. When the service requirement is for emergency or extreme conditions, the maximum load capacity is used; is the power factor, usually between 0.8 and 1, and 0.9 can be taken by default.

[0033] In addition, the system terminal also extracts the operation and maintenance history records of each transmission line from the historical maintenance data, including the historical operation and maintenance frequency, remaining service life, and line operation cycle. The historical operation and maintenance frequency reflects the health status of the line, and the remaining service life is used to show the time window when the line may fail in the future. This line operation cycle is the total time since the line was put into operation.

[0034] Pre-build a transmission risk evaluation function, and the transmission risk evaluation function is as follows: ; By synchronizing the multiple power outage node ratios, multiple line length parameters, multiple line impedances, multiple line transmission capacities, multiple historical operation and maintenance frequencies, multiple remaining service lives, and multiple line operation cycles of the multiple transmission lines to the transmission risk evaluation function, multiple transmission risk coefficients are obtained; The multiple transmission lines and multiple transmission risk coefficients are associated and stored to form the transmission line risk number table.

[0035] Optionally, in order to quantify the transmission risk, the system terminal pre-builds a transmission risk evaluation function, which obtains the comprehensive risk value of the line by considering multiple key parameters affecting the safety and stability of the transmission line. The specific transmission risk evaluation function is as follows: ; where is the transmission risk coefficient. The larger the transmission risk coefficient, the higher the risk of the transmission line; P is the power outage node ratio, which represents the proportion of the number of transmission nodes directly or indirectly affected when the transmission line fails; L is the line length parameter, which represents the physical length of the transmission line; Z is the line impedance, which represents the resistance and reactance characteristics of the transmission line; C is the line transmission capacity, which represents the maximum power that the transmission line can transmit under rated conditions or limit conditions; M is the historical operation and maintenance frequency, which represents the frequency of failures and maintenance of the transmission line in history; R is the remaining service life, which represents the time that the transmission line is expected to continue to operate; T is the line operation cycle, which represents the length of time since the line was put into operation. The system terminal inputs the data such as the power outage node ratio, line length, impedance, transmission capacity, historical operation and maintenance frequency, remaining service life, and operation cycle of each transmission line obtained by analysis into the risk evaluation function, calculates the risk coefficient of each line. Then, each transmission line and its corresponding risk coefficient are associated and stored to generate a transmission line risk number table. This number table can help the system terminal quickly identify high-risk lines for key monitoring.

[0036] The backend server receives the real-time monitoring image transmitted back by the target monitoring device, and based on the real-time monitoring image, it selects the region of interest to locate the precise measurement start area. Among them, the target monitoring device is indirectly connected to the backend server through a wireless transmission relay module, and the precise measurement start area covers the target hazard source.

[0037] In one embodiment, the system terminal transmits the real-time monitoring image transmitted back by the target monitoring device to the backend server through a wireless transmission relay module. After receiving this data, the backend server will activate the pre-deployed two-dimensional hazard source recognition model to select the region of interest and locate the precise measurement start area. Each transmission tower is equipped with a wireless transmission relay module, which collects and transmits the data collected by cameras and sensors through a low-power wireless network, such as LoRa or NB-IoT. During the data transmission process, the wireless transmission relay module not only plays the role of data transmission, but also can perform preliminary processing on the collected data. For example, image data can be compressed to reduce the occupancy of the transmission bandwidth, and at the same time, sensor data can be filtered to ensure that only high-quality or useful information is transmitted to the server. This preprocessing mechanism can improve the efficiency and reliability of data transmission, especially in the case of limited bandwidth, ensuring the integrity and real-time nature of the data.

[0038] Furthermore, this application provides that the backend server receives the real-time monitoring image transmitted back by the target monitoring device, and based on the real-time monitoring image, it selects the region of interest to locate the precise measurement start area, including:

[0039] Pre-deploy a two-dimensional hazard source recognition model in the backend server, where the two-dimensional hazard source recognition model includes a cascaded hazard source selection layer and a hazard distance recognition layer; use the hazard source selection layer of the two-dimensional hazard source recognition model to perform hazard source recognition and selection on the real-time monitoring image. If the real-time monitoring image includes the target hazard source, the hazard source selection layer outputs a real-time selected image; based on the real-time selected image, activate the hazard distance recognition layer to perform image acquisition distance analysis and output real-time distance parameters; perform spatial positioning according to the real-time selected image and the real-time distance parameters to obtain the precise measurement start area.

[0040] Optionally, in the backend server, a two-dimensional hazard identification model is pre-arranged in the system terminal. This model is composed of a hazard bounding box layer and a hazard distance identification layer connected in series, and both are constructed based on the CNN network. The hazard bounding box layer is responsible for identifying and bounding potential hazards in the real-time monitoring image. If the real-time monitoring image contains the target hazard, the bounding box layer will output a real-time image with the bounding box area (i.e., ROI, region of interest). These areas represent potential hazards and are used for further analysis. Subsequently, based on the output real-time bounding box image, the system terminal activates the hazard distance identification layer in the two-dimensional hazard identification model to perform distance analysis on the ROI in the image and calculate the actual distance between the hazard and the monitoring device. In this process, the hazard distance identification layer will output a real-time distance parameter to quantify the distance between the position of the hazard in the image and the monitoring device. Then, based on the ROI in the real-time bounding box image and the obtained real-time distance parameter, combined with the position information of the monitoring device, the system terminal performs spatial positioning through geometric projection, triangulation or other spatial positioning algorithms to calculate the specific position of the hazard in the real world. Taking geometric projection as an example, based on the installation height and field of view angle of the monitoring device, the system terminal calculates the horizontal distance and vertical height between the hazard and the power line through geometric projection. Then, using the installation height of the monitoring device as a reference and combined with the field of view angle of the monitoring device, the ROI area identified in the image is mapped to the real space. In this way, the actual position of the hazard can be determined by geometric projection. Then, the actual position of the determined hazard is compared with the position of the power line to determine the distance between the hazard and the power line. Then, the distance between the hazard and the power line is compared with the preset safety distance. If the distance between the hazard and the power line is less than the preset safety distance, the system terminal will mark the area where the hazard is located as the precise measurement start area. This start area is a region defined based on a two-dimensional plane and is used to represent the position of the hazard that needs further monitoring and measurement. This start area provides an accurate target for subsequent operations such as laser ranging.

[0041] Furthermore, the present application provides a two-dimensional hazard identification model, including:

[0042] Extract the device layout characteristics and device model information of the target monitoring device; extract the line layout characteristics of the target transmission line based on the power grid basic data; using the device layout characteristics, device model information and line layout characteristics as screening constraints, collect multiple sample dangerous intrusion images, wherein the multiple sample dangerous intrusion images include multiple sample hazard source bounding box identifiers and multiple sample hazard source distance identifiers; use the multiple sample dangerous intrusion images and multiple sample hazard source bounding box identifiers as training data, and construct the hazard source bounding box layer based on the CNN network; use the multiple sample dangerous intrusion images and multiple sample hazard source distance identifiers as training data, and construct the dangerous distance recognition layer based on the CNN network; complete the construction of the two-dimensional hazard source recognition model by cascading the hazard source bounding box layer and the dangerous distance recognition layer.

[0043] Optionally, during the construction of the two-dimensional hazard identification model, the system terminal first extracts the device layout features and device model information of the target monitoring device. These information include the installation location, installation method, technical specifications of the device, etc. Then, based on the power grid basic data, the layout features of the transmission line are extracted. These features include but are not limited to the physical location, length, etc. of the transmission line. Subsequently, using the device layout features, model information, and transmission line layout features as screening conditions, multiple sample hazardous intrusion images that match these conditions are collected. Each sample image not only contains the image data of the hazard source, but also marks the location of these hazard sources (i.e., the bounding box identification) and the distance between the hazard source and the monitoring device (i.e., the distance identification). After that, a hazard source bounding box layer is constructed using a convolutional neural network (CNN), including an input layer, an output layer, a convolutional layer, a pooling layer, etc. Then, the weights and biases of the hazard source bounding box layer are initialized using random numbers, etc., and multiple sample hazardous intrusion images and multiple sample hazard source bounding box identifications are used as training data and input into the initialized hazard source bounding box layer for forward propagation to generate real-time bounding box images. Then, the cross-entropy loss function and the intersection over union (IoU) are used to jointly calculate the loss value between the real-time bounding box image and the corresponding sample hazard source bounding box identification. Among them, the cross-entropy loss function is used to measure the difference between the predicted hazard source category and the true category. The IoU is used to measure the degree of overlap of the bounding boxes. The IoU represents the ratio of the intersection to the union of the predicted bounding box and the true bounding box. By minimizing the IoU loss, the hazard source bounding box layer will make the predicted bounding box coincide with the true annotated bounding box as much as possible. Then, through the backpropagation algorithm, the weights are continuously adjusted to gradually reduce the value of the loss function. Commonly used optimization algorithms such as Adam or SGD are used to update the parameters of the network. Through multiple training rounds, continuous iteration is performed on a large amount of sample data until the accuracy of the hazard source bounding box layer in bounding the hazard source in the given image is greater than or equal to the expected accuracy. After training is completed, the system terminal is verified through a validation set (sample images not participating in training) to check the performance of the hazard source bounding box layer on new data. If the accuracy is sufficient, steps such as adjusting the network structure and adding more sample data are taken. When the hazard source bounding box layer can stably output accurate bounding box results on the validation set, the system terminal outputs the current hazard source bounding box layer. At the same time, the system terminal also uses multiple sample hazardous intrusion images and multiple sample hazard source distance identifications as training data, and constructs a hazard distance identification layer based on the CNN network through the same training method as above to analyze and predict the actual distance between the hazard source and the monitoring device. Finally, by cascading the hazard source bounding box layer and the hazard distance identification layer, the system terminal completes the construction of the two-dimensional hazard identification model. This model can simultaneously identify the hazard source in the image and calculate its distance from the transmission line for subsequent monitoring.

[0044] The backend server filters the real-time point cloud data based on the precise measurement start area to obtain spatial point cloud analysis data.

[0045] In one embodiment, in the backend server, the system terminal analyzes the identified precise measurement start area to determine a spatial measurement start area, and performs spatial data filtering in this spatial measurement start area to generate high-precision spatial point cloud analysis data. This spatial point cloud analysis data will be transmitted back to the backend server via wireless transmission. The backend server will receive and store these data for subsequent analysis, processing, and further risk assessment. Through this process, a complete three-dimensional image of the measurement area can be obtained, ensuring more accurate monitoring and assessment of the hazard sources and transmission lines.

[0046] Furthermore, the present application provides that the backend server filters the real-time point cloud data based on the precise measurement start area to obtain spatial point cloud analysis data, including:

[0047] Taking the target monitoring device as a reference point, taking the transmission tower as the depth axis, and taking the spatial direction of the target transmission line as the horizontal axis, a standard three-dimensional space is constructed; the maximum straight-line distance of the precise measurement start area is collected, and the collection result is used as the point cloud filtering distance constraint; the box selection spatial feature is output according to the shape characteristics of the target hazard source; after positioning the precise measurement start area in the standard three-dimensional space according to the device layout characteristics and real-time distance parameters, the spatial measurement start area is box-selected based on the point cloud filtering distance constraint and the box selection spatial feature; the real-time point cloud data is filtered spatially with the box-selected spatial measurement start area as the constraint to obtain the spatial point cloud analysis data.

[0048] Optionally, the system terminal sets the target monitoring device as the reference point, combines the depth of the transmission tower as the axis in the vertical direction, and uses the spatial direction of the target transmission line as the axis in the horizontal direction to construct a standard three-dimensional space model. This three-dimensional space is used to precisely locate and analyze the precise measurement start area. Subsequently, the identified precise measurement start area is analyzed to determine the maximum straight-line distance within this area. This distance represents the longest distance that needs to be covered during the filtering process. To ensure sufficient measurement coverage, the system terminal sets this maximum straight-line distance as the point cloud filtering distance constraint, ensuring that subsequent point cloud data processing only focuses on the range related to the measurement start area. Then, the system terminal analyzes the shape characteristics of the hazard source (such as rectangle, circle, complex shape, etc.) based on the real-time monitoring image, and through the boundary extraction algorithm (RANSAC algorithm), outputs the specific framed spatial characteristics of the hazard source, which describe the spatial range, direction, and size of the hazard source. Then, based on the layout characteristics of the monitoring devices and the distance data between the hazard source and the transmission line collected in real time, the system terminal precisely locates the precise measurement start area within the constructed three-dimensional space. Within the precise measurement start area, through the three-dimensional coordinates of the point cloud data, the system terminal filters out the point cloud data that is within the start area range and is at a certain distance from the transmission line and the hazard source, and eliminates the points that exceed the filtering distance constraint range. This ensures that subsequent processing only focuses on the point cloud data of the relevant area. Then, by combining the framed spatial characteristics with the filtering distance constraint, a more precise spatial measurement area is generated. This process ensures that the measurement start area includes both the hazard source and its influence range, while excluding irrelevant redundant areas. In this way, with the point cloud filtering distance constraint as the boundary and the three-dimensional shape of the framed spatial characteristics as a supplement, the system terminal generates the final range of the spatial measurement start area. The shape of this spatial measurement start area may be a regular geometric body (such as a rectangle, cylinder), or an irregular three-dimensional polygon area, depending on the shape and influence range of the hazard source. The point cloud data inside it will be used for three-dimensional space analysis, ensuring that data processing focuses on the key areas related to the hazard source, reducing the calculation amount and improving the analysis accuracy. Finally, with the framed spatial measurement start area as the constraint condition, the system terminal filters the real-time point cloud data. The purpose of filtering is to remove the spatial points in the point cloud data that are irrelevant to the measurement start area and only retain the point cloud data within the start area. The filtering process is carried out by checking whether each point cloud point is located within the framed measurement start area (according to the three-dimensional space coordinates). Only the points that meet the conditions will be retained to form a refined point cloud data set, that is, the spatial point cloud analysis data. These data focus on the key points within the precise measurement start area, reflecting the three-dimensional characteristics of this area, and providing support for subsequent measurement, analysis, and risk assessment. Through this process, the point cloud data can be efficiently screened and processed to ensure that only the spatial information related to the precise measurement start area is concerned, thus greatly improving the accuracy and efficiency of the analysis.

[0049] The backend server registers the spatial point cloud analysis data and the real-time monitoring image to obtain a spatial information alignment model.

[0050] In one embodiment, in the backend server, the system terminal registers the collected spatial point cloud analysis data (3D data) and the real-time monitoring image (2D image) from the monitoring device. The core objective of the registration is to integrate the information in the 2D image with the 3D point cloud data to ensure the precise correspondence of their spatial positions and features. Specifically, the system terminal filters the spatial point cloud analysis data according to the accurately measured starting area, and then performs feature matching between the filtering result and the real-time monitoring image to complete the registration alignment of the real-time framed image and the spatial point cloud analysis data, ensuring that the objects in the monitoring image can be accurately mapped in the 3D point cloud data. Through this registration, a spatial information alignment model can be generated, that is, an environmental map combining the 2D image and the 3D point cloud data. This model can more intuitively display the environmental information of the measurement area and contribute to further analysis and risk assessment.

[0051] Further, the present application provides the backend server to register the spatial point cloud analysis data and the real-time monitoring image to obtain a spatial information alignment model, including:

[0052] Extract feature points from the spatial point cloud analysis data to obtain a point cloud feature set; extract features from the real-time framed image to obtain a registration feature set; perform feature matching between the point cloud feature set and the registration feature set to locate the registration matrix; perform registration alignment of the real-time framed image and the spatial point cloud analysis data based on the registration matrix, and output the spatial information alignment model.

[0053] Optionally, the system terminal performs feature extraction on the collected spatial point cloud analysis data, identifies key points therein, such as terrain features, object contours, etc., and generates a point cloud feature set. At the same time, the system terminal performs feature extraction on the real-time box-selected image, extracts key features related to the hazard source from the image, such as edges, corner points, etc., to form a registration feature set. Common algorithms for extracting feature points include SIFT (Scale-Invariant Feature Transform), Harris corner detection, etc. After that, the extracted point cloud feature points and image feature points are described. This process represents the local information of each feature point as a vector, including geometric information or texture information of its surrounding environment. The description method of point cloud features usually combines its three-dimensional geometric shape and local curvature information, while the description of image features is represented by information such as gray scale or color change. Then, by comparing the description vectors of each point cloud feature point with the description vectors of each image feature point, matching feature point pairs with similar distances within the similarity distance threshold are found through the Euclidean distance. Based on the matched point pairs, the system terminal minimizes the similarity distance between the matched point pairs and calculates a registration matrix for aligning the point cloud feature set and the registration feature set. This matrix usually includes translation, rotation, and scaling operations to transform the feature points in the two-dimensional image to the corresponding positions in the three-dimensional point cloud space, so that the two are aligned in the same spatial framework. Finally, based on this registration matrix, the system terminal precisely registers and aligns the real-time box-selected image with the spatial point cloud analysis data. After registration is completed, a spatial information alignment model is output, which integrates two-dimensional image and three-dimensional point cloud data. Using this alignment model, the system terminal can accurately evaluate the distance between the hazard source and the transmission line in combination with the safety threshold of the transmission line, and judge the potential threat degree of the hazard source to the transmission line according to the calculated distance and risk level rules, providing a basis for subsequent risk warning.

[0054] Perform spatial ranging of the target hazard source and the target transmission line in the spatial information alignment model, and output hazard spacing quantization data.

[0055] In one embodiment, in the generated spatial information alignment model, the system terminal precisely aligns the three-dimensional spatial position information of the target hazard source and the target transmission line, and uses this alignment model to calculate the actual distance between the hazard source and the transmission line. Specifically, the system terminal extracts the three-dimensional coordinates of the hazard source and the three-dimensional coordinates of the transmission line from the aligned model, and then measures the straight-line distance between the two through geometric calculations. This process ensures the accuracy of the spatial distance and outputs the measurement result in digital form as hazard spacing quantization data. These quantization data are used to evaluate the potential threat of the hazard source to the transmission line, and judge whether the hazard source exceeds the safety threshold according to the size of the distance. If the distance is less than the preset safety distance, a corresponding warning will be issued.

[0056] In summary, the embodiments of the present application have at least the following technical effects:

[0057] In the embodiments of the present application, real-time data collection of the transmission line and its surrounding environment is performed through a laser point cloud device and a target monitoring device. The backend server combines the real-time monitoring images and point cloud data, and through steps such as region of interest selection, spatial data filtering, two-dimensional recognition model, and risk assessment, generates a spatial information alignment model to achieve accurate distance measurement between the hazard source and the transmission line and output quantitative dangerous distance data. At the same time, by constructing the grid basic data and risk assessment model of the transmission line, the monitoring strategy is further optimized, and the accuracy and reliability of hazard source identification and distance measurement are improved, providing strong technical support for the safety management of transmission lines. These technical effects together solve the technical problem that the existing transmission line monitoring equipment cannot accurately locate and timely warn of hazard sources due to the lack of accurate distance measurement and real-time dynamic recognition of the external environment, and achieve the effect of identifying and measuring the distance of hazard sources through a laser point cloud device and an image monitoring device, reducing the accident rate.

[0058] Embodiment 2, based on the same inventive concept as the method for measuring the distance of a hazard source of a transmission line based on point cloud spatial information fusion in the foregoing embodiment, as Figure 2 shown, the present application provides a system for measuring the distance of a hazard source of a transmission line based on point cloud spatial information fusion. The system includes: a point cloud data acquisition unit 1: performing real-time acquisition of point cloud data on a target transmission line through a laser point cloud device to obtain real-time point cloud data; an image acquisition unit 2: performing image acquisition on the target transmission line through a target monitoring device to obtain real-time monitoring images, where the target monitoring device and the laser point cloud device are pre-installed on a transmission tower mechanically connected to the target transmission line; a region selection unit 3: the backend server receives the real-time monitoring images transmitted back by the target monitoring device and performs region of interest selection based on the real-time monitoring images to locate the precise measurement start area, where the target monitoring device is indirectly connected to the backend server through a wireless transmission relay module, and the precise measurement start area covers the target hazard source; a spatial data filtering unit 4: the backend server filters the real-time point cloud data according to the precise measurement start area to obtain spatial point cloud analysis data; a data registration unit 5: the backend server registers the spatial point cloud analysis data and the real-time monitoring images to obtain a spatial information alignment model; a spatial distance measurement unit 6: performing spatial distance measurement between the target hazard source and the target transmission line in the spatial information alignment model and outputting quantitative dangerous distance data.

[0059] Furthermore, the image acquisition unit 2 is further used to execute the following method:

[0060] Interactively obtain the grid basic data of the target transmission line; perform a risk assessment of the transmission line based on the grid basic data to obtain a transmission line risk number table; preset the identification frequencies of multiple levels of hazard sources; after obtaining the target risk level of the target transmission line by calling from the transmission line risk number table, traverse the identification frequencies of multiple levels of hazard sources with the target risk level to configure the image acquisition frequency of the target monitoring device; the target monitoring device performs intermittent image acquisition on the target transmission line with the image acquisition frequency as a constraint.

[0061] Further, the image acquisition unit 2 is further configured to execute the following method:

[0062] The grid basic data includes grid topology data, line attribute data, system operation data, and historical maintenance data.

[0063] Further, the image acquisition unit 2 is further configured to execute the following method:

[0064] Based on the grid topology data, analyze the power outage scope to obtain the power outage node ratios of multiple transmission lines; extract the multiple line length parameters of the multiple transmission lines from the line attribute data, and perform impedance calculation on the multiple transmission lines based on the line attribute data to obtain multiple line impedances; evaluate the transmission capacity of the multiple transmission lines based on the system operation data to obtain multiple line transmission capacities; call from the historical maintenance data to obtain the multiple historical operation and maintenance frequencies, multiple remaining service lives, and multiple line operation cycles of the multiple transmission lines; pre-construct a transmission risk evaluation function, and the transmission risk evaluation function is as follows: ; where is the transmission risk coefficient, P is the power outage node ratio, L is the line length parameter, Z is the line impedance, C is the line transmission capacity, M is the historical operation and maintenance frequency, R is the remaining service life, T is the line operation cycle; by synchronizing the multiple power outage node ratios, multiple line length parameters, multiple line impedances, multiple line transmission capacities, multiple historical operation and maintenance frequencies, multiple remaining service lives, and multiple line operation cycles of the multiple transmission lines to the transmission risk evaluation function, obtain multiple transmission risk coefficients; associatively store the multiple transmission lines and the multiple transmission risk coefficients to form the transmission line risk number table.

[0065] Further, the region selection unit 3 is further configured to execute the following method:

[0066] A two-dimensional hazard identification model is pre-deployed in the backend server, where the two-dimensional hazard identification model includes a cascaded hazard bounding box layer and a hazard distance identification layer; the hazard bounding box layer of the two-dimensional hazard identification model is used to perform hazard identification and bounding on the real-time monitoring image. If the real-time monitoring image includes the target hazard, the hazard bounding box layer outputs a real-time bounding image; based on the real-time bounding image, the hazard distance identification layer is activated to perform image acquisition distance analysis and output real-time distance parameters; based on the real-time bounding image and the real-time distance parameters, spatial positioning is performed to obtain the precise measurement start area.

[0067] Further, the area bounding unit 3 is further configured to perform the following method:

[0068] Extract the device layout characteristics and device model information of the target monitoring device; extract the line layout characteristics of the target transmission line based on the power grid basic data; use the device layout characteristics, device model information, and line layout characteristics as screening constraints to collect multiple sample dangerous intrusion images, where the multiple sample dangerous intrusion images include multiple sample hazard bounding box identifiers and multiple sample hazard distance identifiers; use the multiple sample dangerous intrusion images and multiple sample hazard bounding box identifiers as training data to construct the hazard bounding box layer based on the CNN network; use the multiple sample dangerous intrusion images and multiple sample hazard distance identifiers as training data to construct the hazard distance identification layer based on the CNN network; complete the construction of the two-dimensional hazard identification model by cascading the hazard bounding box layer and the hazard distance identification layer.

[0069] Further, the spatial data filtering 4 is further configured to perform the following method:

[0070] Taking the target monitoring device as the reference point, taking the transmission tower as the depth axis, and taking the spatial direction of the target transmission line as the horizontal axis, a standard three-dimensional space is constructed; the maximum straight-line distance of the precise measurement start area is collected, and the collection result is used as the point cloud filtering distance constraint; the bounding space characteristics are output according to the shape characteristics of the target hazard; after positioning the precise measurement start area in the standard three-dimensional space according to the device layout characteristics and the real-time distance parameters, the spatial measurement start area is bounded based on the point cloud filtering distance constraint and the bounding space characteristics; the real-time point cloud data is spatially filtered with the bounded spatial measurement start area as the constraint to obtain the spatial point cloud analysis data.

[0071] Further, the data registration unit 5 is further configured to perform the following method:

[0072] Extract feature points from the spatial point cloud analysis data to obtain a point cloud feature set; extract features from the real-time selected image to obtain a registration feature set; perform feature matching on the point cloud feature set and the registration feature set to locate the registration matrix; perform registration alignment of the real-time selected image and the spatial point cloud analysis data based on the registration matrix, and output the spatial information alignment model.

[0073] It should be noted that the above sequence of embodiments of the present application is only for description and does not represent the superiority or inferiority of the embodiments. And the above describes specific embodiments of this specification. The processes depicted in the drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0074] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included within the protection scope of the present application.

[0075] This specification and the drawings are only exemplary descriptions of the present application and are considered to have covered any and all modifications, variations, combinations, or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these changes and modifications.

Claims

1. A method for measuring the distance of a dangerous source of a transmission line based on the fusion of point cloud spatial information, characterized in that, The method includes: Performing real-time acquisition of point cloud data on the target transmission line through a laser point cloud device to obtain real-time point cloud data; Performing image acquisition on the target transmission line through a target monitoring device to obtain a real-time monitoring image, where the target monitoring device and the laser point cloud device are pre-installed on a transmission tower mechanically connected to the target transmission line; The backend server receives the real-time monitoring image transmitted back by the target monitoring device, performs region of interest (ROI) selection based on the real-time monitoring image, conducts distance analysis on the image of the selected region, calculates the actual distance between the target hazard source and the monitoring device, and calculates the distance between the target hazard source and the transmission line in combination with the position information of the monitoring device to locate the precise measurement start area, and the precise measurement start area covers the target hazard source; The backend server filters the real-time point cloud data according to the precise measurement start area to obtain spatial point cloud analysis data; The backend server registers the spatial point cloud analysis data and the real-time monitoring image to obtain a spatial information alignment model; Performing spatial distance measurement between the target hazard source and the target transmission line in the spatial information alignment model to output risk distance quantization data; Pre-constructing a transmission risk evaluation function to obtain a transmission line risk number table, and configuring the image acquisition frequency of the target monitoring device. The transmission risk evaluation function is as follows: ； Among them, is the transmission risk coefficient, P is the proportion of power outage nodes, L is the line length parameter, Z is the line impedance, C is the line transmission capacity, M is the historical operation and maintenance frequency, R is the remaining service life, and T is the line operation cycle.

2. The method for measuring the distance of a transmission line hazard source based on the fusion of point cloud spatial information according to claim 1, wherein Before performing image acquisition on the target transmission line through the target monitoring device, the method includes: Interactively obtaining the grid basic data of the target transmission line; Performing transmission line risk assessment based on the grid basic data to obtain a transmission line risk number table; Presetting multi-level hazard source identification frequencies; After obtaining the target risk level of the target transmission line by calling from the transmission line risk number table, traversing the multi-level hazard source identification frequencies with the target risk level to configure the image acquisition frequency of the target monitoring device; The target monitoring device performs intermittent image acquisition on the target transmission line with the image acquisition frequency as a constraint.

3. The method for measuring the distance of a dangerous source of a transmission line based on the fusion of point cloud spatial information according to claim 2, wherein, The backend server receives the real-time monitoring image transmitted back by the target monitoring device and performs ROI selection based on the real-time monitoring image to locate the precise measurement start area. The method includes: Pre-laying a hazard source two-dimensional identification model in the backend server, where the hazard source two-dimensional identification model includes a cascaded hazard source selection layer and a hazard distance identification layer; Using the hazard source selection layer of the hazard source two-dimensional identification model to perform hazard source identification and selection on the real-time monitoring image. If the real-time monitoring image includes the target hazard source, the hazard source selection layer outputs a real-time selected image; Activating the hazard distance identification layer based on the real-time selected image to conduct image acquisition distance analysis and output real-time distance parameters for quantifying the distance between the position of the hazard source in the image and the monitoring device; Performing spatial positioning according to the real-time selected image and the real-time distance parameters to obtain the precise measurement start area.

4. The method for measuring the distance of a transmission line hazard source based on the fusion of point cloud spatial information according to claim 3, characterized in that, The method includes: Extracting the device layout characteristics and device model information of the target monitoring device; Extract the line layout characteristics of the target transmission line based on the grid basic data; Taking the equipment layout characteristics, equipment model information, and line layout characteristics as screening constraints, collect multiple sample dangerous intrusion images, where the multiple sample dangerous intrusion images include multiple sample hazard source bounding box identifiers and multiple sample hazard source distance identifiers; Taking the multiple sample dangerous intrusion images and multiple sample hazard source bounding box identifiers as training data, construct the hazard source bounding box layer based on the CNN network; Taking the multiple sample dangerous intrusion images and multiple sample hazard source distance identifiers as training data, construct the dangerous distance recognition layer based on the CNN network; Complete the construction of the two-dimensional hazard source recognition model by cascading the hazard source bounding box layer and the dangerous distance recognition layer.

5. The method for measuring the distance of the transmission line hazard source based on the fusion of point cloud spatial information according to claim 4, characterized in that, The backend server filters the real-time point cloud data according to the precise measurement activation area to obtain spatial point cloud analysis data. The method includes: Taking the target monitoring device as the reference point, taking the transmission tower as the depth axis, and taking the spatial direction of the target transmission line as the horizontal axis, construct a standard three-dimensional space; Collect the maximum straight-line distance of the precise measurement activation area, and use the collection result as the point cloud filtering distance constraint; Output the bounding box selection spatial characteristics according to the shape characteristics of the target hazard source; After positioning the precise measurement activation area in the standard three-dimensional space according to the equipment layout characteristics and real-time distance parameters, select the spatial measurement activation area based on the point cloud filtering distance constraint and the bounding box selection spatial characteristics; Filter the real-time point cloud data according to the selected spatial measurement activation area to obtain the spatial point cloud analysis data.

6. The method for measuring the distance of a dangerous source of a transmission line based on the fusion of point cloud spatial information according to claim 2, wherein, The grid basic data includes grid topology data, line attribute data, system operation data, and historical maintenance data.

7. The method for measuring the distance of a transmission line hazard source based on point cloud space information fusion according to claim 6, wherein Conduct a risk assessment of the transmission line based on the grid basic data to obtain a transmission line risk number table. The method includes: Conduct an outage range analysis based on the grid topology data to obtain the outage node ratios of multiple transmission lines; Extract the line length parameters of the multiple transmission lines from the line attribute data, and calculate the impedance of the multiple transmission lines based on the line attribute data to obtain multiple line impedances; Evaluate the transmission capacity of the multiple transmission lines based on the system operation data to obtain multiple line transmission capacities; Call the multiple historical operation and maintenance frequencies, multiple remaining service lives, and multiple line operation cycles of the multiple transmission lines from the historical maintenance data; Synchronize the outage node ratios, line length parameters, line impedances, line transmission capacities, historical operation and maintenance frequencies, remaining service lives, and line operation cycles of the multiple transmission lines to the transmission risk assessment function to obtain multiple transmission risk coefficients; Associate and store the multiple transmission lines and multiple transmission risk coefficients to form the transmission line risk number table.

8. The method for measuring the distance of a dangerous source of a transmission line based on the fusion of point cloud spatial information according to claim 5, characterized in that, The backend server registers the spatial point cloud analysis data and the real-time monitoring image to obtain a spatial information alignment model. The method includes: Extract feature points from the spatial point cloud analysis data to obtain a point cloud feature set; Extract features from the real-time boxed image to obtain a registration feature set; Perform feature matching on the point cloud feature set and the registration feature set to locate the registration matrix; Based on the registration matrix, perform registration alignment of the real-time boxed image and the spatial point cloud analysis data, and output the spatial information alignment model.

9. A transmission line hazard distance measurement system based on the fusion of point cloud spatial information, characterized in that, Steps for implementing the transmission line hazard distance measurement method based on point cloud spatial information fusion according to any one of claims 1 to 8, including: Point cloud data acquisition unit: Real-time collect point cloud data of the target transmission line through a laser point cloud device to obtain real-time point cloud data; Image acquisition unit: Collect images of the target transmission line through a target monitoring device to obtain real-time monitoring images, wherein the target monitoring device and the laser point cloud device are pre-installed on a transmission tower mechanically connected to the target transmission line; Region box selection unit: The backend server receives the real-time monitoring image transmitted back by the target monitoring device, and based on the real-time monitoring image, performs region of interest box selection to locate the precise measurement start area, wherein the target monitoring device is indirectly connected to the backend server through a wireless transmission relay module, and wherein the precise measurement start area covers the target hazard; Spatial data filtering unit: The backend server filters the real-time point cloud data according to the precise measurement start area to obtain spatial point cloud analysis data; Data registration unit: The backend server registers the spatial point cloud analysis data and the real-time monitoring image to obtain a spatial information alignment model; Spatial distance measurement unit: Perform spatial distance measurement between the target hazard and the target transmission line in the spatial information alignment model, and output the quantified data of the dangerous distance.

Images