Power transmission line hidden danger analysis method based on distance measurement of pictures monitored and shot by visual device

Abstract

The invention discloses a power transmission line hidden danger analysis method based on visual device monitoring picture distance measurement, and the method comprises the steps: obtaining the high-precision three-dimensional information of a power transmission line and the surrounding environment through a laser point cloud technology, and providing a distance measurement coordinate for a visual device monitoring picture; developing a fusion algorithm of laser point cloud data of the power transmission channel and a picture monitored and shot by the visual device, and fusing the two kinds of data to realize distance measurement and accurate positioning of a hidden danger point target from the power transmission line; then, detailed path data of the power transmission line are collected, meanwhile, various hidden danger data are arranged, the hidden danger data and the path data of the power transmission line are subjected to overlay analysis through a spatial analysis technology, the hidden danger data and the path data of the power transmission line are subjected to fusion comparison through a data association model, and accurate judgment of hidden danger levels and pushing of management and control measures are achieved; the method has the advantages of being accurate in hidden danger point positioning, high in hidden danger grade judgment accuracy and high in management and control measure pertinence, and intelligent analysis and accurate pushing of hidden dangers can be achieved.

Description

Technical Field

[0001] This invention belongs to the technical field of power transmission line hazard analysis, specifically relating to a method for power transmission line hazard analysis based on distance measurement from images captured by a visualization device. Background Technology

[0002] Currently, the panoramic monitoring system for power transmission lines uses artificial intelligence analysis of images captured by visualization devices to identify and push information about potential hazards such as construction equipment, foreign objects, and smoke / fire, greatly improving the efficiency of hazard control along the transmission line. However, the following problems still exist: 1. A flood of data has emerged, with an astonishing number of potential hazard images: Currently, the system has achieved 2,000 video surveillance inspections. The massive monitoring scale has brought about a huge amount of data, generating about 300,000 images every day, of which about 30,000 are potential hazard images. These potential hazard images carry critical information about line safety, but also bring great challenges to subsequent processing. 2. Inaccurate judgment, time-consuming, labor-intensive and inefficient: Because the location and distance of the hidden danger cannot be determined, it is difficult for staff to accurately grasp the severity and scope of the hidden danger, and empty runs often occur. According to more than two years of operation experience of the intelligent management system for transmission lines, a large number of images have wasted a lot of human and material resources in judging hidden danger points and transmission lines. Therefore, there is an urgent need to research new technologies to reprocess and reanalyze images of potential hazards, to achieve distance measurement from potential hazard points to the line, and thus to realize intelligent analysis and accurate push of potential hazards. To solve the above problems, it is necessary to develop a method for analyzing potential hazards in transmission lines based on distance measurement of images captured by a visual device. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method for analyzing transmission line hazards based on distance measurement of images captured by a visualization device. This method is characterized by accurate hazard location, high accuracy in hazard level judgment, and targeted control measures, thereby achieving intelligent analysis and precise push of hazards.

[0004] The objective of this invention is achieved as follows: a method for analyzing potential hazards in power transmission lines based on distance measurement using images captured by a visualization device, comprising the following steps: S1, Data Fusion and Precise Positioning: Utilizing laser point cloud technology, high-precision three-dimensional information of transmission lines and their surrounding environment is obtained. At the same time, ranging coordinates are provided for images captured by visualization devices. A fusion algorithm is developed to integrate laser point cloud data of transmission channels with images captured by visualization devices. By fusing these two types of data, the distance between transmission lines and potential hazard points can be measured and precisely positioned. S2, Hazard Assessment and Decision-Making: Collect detailed route data of transmission lines, and organize various hazard data obtained in step S1. Using spatial analysis technology, overlay and analyze the hazard data with the transmission line route data. Through data association model, fuse and compare the hazard data with the transmission line route data to accurately determine the hazard level and push control measures.

[0005] Preferably, step S1 includes the following steps: S11, Data Acquisition: S111, Laser point cloud data: A 3D laser scanner is used to scan the power transmission line and its surrounding environment to obtain high-precision laser point cloud data; S112, Visualization Device Images: By using the deployed visualization devices, images are taken at set time intervals to obtain a large number of visualization device images, and the large number of visualization device images are deduplicated. S12, Laser point cloud data processing: S121, Noise Removal: A bilateral filtering algorithm is used to remove noise while preserving the edge features of the point cloud data, and the filtering parameters are determined through multiple experiments. S122, Feature Extraction: Using methods based on region growing and morphological operations, geometric features of key parts of the line are extracted; S13, Feature extraction from the deduplicated visualization device image: S131, Shape Feature Extraction: The Canny edge detection algorithm is used to extract the edges of objects in the image, and then the contour detection algorithm is used to obtain the contour information of the object, thereby describing the shape of the object; S132, Color Feature Extraction: Convert the image from the RGB color space to the HSV color space, and then calculate the HSV histogram as a representation of color features; S14, Data Fusion and Scene Ranging: S141, Coordinate system establishment: A unified three-dimensional coordinate system is established with the starting tower of the transmission line as the origin, the line direction as the X-axis, and the vertical upward direction as the Z-axis. At the same time, the images of the visualization device are calibrated to determine the conversion relationship between image pixels and actual physical coordinates. S142, Feature Matching: The SIFT algorithm is used to extract feature points from laser point cloud data and images. Then, the KD tree algorithm is used to quickly match feature points. For matched point pairs, the coordinates of the hazard point in the three-dimensional coordinate system are calculated using the triangulation principle, thereby realizing the target distance measurement and accurate positioning of the hazard point.

[0006] More preferably, in step S111, the scanning frequency of the 3D laser scanner is set according to the complexity of the circuit, and is set to scan once every 10 to 20 seconds; in step S112, the time interval is set to take an image once every 5 to 10 minutes.

[0007] More preferably, in step S131, the contour detection algorithm uses the findContours function in OpenCV.

[0008] Preferably, step S2 includes the following steps: S21, Data Collection and Organization: S211, Transmission line route data: Obtain transmission line route data from transmission line design drawings and geographic information databases, and perform format standardization and data cleaning; S212, Hazard Data: Organize the hazard data identified in the images captured by the visualization device in step S1. For hazard data with inaccurate location information, correct it by combining the positioning results after fusion of laser point cloud and image. S22, Spatial Analysis: S221, Import data into GIS platform: Import the organized transmission line route data and hidden danger data into a professional GIS platform, convert the transmission line data into a vector layer, and the hidden danger data into a point layer, and ensure that the two layers have the same coordinate system; S222, Overlay Analysis: Overlay analysis is performed on the GIS platform, overlaying the hazard point layer onto the transmission line vector layer. By setting different symbols, colors and transparency, the distribution of hazards around the transmission line is displayed intuitively. At the same time, the spatial analysis tools of the GIS platform are used to calculate the spatial relationship between the hazard points and the transmission line. S23, Data Correlation and Hazard Level Assessment: S231, Establish a data association model: Using the geographical location of the potential hazard point as a bridge, establish the association between the potential hazard data and the transmission line routing data; S232, Construct a hazard level assessment index system: comprehensively consider factors such as hazard type, distance between hazard point and transmission line, and frequency of hazard occurrence to construct a hazard level assessment index system; S233, Hazard Level Judgment: The weight of each evaluation indicator is determined by the Analytic Hierarchy Process (AHP), and then the fuzzy comprehensive evaluation method is used to quantitatively evaluate the hazards and determine the hazard level. S234, Control Measures Push: Based on different hazard levels, corresponding control measures are preset. Specifically, for high-level hazards, an SMS is immediately sent to notify maintenance personnel to go to the site for handling; for medium-level hazards, an inspection is arranged in the near future and close monitoring is carried out; and for low-level hazards, they are recorded and reviewed regularly.

[0009] More preferably, in step S211, the transmission line route data includes the starting point, ending point, tower location, and route direction of the line; in step S212, the hazard data includes the hazard type, occurrence time, and latitude and longitude location.

[0010] In a further preferred embodiment, in step S221, the professional GIS platform used is ArcGIS; in step S222, the spatial relationship between the potential hazard point and the transmission line is distance and angle.

[0011] In a further preferred embodiment, in step S231, the association relationship is specifically achieved by writing SQL statements or using the data association function of a GIS platform, matching the latitude and longitude information in the hidden danger data with the coordinate information in the transmission line route data, thereby achieving deep data fusion.

[0012] Due to the adoption of the above technical solution, the beneficial effects of the present invention are: This invention employs a fusion algorithm combining laser point clouds and images from a visualization device to achieve precise measurement of the location and distance of potential hazards relative to transmission lines. Actual testing shows a ranging accuracy of ±0.5 meters, and the positioning accuracy meets the needs of on-site hazard analysis. This allows staff to accurately grasp the severity and impact range of hazards, effectively reducing wasted trips and improving work efficiency. Statistics show that after applying this technology, the number of invalid inspections due to inaccurate positioning has decreased by approximately 60%. This invention uses the fusion and comparison of hidden danger data and transmission line routes to achieve accurate judgment of hidden danger level by combining hidden danger type. By establishing a scientific hidden danger level assessment index system and data association model, the accuracy rate of hidden danger level judgment is increased to more than 90%. At the same time, targeted control measures are automatically pushed according to different levels, realizing differentiated management of hidden dangers and greatly improving the efficiency and effectiveness of hidden danger handling. For example, the response time for handling high-level hidden dangers has been shortened from the original average of 2 hours to less than 30 minutes. In summary, this invention has the advantages of accurate hazard location, high accuracy in hazard level judgment, and highly targeted control measures, enabling intelligent analysis and precise delivery of hazard information. Attached Figure Description

[0013] Figure 1 This is a schematic diagram of the method steps of the present invention. Detailed Implementation

[0014] The technical solution of the present invention will be further described in detail below through embodiments and in conjunction with the accompanying drawings.

[0015] like Figure 1 As shown, this invention provides a method for analyzing potential hazards in power transmission lines based on distance measurement from images captured by a visualization device, characterized by the following steps: S1, Data Fusion and Precise Positioning: Utilizing laser point cloud technology, high-precision three-dimensional information of transmission lines and their surrounding environment is obtained. At the same time, ranging coordinates are provided for images captured by visualization devices. A fusion algorithm is developed to integrate laser point cloud data of transmission channels with images captured by visualization devices. By fusing these two types of data, the distance between the transmission line and the target of the hidden danger point can be measured and precisely positioned.

[0016] S11, Data Acquisition: Used to acquire raw laser point cloud data and images from visualization devices.

[0017] S111, Laser Point Cloud Data: A 3D laser scanner is used to scan the power transmission line and its surrounding environment to obtain high-precision laser point cloud data. The scanning frequency is set according to the complexity of the line, and can generally be set to scan once every 10 to 20 seconds to ensure full coverage of the line and surrounding areas where potential hazards may occur.

[0018] S112, Visualization Device Images: Using 2000 deployed visualization devices, images are captured at set time intervals (e.g., every 5-10 minutes) to obtain a massive amount of visualization device images. This ensures that the images can reflect the line status in real time. The massive amount of visualization device images is deduplicated to reduce the amount of subsequent image processing work.

[0019] S12, Laser Point Cloud Data Processing: Used to acquire clean point cloud data with key feature labels.

[0020] S121, Noise Removal: A bilateral filtering algorithm is adopted to remove noise while preserving the edge features of the point cloud data. The filtering parameters are determined through multiple experiments to ensure that the extraction of geometric features is not affected while effectively removing noise.

[0021] S122, Feature Extraction: Using methods based on region growing and morphological operations, geometric features of key parts of the line such as towers and conductors are extracted; for example, the position and height of towers are determined by identifying planar regions in point cloud data, and the direction of conductors is determined by using curve fitting algorithms.

[0022] S13, Feature extraction of deduplicated visualization device images: used to obtain image data containing object shape and color features.

[0023] S131, Shape Feature Extraction: The Canny edge detection algorithm is used to extract the edges of objects in the image, and then a contour detection algorithm (such as the findContours function in OpenCV) is used to obtain the contour information of the object, thereby describing the shape of the object.

[0024] S132, Color Feature Extraction: Convert the image from the RGB color space to the HSV color space, and then calculate the HSV histogram as a representation of color features.

[0025] S14, Data Fusion and Scene Ranging: Used to obtain information on potential hazard points with precise three-dimensional coordinates.

[0026] S141, Coordinate system establishment: A unified three-dimensional coordinate system is established with the starting tower of the transmission line as the origin, the line direction as the X-axis, and the vertical upward direction as the Z-axis. At the same time, the images of the visualization device are calibrated to determine the conversion relationship between image pixels and actual physical coordinates.

[0027] S142, Feature Matching: The SIFT algorithm is used to extract feature points from laser point cloud data and images. Then, the KD tree algorithm is used to quickly match feature points. For matched point pairs, the coordinates of the hazard point in the three-dimensional coordinate system are calculated using the triangulation principle, thereby realizing the target distance measurement and accurate positioning of the hazard point.

[0028] S2, Hazard Assessment and Decision-Making: Collect detailed route data of transmission lines, and organize various hazard data obtained in step S1. Using spatial analysis technology, overlay and analyze the hazard data with the transmission line route data. Through data association model, fuse and compare the hazard data with the transmission line route data to accurately determine the hazard level and push control measures.

[0029] S21, Data Collection and Organization: Used to obtain standardized hazard layers and route layers that can be used for spatial analysis.

[0030] S211, Transmission line route data: Obtain detailed route data such as the starting point, ending point, tower location, and route direction of the transmission line from sources such as transmission line design drawings and geographic information databases, and perform format standardization and data cleaning to ensure the accuracy and completeness of the data.

[0031] S212, Hazard Data: Organize the hazard data identified in the images captured by the visualization device in step S1, including information such as hazard type, occurrence time, and latitude and longitude location. For hazard data with inaccurate location information, correct it by combining the positioning results after fusion of laser point cloud and image.

[0032] S22, Spatial Analysis: Used to obtain a visualized map of hazard distribution and a hazard attribute table containing spatial relationship data.

[0033] S221, Import data into GIS platform: Import the organized transmission line route data and hazard data into a professional GIS platform, such as ArcGIS, convert the transmission line data into a vector layer and the hazard data into a point layer, and ensure that the two layers have the same coordinate system.

[0034] S222, Overlay Analysis: Overlay analysis is performed on the GIS platform, overlaying the hazard point layer onto the transmission line vector layer. By setting different symbols, colors, and transparency, the distribution of hazards around the transmission line is displayed intuitively. At the same time, the spatial analysis tools of the GIS platform are used to calculate the spatial relationship between the hazard points and the transmission line, such as the distance and angle.

[0035] S23, Data Correlation and Hazard Level Assessment: S231, Establish a data association model: Using the geographical location of the potential hazard points as a bridge, establish a relationship between the potential hazard data and the transmission line path data; achieve deep data fusion by writing SQL statements or using the data association function of the GIS platform to match the latitude and longitude information in the potential hazard data with the coordinate information in the transmission line path data.

[0036] S232, Construct a hazard level assessment index system: Comprehensively consider factors such as hazard type, distance between hazard point and transmission line, and frequency of hazard occurrence to construct a hazard level assessment index system; for example, for construction equipment hazards, those within 10 meters of the line are high-level hazards, those 10 to 20 meters are medium-level hazards, and those above 20 meters are low-level hazards; for foreign object hanging hazards, the level is determined according to the hanging location and the degree of impact on line operation.

[0037] S233, Hazard Level Judgment: The weight of each evaluation indicator is determined by the Analytic Hierarchy Process (AHP), and then the fuzzy comprehensive evaluation method is used to quantitatively evaluate the hazard and determine the hazard level.

[0038] S234, Control Measures Push: Based on different hazard levels, corresponding control measures are preset. For example, for high-level hazards, an SMS is immediately sent to notify maintenance personnel to go to the site for handling; for medium-level hazards, an inspection is arranged in the near future and close attention is paid; and for low-level hazards, they are recorded and reviewed regularly.

[0039] In summary, this invention first achieves precise digitization of the physical world through advanced sensor budgeting, and then drives scientific business decisions based on high-quality digitized information, providing an intelligent, accurate, and efficient closed-loop management method for the automatic identification and control of hidden dangers in power transmission lines.

[0040] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of the claims of the present invention.

Claims

1. A method for analyzing potential hazards in power transmission lines based on distance measurement from images captured by a visual monitoring device, characterized in that, Includes the following steps: S1, Data Fusion and Precise Positioning: Utilizing laser point cloud technology, high-precision three-dimensional information of transmission lines and their surrounding environment is obtained. At the same time, ranging coordinates are provided for images captured by visualization devices. A fusion algorithm is developed to integrate laser point cloud data of transmission channels with images captured by visualization devices. By fusing these two types of data, the distance between transmission lines and potential hazard points can be measured and precisely positioned. S2, Hazard Assessment and Decision-Making: Collect detailed route data of transmission lines, and organize various hazard data obtained in step S1. Using spatial analysis technology, overlay and analyze the hazard data with the transmission line route data. Through data association model, fuse and compare the hazard data with the transmission line route data to accurately determine the hazard level and push control measures.

2. The method for analyzing potential hazards in transmission lines based on distance measurement using images captured by a visualization device, as described in claim 1, is characterized in that... Step S1 includes the following steps: S11, Data Acquisition: S111, Laser point cloud data: A 3D laser scanner is used to scan the power transmission line and its surrounding environment to obtain high-precision laser point cloud data; S112, Visualization Device Images: By using the deployed visualization devices, images are taken at set time intervals to obtain a large number of visualization device images, and the large number of visualization device images are deduplicated. S12, Laser point cloud data processing: S121, Noise Removal: A bilateral filtering algorithm is used to remove noise while preserving the edge features of the point cloud data, and the filtering parameters are determined through multiple experiments. S122, Feature Extraction: Using methods based on region growing and morphological operations, geometric features of key parts of the line are extracted; S13, Feature extraction from the deduplicated visualization device image: S131, Shape Feature Extraction: The Canny edge detection algorithm is used to extract the edges of objects in the image, and then the contour detection algorithm is used to obtain the contour information of the object, thereby describing the shape of the object; S132, Color Feature Extraction: Convert the image from the RGB color space to the HSV color space, and then calculate the HSV histogram as a representation of color features; S14, Data Fusion and Scene Ranging: S141, Coordinate system establishment: A unified three-dimensional coordinate system is established with the starting tower of the transmission line as the origin, the line direction as the X-axis, and the vertical upward direction as the Z-axis. At the same time, the images of the visualization device are calibrated to determine the conversion relationship between image pixels and actual physical coordinates. S142, Feature Matching: The SIFT algorithm is used to extract feature points from laser point cloud data and images. Then, the KD tree algorithm is used to quickly match feature points. For matched point pairs, the coordinates of the hazard point in the three-dimensional coordinate system are calculated using the triangulation principle, thereby realizing the target distance measurement and accurate positioning of the hazard point.

3. The method for analyzing potential hazards in transmission lines based on distance measurement using images captured by a visualization device, as described in claim 2, is characterized in that: In step S111, the scanning frequency of the 3D laser scanner is set according to the complexity of the circuit, and is set to scan once every 10 to 20 seconds; in step S112, the time interval is set to take an image once every 5 to 10 minutes.

4. The method for analyzing potential hazards in transmission lines based on distance measurement using images captured by a visualization device, as described in claim 2, is characterized in that: In step S131, the contour detection algorithm uses the findContours function in OpenCV.

5. The method for analyzing potential hazards in transmission lines based on distance measurement using images captured by a visualization device, as described in claim 1, is characterized in that... Step S2 includes the following steps: S21, Data Collection and Organization: S211, Transmission line route data: Obtain transmission line route data from transmission line design drawings and geographic information databases, and perform format standardization and data cleaning; S212, Hazard Data: Organize the hazard data identified in the images captured by the visualization device in step S1. For hazard data with inaccurate location information, correct it by combining the positioning results after fusion of laser point cloud and image. S22, Spatial Analysis: S221, Import data into GIS platform: Import the organized transmission line route data and hidden danger data into a professional GIS platform, convert the transmission line data into a vector layer, and the hidden danger data into a point layer, and ensure that the two layers have the same coordinate system; S222, Overlay Analysis: Overlay analysis is performed on the GIS platform, overlaying the hazard point layer onto the transmission line vector layer. By setting different symbols, colors and transparency, the distribution of hazards around the transmission line is displayed intuitively. At the same time, the spatial analysis tools of the GIS platform are used to calculate the spatial relationship between the hazard points and the transmission line. S23, Data Correlation and Hazard Level Assessment: S231, Establish a data association model: Using the geographical location of potential hazards as a bridge, establish the association between potential hazard data and transmission line routing data; S232, Construct a hazard level assessment index system: comprehensively consider factors such as hazard type, distance between hazard point and transmission line, and frequency of hazard occurrence to construct a hazard level assessment index system; S233, Hazard Level Judgment: The weight of each evaluation indicator is determined by the Analytic Hierarchy Process (AHP), and then the fuzzy comprehensive evaluation method is used to quantitatively evaluate the hazards and determine the hazard level. S234, Control Measures Push: Based on different hazard levels, corresponding control measures are preset. Specifically, for high-level hazards, an SMS is immediately sent to notify maintenance personnel to go to the site for handling; for medium-level hazards, an inspection is arranged in the near future and close monitoring is carried out; and for low-level hazards, they are recorded and reviewed regularly.

6. The method for analyzing potential hazards in transmission lines based on distance measurement using images captured by a visualization device, as described in claim 5, is characterized in that: In step S211, the transmission line route data includes the starting point, ending point, tower location, and route direction of the line; in step S212, the hazard data includes the hazard type, occurrence time, and latitude and longitude location.

7. The method for analyzing potential hazards in transmission lines based on distance measurement using images captured by a visualization device, as described in claim 5, is characterized in that: In step S221, the professional GIS platform used is ArcGIS; in step S222, the spatial relationship between the hidden danger point and the transmission line is distance and angle.

8. The method for analyzing potential hazards in transmission lines based on distance measurement using images captured by a visualization device, as described in claim 5, is characterized in that: In step S231, the association relationship is specifically achieved by writing SQL statements or using the data association function of the GIS platform, matching the latitude and longitude information in the hidden danger data with the coordinate information in the transmission line route data, thereby achieving deep data fusion.

Images