Line wind deviation and sway monitoring and early warning device combining straight-line laser and monocular vision

Abstract

The present invention discloses a line wind yaw and sway monitoring and early warning device that combines a straight-line laser with a monocular vision system, and relates to the technical field of power transmission line monitoring. The device comprises: a setting module for determining key monitoring points in a transmission line area and configuring a joint monitoring group; an image establishment and enhancement module for reading a monitoring data set from a monocular camera and synchronously reading real-time wind data from a wind monitoring sensor, and performing adaptive enhancement of the monitoring data set; a wind yaw and sway recognition module for establishing wind yaw and sway data; and an early warning module for performing environmentally sensitive threshold trigger analysis and using the trigger analysis results to report an early warning signal. This solves the technical problem of the existing line wind yaw and sway monitoring system that it is difficult to accurately capture the global sway characteristics of a large-scale transmission line area under complex climatic conditions, which in turn leads to limited accuracy, reliability, and monitoring range of wind yaw and sway monitoring, thereby achieving the technical effect of improving the monitoring range, accuracy, and reliability of wind yaw and sway monitoring.

Description

Technical Field

[0001] The present application relates to the technical field related to transmission line monitoring, and specifically to a line wind deviation and swaying monitoring and early warning device combining a straight-line laser with monocular vision. Background Art

[0002] With the growing demand for electricity due to social development, the safe and stable operation of transmission lines, especially long-distance transmission lines of high-voltage transmission lines, is crucial to the reliability of the power system. However, transmission lines are usually located in the wild environment and are exposed to various complex natural conditions all year round. For example, the influence of severe weather such as strong winds, rain, snow, hail, etc. can easily cause line dancing, resulting in a reduction in the distance between conductors, and even flashovers, line damage and other faults. Especially under the action of wind, transmission lines are prone to wind-induced dancing, which not only poses a threat to the safe operation of transmission lines, but may also cause serious consequences such as large-scale power outages. Wind-induced fluctuations refer to the horizontal and vertical deviations and vibrations of transmission line conductors caused by wind. When the wind speed is high, the conductors of the transmission line will produce periodic or non-periodic swings, especially in areas with strong winds such as valleys and coasts. The fluctuations of the transmission line are more frequent and larger in amplitude. The fluctuations of the transmission line may also cause fatigue damage to the line structure, such as tower shaking, conductor fatigue cracking, and excessive stress on insulators. Existing monitoring and early warning of wind-induced fluctuations of transmission lines face harsh natural environments and are mostly based on the local vibration characteristics of the line. It is difficult to accurately obtain the fluctuation data of the entire transmission line. In particular, the monitoring accuracy is limited over a large area, and the overall wind-induced fluctuation characteristics of the transmission line cannot be accurately reflected. In addition, the line status can only be monitored at specific locations. There are blind spots in the fluctuation monitoring over a large area, making it difficult to achieve full coverage monitoring of the transmission line, thereby affecting the monitoring accuracy of wind-induced fluctuations and the accuracy and reliability of early warning.

[0003] Therefore, in the current technologies related to line wind yaw fluctuation monitoring and early warning, there is a problem that it is difficult to accurately capture the global fluctuation characteristics of transmission line areas under complex climatic conditions and large-scale transmission line areas, which leads to technical problems such as limited accuracy, reliability and monitoring range of wind yaw fluctuation monitoring. Summary of the Invention

[0004] This application solves the technical problem of the existing line wind yaw dance monitoring that is difficult to accurately capture the global dance characteristics of large-scale transmission line areas under complex climatic conditions, which leads to limited accuracy, reliability and monitoring range of wind yaw dance monitoring, by providing a line wind yaw dance monitoring and early warning device that combines a straight-line laser with monocular vision. It achieves the technical effect of improving the wind yaw dance monitoring range, monitoring accuracy and reliability.

[0005] The present application provides a line wind deviation and sway monitoring and early warning device combining a straight line laser with a monocular vision, the device comprising: a setting module for determining key monitoring points in a transmission line area, configuring a joint monitoring group at the key monitoring points, the joint monitoring group integrating a straight line laser and a monocular camera, the projection direction of the straight line laser being set along the transverse direction of the transmission line to constrain the laser beam of the straight line laser to cover the key monitoring points, the optical axis of the monocular camera and the projection plane of the laser beam being set at a predetermined angle, and the monocular camera being able to completely capture the intersection of the laser beam and the key monitoring points; an image establishment enhancement module A block is used to activate the joint monitoring group, read the monitoring data set of the monocular camera, and synchronously read the real-time wind data of the wind monitoring sensor, configure the enhancement constraints with the real-time wind data, optimize the image processing network using the enhancement constraints, perform adaptive enhancement of the monitoring data set, and establish an adaptive enhancement result; a wind yaw and dance recognition module is used to use the adaptive enhancement result to establish wind yaw and dance data, and the wind yaw and dance data includes wind yaw angle, vertical dance amplitude, horizontal dance amplitude, and elliptical tilt angle; an early warning module is used to perform environmental sensitive threshold trigger analysis based on the wind yaw and dance data, and use the trigger analysis results to report an early warning signal.

[0006] In a possible implementation, the image establishment enhancement module also performs the following processing: calling the edge detection channel of the image processing network to perform contour recognition and establish a basic contour recognition result; calling the local segmentation channel to perform similar pixel aggregation extraction of the monitoring data set and establish a basic local segmentation result; synchronizing the basic contour recognition result and the basic local segmentation result to the joint analysis channel to establish a regional growth starting point; calculating the similarity between the growth starting point and the adjacent pixel points, reconstructing the segmentation based on the similarity, establishing a reconstructed segmentation result, and using the reconstructed segmentation result and enhancement constraints to perform local enhancement processing to complete adaptive enhancement.

[0007] In a possible implementation, the image establishment enhancement module also performs the following processing: performing gradient direction analysis of contour points on the basic contour recognition results to generate standard consistency analysis results of horizontal boundary segments; after screening contour points on the standard consistency analysis results, re-screening contour points using non-maximum suppression to establish initial seed points; using the segmentation labels of the basic local segmentation results to perform proximity authentication correction on the initial seed points, and establishing the starting point of regional growth based on the proximity authentication correction results.

[0008] In a possible implementation, the image creation enhancement module further performs the following processing: performing similarity calculation using a similarity calculation formula, as follows:

[0009] ;

[0010] in, Characterizing the starting point of growth and adjacent pixels The similarity, and Characterize the growth starting point and adjacent pixels The color value of Characterizing the starting point of growth and adjacent pixels The gradient difference, , Starting point for growth The horizontal gradient of Brightness value in the direction The rate of change, For adjacent pixels The horizontal gradient, Starting point for growth The vertical gradient of Brightness value in the direction The rate of change, For adjacent pixels The vertical gradient, is the similarity of local texture patterns, , Starting point for growth or adjacent pixels The number of pixels in the local area created, Represent any pixel point in the local area, and The starting point of growth and adjacent pixels The corresponding eigenvalue represents the starting point of growth and adjacent pixels The texture features of the pixels in the local area centered are Characterizing the starting point of growth and adjacent pixels The local feature similarity score of , is the Euclidean distance between descriptors, To adjust the parameters, 、 、 、 They are the weight factors of color feature, gradient difference feature, local texture feature, and local similarity feature respectively.

[0011] In a possible implementation, the setting module further performs the following processing: the angle range of the predetermined angle is 10° to 30°.

[0012] In a possible implementation, the wind deviation and waving recognition module further performs the following processing: establishing an initial frame position of the stationary state of the transmission line; extracting the updated frame position of the transmission line in each frame image using the adaptive enhancement result; and calculating the time series wind deviation angle using the initial frame position and the updated frame position, as follows:

[0013] ;

[0014] in, Representational Moment The windage angle, is the initial time point, and Respectively indicate time The horizontal and vertical coordinates of the starting point of the transmission line, and Respectively indicate time The horizontal and vertical coordinates of the transmission line terminal, and Respectively indicate time The horizontal and vertical coordinates of the starting point of the transmission line, and Respectively indicate time The horizontal coordinates and vertical coordinates of the transmission line terminal are obtained; the calculated wind deviation angle is used as a kind of wind deviation swing data to establish wind deviation swing data.

[0015] In a possible implementation, the line wind deviation and swaying monitoring and early warning device combining the straight-line laser with monocular vision also performs the following processing: an early warning database construction module is used to establish an environmentally sensitive threshold database mapped to the weather based on the weather history database. When the wind deviation and swaying data is generated, the early warning module calls the environmentally sensitive threshold database based on real-time weather information to perform environmentally sensitive threshold trigger analysis.

[0016] In a possible implementation, the line wind deviation and swaying monitoring and early warning device combining the straight-line laser and monocular vision also performs the following processing: the device also includes a self-update module for extracting the feedback signal of the early warning signal and establishing a self-update signal, and updating the environmental sensitive threshold database in the early warning database construction module based on the self-update signal.

[0017] The proposed line wind yaw and sway monitoring and early warning device, which combines a straight-line laser with a monocular vision, includes a setting module for determining key monitoring points in the transmission line area and configuring a joint monitoring group; an image establishment and enhancement module for reading the monitoring data set of the monocular camera and simultaneously reading the real-time wind data of the wind monitoring sensor to perform adaptive enhancement of the monitoring data set; a wind yaw and sway identification module for establishing wind yaw and sway data; and an early warning module for performing environmentally sensitive threshold trigger analysis and using the trigger analysis results to issue an early warning signal. This solves the technical problem of the existing line wind yaw and sway monitoring, which is that it is difficult to accurately capture the global sway characteristics of a large transmission line area under complex climatic conditions, which leads to limited accuracy, reliability, and monitoring range of wind yaw and sway monitoring, thereby achieving the technical effect of improving the monitoring range, accuracy, and reliability of wind yaw and sway. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] To more clearly illustrate the technical solutions of the embodiments of the present disclosure, the accompanying drawings of the embodiments of the present disclosure are briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the apparatus according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in precise order. Instead, various steps may be processed in reverse order or simultaneously as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0019] Figure 1 A schematic diagram of the structure of a line windage and sway monitoring and early warning device combining a straight-line laser with monocular vision provided in an embodiment of the present application;

[0020] Figure 2 This is a schematic diagram of the execution flow of the image creation and enhancement module in the line wind deviation and swaying monitoring and early warning device combining a straight line laser and monocular vision provided in an embodiment of the present application.

[0021] Description of reference numerals: setting module 10 , image creation and enhancement module 20 , wind deviation and fluctuation recognition module 30 , early warning module 40 . DETAILED DESCRIPTION

[0022] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0023] In order to make the purpose, technical solutions and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limiting this application. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0024] In the following description, reference is made to “some embodiments”, which describes a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict, and the terms “first\second” involved are merely used to distinguish similar objects and do not represent a specific ordering of the objects. The terms “including” and “having” and any variations are intended to cover non-exclusive inclusions. For example, a process, device, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or components that are not clearly listed or that are inherent to these processes, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs. The terms used herein are for the purpose of describing the embodiments of this application only.

[0025] The embodiment of the present application provides a line wind deviation and swaying monitoring and early warning device combining a line laser and a monocular vision, such as Figure 1 As shown, the line wind deviation and waving monitoring and early warning device combining a straight line laser with monocular vision includes: a setting module 10, an image creation and enhancement module 20, a wind deviation and waving recognition module 30, and an early warning module 40.

[0026] The setting module 10 is used to determine the key monitoring points in the transmission line area, and configure a joint monitoring group at the key monitoring points. The joint monitoring group integrates a line laser and a monocular camera. The projection direction of the line laser is set along the transverse direction of the transmission line to constrain the laser beam of the line laser to cover the key monitoring points. The optical axis of the monocular camera and the projection plane of the laser beam are set at a predetermined angle, and the monocular camera can completely capture the intersection of the laser beam and the key monitoring points.

[0027] The specific configuration of the setting module 10 also includes that the angle range of the predetermined angle is 10° to 30°.

[0028] Preferably, determining the key monitoring points in the transmission line area means selecting areas with strong winds, locations with larger line spans, or locations with greater mechanical stress in the transmission line area. These locations are key locations for monitoring the swaying of the transmission line. After the locations of the key monitoring points are determined, these key monitoring points are monitored to ensure that key parameters such as the swaying amplitude and wind angle of the transmission line can be accurately captured in these areas. In addition, the key monitoring points do not just monitor the data of one point, but collect data of a section of the transmission line centered on this monitoring point.

[0029] Preferably, a joint monitoring group is configured at the key monitoring points. The joint monitoring group refers to an integrated equipment group configured at each key monitoring point, including a straight line laser and a monocular camera. Through laser projection and camera capture, the dancing of the transmission line in the area is monitored in real time to ensure accurate monitoring in areas where the wind deviation and dancing of the transmission line are more significant. The straight line laser projects a straight laser beam, and the projection direction is along the transverse direction of the transmission line, that is, perpendicular to the longitudinal direction of the transmission line. The purpose is to allow the laser beam to sweep across the transmission line laterally. When the transmission line dances, the intersection position of the laser beam and the transmission line will change accordingly to help capture the dancing position of the transmission line, especially the displacement change of the transmission line in the transverse and vertical directions; constraining the laser beam of the straight line laser to cover the key monitoring points means accurately setting the projection range of the laser beam to ensure that it can cover the transmission line area around the key monitoring points, that is, the laser beam will be projected to the area where the transmission line is located to ensure that the laser and the transmission line continue to generate intersections during the dancing process.

[0030] Preferably, a predetermined angle is set between the optical axis of the monocular camera (i.e., the direction of the lens) and the projection plane of the laser beam to ensure that the monocular camera can most effectively capture the changes in the intersection of the laser beam and the transmission line, wherein the angle range of the predetermined angle is 10° to 30°. Since the laser beam is projected horizontally, the transmission line will dance when affected by wind, thereby forming a dynamic intersection on the laser beam. The barrage camera can capture the changes in the intersection of the transmission line and the laser beam from the best angle, thereby ensuring that the camera can monitor the offset and dancing of the transmission line in the area, record the displacement of the transmission line during wind deviation and dancing, and thus provide high-precision wind deviation and dancing monitoring data.

[0031] The image establishment enhancement module 20 is used to activate the joint monitoring group, read the monitoring data set of the monocular camera, and synchronously read the real-time wind data of the wind monitoring sensor, configure the enhancement constraints based on the real-time wind data, optimize the image processing network using the enhancement constraints, perform adaptive enhancement of the monitoring data set, and establish the adaptive enhancement results.

[0032] Preferably, activating the joint monitoring group means starting a straight-line laser and a monocular camera, the joint monitoring group starts running, and starts collecting images and monitoring data of the transmission line. The entire monitoring group is in working condition, the laser projects a laser beam, and the camera captures the image of the intersection of the transmission line and the laser beam to form a monitoring data set, which includes the dancing information of the transmission line at different time points, that is, the image data of the changes in light and image during the dancing of the transmission line with the wind. By reading these image data, the dancing of the transmission line (such as the dancing amplitude, offset angle, etc.) can be analyzed. The wind speed, wind direction and other data of the current environment are also collected in real time through the wind monitoring sensor to reflect the wind conditions of the current environment. Synchronous reading indicates that the image data and wind data are obtained at the same time and combined into the same analysis and optimization process. The wind data serves as an external influencing factor of the dancing of the transmission line.

[0033] Preferably, the enhanced constraints are configured with real-time wind data. The enhanced constraints refer to the appropriate adjustment and optimization of the image processing process according to the wind data obtained in real time. The wind data directly affects the dancing characteristics of the transmission line. The image processing parameters can be dynamically adjusted according to the current wind speed changes. For example, when the wind speed is high, the dancing amplitude of the transmission line may be larger, and more motion blur will appear in the image. The enhanced parameters of the image processing network are automatically adjusted according to the real-time wind data to adapt to different dancing amplitudes. The image processing network refers to the model used to process the camera image data. The image processing network is responsible for extracting the transmission line from the original image monitoring data. The dancing characteristics of power lines, such as detecting the edges of power lines in the image through edge detection (such as the Canny algorithm), dividing the image into different areas, identifying the power line area and the background area, and then using feature tracking (such as the optical flow method) to track the changes of feature points in the image, calculate the offset and dancing amplitude of the power lines, and then obtain the dancing amplitude, wind angle, displacement, etc. Image processing network optimization specifically refers to dynamically adjusting the parameters of the image processing network through configured enhancement constraints to better process image data under current environmental conditions, including adjusting image enhancement technology, noise filtering, feature extraction, etc. to adapt to different wind speeds, wind directions and lighting conditions.

[0034] Preferably, adaptive enhancement of the monitoring data set is performed, that is, during the image processing process, the image processing process is adaptively adjusted according to environmental data (such as wind force, light, etc.) to improve the quality of the processing results, especially in scenes with drastic changes in lighting conditions and wind speed. Adaptive enhancement can ensure the extraction of accurate wind deviation and dancing data. For example, when the wind speed is high or the light is weak, the adaptive enhancement module can improve the contrast and sharpness of the image, or remove noise through a filtering algorithm to more clearly capture the dancing of the transmission line, and finally establish an adaptive enhancement result. The adaptive enhancement result refers to the image processing result after enhancement and optimization. After being constrained by wind data, optimized by the image processing network, and adaptively enhanced, an enhanced image data result containing dancing features is finally generated, including clear dancing trajectories, motion trajectories of key points, and data such as the dancing amplitude, wind deviation angle, and displacement of the transmission line under different wind speed conditions.

[0035] like Figure 2 As shown, the specific configuration of the image establishment enhancement module 20 also includes calling the edge detection channel of the image processing network to perform contour recognition and establish a basic contour recognition result; calling the local segmentation channel to perform similar pixel aggregation extraction of the monitoring data set and establish a basic local segmentation result; synchronizing the basic contour recognition result and the basic local segmentation result to the joint analysis channel to establish a regional growth starting point; calculating the similarity between the growth starting point and the adjacent pixel points, reconstructing the segmentation based on the similarity, establishing a reconstructed segmentation result, and using the reconstructed segmentation result and the enhancement constraint to perform local enhancement processing to complete adaptive enhancement.

[0036] Preferably, the edge detection channel is a module in the image processing network responsible for detecting the edges of objects in the image. The edge detection algorithm is used to identify the contours of the power lines and laser lines in the image. Specifically, algorithms (such as Canny, Sobel, Prewitt, etc.) are used to detect areas in the image where pixel values ​​change significantly, highlighting the edges of the power lines. On the edge image, a contour tracking algorithm (such as the chain code method) is used to extract continuous contour lines. The extracted contour lines are aggregated to ultimately form a basic contour recognition result, that is, preliminary contour information is obtained, which includes the position and shape information of the power lines in the image. The local segmentation channel refers to a module in the image processing network for dividing the image into several regions. Based on the similarity of pixels (such as color, brightness, texture), similar pixels are aggregated into regions. Specifically, an algorithm (such as the SLIC superpixel algorithm) is used to divide the image into multiple superpixel regions, each region containing similar pixels. Similar pixels are aggregated. Within each superpixel, pixels have similar features, which is conducive to maintaining the consistency of the local area of ​​the image. Ultimately, a superpixel segmentation image is generated, which is the basic local segmentation result and provides the regional information of the image.

[0037] Preferably, the joint analysis channel refers to a module for comprehensively analyzing the contour recognition and local segmentation results, and then determining the starting point of region growth. Specifically, the contour obtained by edge detection is aligned with the superpixel segmentation area, compared and fused, and high-confidence edge points are selected from the contour results as candidate seed points. Then, it is verified whether the seed point is located in the correct superpixel area to ensure that it is located in the transmission line area rather than the background area. The verified seed point is used as the starting point (starting position) of region growth; by comparing the similarity (such as color, brightness, gradient, etc.) between the starting point of region growth and the adjacent pixel points, it is determined whether the adjacent pixels are included in the growth area, where the adjacent pixel points are the pixel points around the starting point of region growth, and then reconstruction segmentation is performed based on the similarity calculation results, that is, starting from the seed point, similar adjacent pixels are included in the same area according to the similarity, the similarity is repeatedly calculated, and the area is gradually expanded until the stopping condition is met (such as the similarity is lower than the threshold), and finally an accurate image region segmentation is obtained, that is, a reconstructed segmentation result.

[0038] Preferably, the reconstructed segmentation results and enhancement constraints are used to perform local image enhancement processing on the target area. Specifically, according to external data (such as real-time wind data) and image features, image enhancement processing parameters are set, and the target area (transmission line part) obtained by reconstructing the segmentation is enhanced, such as contrast enhancement (increasing the contrast of the target area and highlighting details), sharpening (enhancing edges and improving contour clarity) and denoising (reducing noise interference and improving image quality), to complete adaptive enhancement, that is, dynamically adjusting the enhancement strategy according to the real-time environment and image data to ensure that the best effect can be obtained under different conditions, thereby improving the accuracy of subsequent feature extraction and dancing parameter calculation.

[0039] The specific configuration of the image establishment enhancement module 20 also includes performing gradient direction analysis of the contour points on the basic contour recognition results to generate standard consistency analysis results of the horizontal boundary segments; after screening the contour points of the standard consistency analysis results, re-screening the contour points using non-maximum suppression to establish initial seed points; using the segmentation labels of the basic local segmentation results to perform proximity authentication correction on the initial seed points, and establishing the starting point of regional growth based on the proximity authentication correction results.

[0040] Preferably, the gradient direction analysis of the contour points is performed on the basic contour recognition results, that is, the gradient direction of each contour point on the contour line is calculated to determine the edge direction of the point in the image. The gradient direction refers to the direction in which the grayscale change in the image is most significant. By analyzing the gradient directions of all contour points, horizontal line segments (that is, line segments with gradient directions close to horizontal) are screened out. Standard consistency analysis is used to determine whether these contour segments meet the preset boundary standards to ensure that the contour segments are consistent with the actual structures in the image (such as the horizontal sections of power lines); then, non-maximum suppression is used to further screen the contour points obtained by the standard consistency analysis to remove contour points that do not meet the conditions. Non-maximum suppression is a technology in image processing used to remove local non-extreme points to retain the most significant edge points. Specifically, by comparing the local gradient amplitudes of the contour points, the local maximum points are retained and those relatively weak edge points are removed to ensure that only the strongest edge points are left, and these contour points are used as initial seed points for the region growing algorithm.

[0041] Preferably, during the image segmentation process, each pixel is assigned a segmentation label. These segmentation labels indicate which area of ​​the image the pixel belongs to (such as the background or the target object, such as the power line), thereby determining which areas are important target areas and which are the background. Proximity authentication correction refers to comparing the position of the initial seed point with the segmentation labels of the surrounding pixels to verify whether the seed point falls in the correct area (i.e., whether it belongs to the area where the power line is located). Specifically, for each initially selected seed point (x, y), check whether the point belongs to the same area in the local segmentation result (such as the superpixel area has the same segmentation label). If the seed point is inconsistent with the label of the segmentation area (the seed point is in the wrong area), correction is performed by finding a neighboring similar pixel aggregation area (i.e., the power line area in the local segmentation result). The seed point is then adjusted using a distance metric or color similarity to ensure that the final seed point is located in the correct segmentation area. These contour points are used as the starting point for region growth, and the similar neighboring areas are expanded to ensure that the segmented area accurately covers the target object (such as the power line).

[0042] The specific configuration of the image creation enhancement module 20 also includes performing similarity calculation using a similarity calculation formula as follows:

[0043] ;

[0044] in, Characterizing the starting point of growth and adjacent pixels The similarity, and Characterize the growth starting point and adjacent pixels The color value of Characterizing the starting point of growth and adjacent pixels The gradient difference, , Starting point for growth The horizontal gradient of Brightness value in the direction The rate of change, For adjacent pixels The horizontal gradient, Starting point for growth The vertical gradient of Brightness value in the direction The rate of change, For adjacent pixels The vertical gradient of is the similarity of local texture patterns, , Starting point for growth or adjacent pixels The number of pixels in the local area created, Represent any pixel point in the local area, and The starting point of growth and adjacent pixels The corresponding eigenvalue represents the starting point of growth and adjacent pixels The texture features of the pixels in the local area centered are Characterizing the starting point of growth and adjacent pixels The local feature similarity score of , is the Euclidean distance between descriptors, To adjust the parameters, 、 、 、 They are the weight factors of color feature, gradient difference feature, local texture feature, and local similarity feature respectively.

[0045] The windage dance recognition module 30 is used to create windage dance data using the adaptive enhancement result. The windage dance data includes windage angle, vertical dance amplitude, horizontal dance amplitude, and ellipse tilt angle.

[0046] Preferably, when processing the wind-induced swing monitoring image of the transmission line, the key motion parameters of the transmission line are extracted by adaptively enhancing the image processing network, reflecting the offset and swing state of the transmission line under the action of wind. The wind-induced swing data specifically include the wind angle, vertical swing amplitude, horizontal swing amplitude, and elliptical tilt angle. The wind angle refers to the horizontal deflection angle of the transmission line relative to the initial static state under the action of wind, which usually reflects the overall swing direction of the transmission line under the action of wind. By detecting the relative position changes of the starting point and the end point of the transmission line in different frame images, the angle between the transmission line and the horizontal direction is calculated. The calculation formula is as follows:

[0047] ;

[0048] in, , , , Indicates the coordinates of the starting and ending points of the transmission line in a certain frame of image.

[0049] Preferably, the vertical waving amplitude refers to the vertical displacement of the transmission line, that is, the amplitude of the transmission line swinging up and down with the wind. In different frame images, a specific point of the transmission line (such as the starting point or the end point) is selected to calculate its vertical displacement difference. By comparing the frame images at multiple moments, the maximum amplitude of the vertical waving can be obtained. , the formula is as follows:

[0050] ;

[0051] in, It is the current moment The vertical coordinate of is the reference time The vertical coordinate of the transmission line; the horizontal gallop amplitude refers to the horizontal offset of the transmission line, reflecting the swing of the transmission line in the left and right directions. Similar to the vertical gallop amplitude, a specific point of the transmission line (such as the starting point or the end point) is selected, and the horizontal displacement difference of the point is calculated. By comparing the frame images, the maximum amplitude of the horizontal gallop can be calculated. , the formula is:

[0052] ;

[0053] in, It is the current moment The horizontal coordinate of is the reference time The horizontal coordinate of .

[0054] Preferably, the ellipse inclination angle refers to the inclination angle of the ellipse trajectory formed by the transmission line during the wind yaw process relative to a certain axis. Since the yaw of the transmission line under the action of wind usually presents an elliptical swinging trajectory, the inclination angle of the ellipse reflects the direction of the yaw trajectory. By analyzing the yaw trajectory of the transmission line at multiple time points, the ellipse shape of its motion trajectory is fitted, and the angle between the major axis of the ellipse and the horizontal line is calculated. Through multiple frame images in the time series, the trajectory of the feature point is fitted into an ellipse. The formula is:

[0055] ;

[0056] in, , is the change of the ellipse center over time, are the lengths of the major and minor axes; the inclination angle Represents the rotation angle of the ellipse. The tilt angle is calculated over time by fitting the parameters of the ellipse equation. The formula is:

[0057] ;

[0058] here, are the ellipse parameters fitted in each frame of the image. Tracking the changes of these parameters over time can capture the dynamic characteristics of the ellipse tilt angle.

[0059] The specific configuration of the wind yaw motion recognition module 30 also includes establishing an initial frame position of the stationary state of the transmission line; using the adaptive enhancement result to extract the updated frame position of the transmission line in each frame image; and using the initial frame position and the updated frame position to calculate the time series wind yaw angle as follows:

[0060] ;

[0061] in, Representational Moment The windage angle, is the initial time point, and Respectively indicate time The horizontal and vertical coordinates of the starting point of the transmission line, and Respectively indicate time The horizontal and vertical coordinates of the transmission line terminal, and Respectively indicate time The horizontal and vertical coordinates of the starting point of the transmission line, and Respectively indicate time The horizontal coordinates and vertical coordinates of the transmission line terminal are obtained; the calculated wind deviation angle is used as a kind of wind deviation swing data to establish wind deviation swing data.

[0062] Preferably, in the absence of wind interference, an image of the transmission line is captured and defined as the initial frame position. The initial frame position refers to the image frame when the transmission line is in a stationary state before the wind acts, and is used as a reference state (benchmark) for wind deviation and yaw monitoring. When there is no wind deviation, the position and angle of the transmission line will not change. In the initial frame, the coordinates of the starting point and the end point of the transmission line are determined for subsequent comparison and wind deviation angle calculation; then, the adaptive enhancement result (the processing result after each frame image enhancement) is used to extract the updated frame position of the transmission line, that is, the new position under the current wind force (the actual position at the current moment), and the initial frame position and the updated frame position are used to perform time-series wind deviation angle calculation. The formula is as follows:

[0063] ;

[0064] in, Representational Moment The windage angle, is the initial time point, and Respectively indicate time The horizontal and vertical coordinates of the starting point of the transmission line, and Respectively indicate time The horizontal and vertical coordinates of the transmission line terminal, and Respectively indicate time The horizontal and vertical coordinates of the starting point of the transmission line, and Respectively indicate time The horizontal and vertical coordinates of the transmission line terminal are calculated; the calculated wind yaw angle is used as a type of wind yaw fluctuation data to establish wind yaw fluctuation data. The time-series wind yaw angle refers to the deflection angle of the transmission line relative to its initial static state under the action of wind. The calculated wind yaw angle is used as a type of wind yaw fluctuation data. Specifically, the wind yaw angle of each frame is stored and added to the wind yaw fluctuation data set to form the wind yaw fluctuation data. In addition to the wind yaw angle, the wind yaw fluctuation data also includes vertical yaw amplitude, horizontal yaw amplitude, and elliptical tilt angle.

[0065] The early warning module 40 is used to perform environmental sensitivity threshold trigger analysis based on the wind yaw and fluctuation data, and report an early warning signal using the trigger analysis result.

[0066] Preferably, the environmental sensitive threshold trigger analysis refers to comparing the wind deviation dance data with the current threshold. When a dance parameter (such as wind deviation angle or dance amplitude) exceeds the preset threshold, the system triggers the analysis result to determine whether the transmission line is in a dangerous state and whether an early warning is required. The environmental sensitive threshold is dynamically adjusted according to real-time environmental conditions (such as wind speed, wind direction, temperature, etc.). For example, in extreme weather such as strong winds and gale, the threshold is lowered to improve the sensitivity of monitoring; in the case of weak wind, the threshold is increased to reduce false alarms; if the trigger analysis result determines that the dance parameters such as the dance amplitude and wind deviation angle of the transmission line exceed the safety threshold, the analysis result is judged to be abnormal and a warning signal is automatically generated to prevent the transmission line from breaking, discharging and other accidents due to excessive dance or deflection.

[0067] The line wind deviation and sway monitoring and early warning device combining a straight-line laser with monocular vision also includes an early warning database construction module, which is used to establish an environmentally sensitive threshold database mapped to the weather based on the weather history database. When wind deviation and sway data are generated, the early warning module calls the environmentally sensitive threshold database based on real-time weather information to perform environmentally sensitive threshold trigger analysis.

[0068] Preferably, weather information (such as wind speed, wind direction, temperature, etc.) is associated with monitoring parameters of wind yaw (such as wind yaw angle, yaw amplitude, etc.) to establish an environmentally sensitive threshold database that can be dynamically adjusted. Specifically, based on the weather information in the weather history database, it is associated with the wind yaw yaw situation of the transmission line under different weather conditions and mapped. A reasonable safety threshold is set for each weather condition to form an environmentally sensitive threshold database, including the upper limit of the wind yaw angle, the limit of the yaw amplitude, etc. For example, when the wind speed is 5-10m / s, the wind yaw angle threshold may be set to When the wind speed exceeds 15m / s, the wind deviation angle threshold may be reduced to 3 degrees to improve the monitoring sensitivity. When the wind deviation swing monitoring parameters of the transmission line (wind deviation angle, vertical swing amplitude, horizontal swing amplitude, etc.) are collected through real-time monitoring, the early warning module will call out the threshold matching the current weather conditions from the environmental sensitive threshold database based on real-time weather data, perform environmental sensitive threshold trigger analysis, and determine whether the current wind deviation swing data exceeds the safety range. When the trigger analysis determines that the wind deviation swing data exceeds the safety threshold, a warning signal will be issued.

[0069] The line wind deviation and sway monitoring and early warning device combining a straight line laser with monocular vision also includes a self-update module for extracting the feedback signal of the early warning signal and establishing a self-update signal, and updating the environmental sensitive threshold database in the early warning database construction module based on the self-update signal.

[0070] Preferably, after the warning signal is issued, a feedback signal is obtained based on the actual situation. That is, by recording the actual response after the warning (such as maintenance results, monitoring results, etc.), these feedback signals are extracted. These feedback signals reflect the accuracy and response of the warning. For example, if no abnormality is found after a warning, the warning is recorded as a false alarm. If a problem is found and measures are taken after the warning, the warning is recorded as effective. By analyzing the warning feedback signal, a corresponding self-update signal is generated. The self-update signal refers to a signal established based on the feedback signal for optimizing and updating system parameters. It is used to adjust according to the feedback information so that the sensitivity and accuracy of the warning are continuously optimized. The self-update signal dynamically adjusts the parameters in the environmental sensitive threshold database based on the accuracy of the warning and the information in the feedback signal. For example, if there are multiple false alarms under certain weather conditions, the threshold for that condition is appropriately relaxed. Specifically, after receiving the self-update signal, the information contained in the signal is analyzed (such as whether the threshold is too strict or loose, whether the warning signal is a false alarm, etc.). Based on this information, the corresponding threshold in the environmental sensitive threshold database is automatically adjusted, so that the environmental sensitive threshold database can continuously self-optimize over time, thereby improving the accuracy and reliability of the warning system.

[0071] Although this application makes various references to certain components in the apparatus according to the embodiments of this application, any number of different components may be used and run on the user terminal and / or server, and the various units and components included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other and are not used to limit the scope of protection of the present invention.

[0072] The above specific embodiments do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application shall be included within the scope of protection of this application.

Claims

1. A line wind deviation and sway monitoring and early warning device combining a straight line laser and monocular vision is characterized by: The device comprises: A setting module is used to determine key monitoring points in the transmission line area and deploy a joint monitoring group at the key monitoring points. The joint monitoring group integrates a line laser and a monocular camera. The projection direction of the line laser is set along the transverse direction of the transmission line to constrain the laser beam of the line laser to cover the key monitoring points. The optical axis of the monocular camera is set at a predetermined angle with the projection plane of the laser beam, and the monocular camera can completely capture the intersection of the laser beam and the key monitoring points. The image enhancement module is used to activate the joint monitoring group, read the monitoring data set of the monocular camera, and simultaneously read the real-time wind data of the wind monitoring sensor. It configures the enhancement constraints based on the real-time wind data, optimizes the image processing network using the enhancement constraints, performs adaptive enhancement of the monitoring data set, and creates the adaptive enhancement results. A wind yaw and dance recognition module is used to establish wind yaw and dance data using the adaptive enhancement result, wherein the wind yaw and dance data includes wind yaw angle, vertical dance amplitude, horizontal dance amplitude, and elliptical tilt angle; The early warning module is used to perform environmental sensitivity threshold trigger analysis based on wind yaw and fluctuation data, and use the trigger analysis results to issue early warning signals; The method of optimizing the image processing network using the enhancement constraint and then performing adaptive enhancement of the monitoring data set further includes: Call the edge detection channel of the image processing network to perform contour recognition and establish basic contour recognition results; Call the local segmentation channel to perform similar pixel aggregation extraction of the monitoring data set and establish the basic local segmentation result; Synchronize the basic contour recognition results and basic local segmentation results to the joint analysis channel to establish the starting point of region growth; Calculate the similarity between the growth starting point and the adjacent pixel points, reconstruct the segmentation based on the similarity, establish the reconstructed segmentation result, and perform local enhancement processing using the reconstructed segmentation result and enhancement constraints to complete adaptive enhancement; The wind deviation and dancing recognition module is further used for: Establishing an initial frame position of a stationary state of the transmission line; The adaptive enhancement results are used to extract the transmission line in each frame image and update the frame position. The initial frame position and the updated frame position are used to calculate the time series wind angle as follows: ； in, Representational Moment The windage angle, is the initial time point, and Respectively indicate time The horizontal and vertical coordinates of the starting point of the transmission line, and Respectively indicate time The horizontal and vertical coordinates of the transmission line terminal, and Respectively indicate time The horizontal and vertical coordinates of the starting point of the transmission line, and Respectively indicate time The horizontal and vertical coordinates of the transmission line terminal; The calculated windage angle is used as a type of windage dance data to establish windage dance data.

2. The line wind deviation and sway monitoring and early warning device combining a straight line laser and monocular vision according to claim 1 is characterized in that: The step of synchronizing the basic contour recognition results and the basic local segmentation results to the joint analysis channel to establish the region growth starting point further includes: Performing gradient direction analysis of contour points on the basic contour recognition result to generate standard consistency analysis results of horizontal boundary segments; After screening the contour points of the standard consistency analysis results, the contour points are re-screened using non-maximum suppression to establish the initial seed points; The segmentation labels of the basic local segmentation results are used to perform proximity authentication correction on the initial seed points, and the starting point of region growing is established based on the proximity authentication correction results.

3. The line wind deviation and sway monitoring and early warning device combining a straight line laser and monocular vision as claimed in claim 1 is characterized in that: The calculating of the similarity between the growth starting point and the adjacent pixel points, reconstructing the segmentation based on the similarity, and establishing the reconstructed segmentation result further includes: The similarity calculation formula is used to calculate the similarity as follows: ； in, Characterizing the starting point of growth and adjacent pixels The similarity, and Characterize the growth starting point and adjacent pixels The color value of Characterizing the starting point of growth and adjacent pixels The gradient difference, , Starting point for growth The horizontal gradient of Brightness value in the direction The rate of change, For adjacent pixels The horizontal gradient, Starting point for growth The vertical gradient of Brightness value in the direction The rate of change, For adjacent pixels The vertical gradient of is the similarity of local texture patterns, , Starting point for growth or adjacent pixels The number of pixels in the local area created, Represent any pixel point in the local area, and The starting point of growth and adjacent pixels The corresponding eigenvalue represents the starting point of growth and adjacent pixels The texture features of the pixels in the local area centered are Characterizing the starting point of growth and adjacent pixels The local feature similarity score of , is the Euclidean distance between descriptors, To adjust the parameters, 、 、 、 They are the weight factors of color feature, gradient difference feature, local texture feature, and local similarity feature respectively.

4. The line wind deviation and sway monitoring and early warning device combining a straight line laser and monocular vision as claimed in claim 1 is characterized in that: The predetermined angle ranges from 10° to 30°.

5. The line wind deviation and sway monitoring and early warning device combining a straight line laser and monocular vision as claimed in claim 1 is characterized in that: The device further comprises: The early warning database construction module is used to establish an environmentally sensitive threshold database mapped to the weather based on the weather history database. When wind deviation and fluctuation data are generated, the early warning module calls the environmentally sensitive threshold database based on real-time weather information to perform environmentally sensitive threshold trigger analysis.

6. The line wind deviation and sway monitoring and early warning device combining a straight line laser and monocular vision as claimed in claim 1 is characterized in that: The device further comprises: The self-update module is used to extract the feedback signal of the early warning signal and establish a self-update signal, and update the environmental sensitive threshold database in the early warning database construction module based on the self-update signal.

Images