An amplitude quantization method and system for galloping video analysis of power transmission lines

Abstract

The application provides an amplitude quantization method and system for galloping video analysis of a power transmission line, and belongs to the technical field of data processing and data representation. The conductor closest to a tracking point is taken as the conductor to which the tracking point belongs, the ordinate of the tracking point is substituted into the scale equation of the conductor to which the tracking point belongs, and the scale of the conductor to which the tracking point belongs at the tracking point is obtained; the two-dimensional pixel amplitude is converted into three-dimensional real amplitude according to the scale; the three-dimensional point cloud data and the monocular camera video are combined for use, the three-dimensional linear diameter of the single-phase conductor and the continuous two-dimensional linear diameter are calculated, the scale of the conductor at each position is solved, the function of quantizing the pixel amplitude obtained by analyzing the conductor galloping video into real physical amplitude is realized, and the problem of difficult monocular camera ranging marking in the field is solved.

Description

Technical Field

[0001] This invention relates to the field of power line monitoring technology, and in particular to an amplitude quantification method and system for video analysis of power line galloping. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] Transmission line galloping is a power grid disaster caused by natural factors, which seriously threatens the safe and stable operation of the power grid. Therefore, effective transmission line galloping monitoring schemes have become a research focus. Currently, transmission line galloping monitoring schemes are mainly divided into two categories: sensor-based galloping monitoring schemes and video analysis-based galloping monitoring schemes.

[0004] Sensor-based galloping monitoring solutions primarily analyze the degree of conductor galloping by installing acceleration, tilt, and tension sensors along the power line. While these solutions offer high accuracy, they require sensor installation on the conductors and suffer from issues such as numerous sensor locations, short power supply times, and unstable data transmission, resulting in high construction and maintenance costs and hindering large-scale deployment. Video analytics-based galloping monitoring solutions mainly utilize cameras to capture video footage of transmission lines and analyze the degree of conductor galloping within the video. These solutions offer advantages such as being non-contact, low-cost, and reusable; however, the acquired video footage lacks accurate physical dimensions, presenting a technical bottleneck in quantifying galloping amplitude.

[0005] Galloping amplitude quantization mainly refers to converting the amplitude of the galloping image of a conductor analyzed from video into the physical amplitude of the real world. Due to the limitations of the field environment, binocular ranging has low accuracy and poor robustness; therefore, research focuses mainly on monocular vision, using calibration or other prior information to solve for the scale of distance transformation, thereby achieving amplitude quantization.

[0006] Reference 1, "Transmission Line Galloping Measurement Based on Monocular Vision Analysis," describes a method for calculating the actual amplitude of conductor galloping by calibrating a monocular camera to determine its intrinsic and extrinsic parameters. Reference 2, "A Short Video-Based Method for Detecting Transmission Line Galloping," describes a method for amplitude quantization: selecting key points on the conductor to be detected within the field of view and recording their coordinates and conversion ratio η, where conversion ratio η = actual length of the key point / pixel length of the key point; then converting the pixel amplitude obtained from optical flow analysis into the actual amplitude according to the conversion ratio η. Reference 3, "Research on Transmission Line Galloping Monitoring Technology Based on Video Tracking," uses a monocular vision geometric similarity method to achieve amplitude quantization. It selects a spacer bar as a standard component with a known true length, calculates the ratio of its true length to the pixel scale, obtains the true length of each pixel in the image mapped to the actual space, and then multiplies it by the pixel distance the spacer bar moves in the image to finally obtain the actual distance the object deviates from its equilibrium position.

[0007] The method described in Reference 1 is a conventional monocular camera calibration method. Its drawback is that the distance from the camera to the measured point and the camera's elevation angle are unknown during the calculation process and need to be measured on-site. It also has strict requirements regarding the camera's installation angle. The methods described in References 2 and 3 are essentially the same: selecting key points for calibration and calculating the conversion ratio at those key points. Their disadvantage is that the calibration of key points is entirely manual and only the distance conversion ratio at those key points can be obtained, not a continuous scale at any location on the traverse line. Summary of the Invention

[0008] To address the shortcomings of existing technologies, this invention provides an amplitude quantization method and system for video analysis of power transmission line galloping. It combines three-dimensional point cloud data with monocular camera video, calculates the three-dimensional diameter and continuous two-dimensional diameter of a single-phase conductor, and solves for the scale at various points on the conductor. This enables the quantization of pixel amplitudes obtained from conductor galloping video analysis into real physical amplitudes, solving the problem of difficult distance measurement and annotation using monocular cameras in this field.

[0009] To achieve the above objectives, the present invention adopts the following technical solution:

[0010] The first aspect of this invention provides an amplitude quantization method for video analysis of power transmission line galloping.

[0011] An amplitude quantization method for video analysis of power transmission line galloping includes the following steps:

[0012] Generate the scale equation;

[0013] The traverse closest to the tracking point is taken as the traverse to which the tracking point belongs. The ordinate of the tracking point is substituted into the scale equation of the traverse to which the tracking point belongs to obtain the scale of the traverse to which the tracking point belongs at the tracking point. Based on the scale, the two-dimensional pixel amplitude is converted into the three-dimensional real amplitude.

[0014] The generation of the scale equation includes:

[0015] Acquire 3D point cloud data of the power transmission line scene after processing, where each point includes 3D world coordinates and 2D image coordinates;

[0016] The point cloud is clustered based on the number of conductor clusters in the current scene, and each point is assigned to the corresponding conductor cluster.

[0017] Calculate the two-dimensional approximate straight line equation and its normal vector for each cluster of wires in the image;

[0018] Based on the two-dimensional approximate straight line equation and its normal vector, and using the integer values ​​of the two-dimensional image ordinates of each point as indices, calculate the two-dimensional line diameter of each cluster of wires at all two-dimensional image ordinates.

[0019] Calculate the approximate vector of each cluster of wires in three-dimensional space and the normal plane of that vector at the origin;

[0020] Based on the approximate vector of each cluster of conductors in three-dimensional space and the normal plane of that vector at the origin, the integer values ​​of the ordinates of the two-dimensional images of each point are used as indices to calculate the three-dimensional diameter of each cluster of conductors at all ordinates of the two-dimensional images, and the average value is taken as the overall three-dimensional diameter of each cluster of conductors.

[0021] Calculate the scale of each cluster of conductors at the vertical coordinate of all two-dimensional images, and obtain the scale equation after fitting.

[0022] As an optional implementation, the processed 3D point cloud data includes:

[0023] Point cloud data containing only the guide wires in the scene is obtained by calibrating, calculating, and filtering the original 3D point cloud data and video images.

[0024] As an optional implementation, the two-dimensional approximate straight line equation and its normal vector for each cluster of wires in the image are calculated, including:

[0025] Based on the vertical coordinate of the two-dimensional image, remove the portion of each cluster of wires at the bottom of the image with a predetermined proportion, and calculate the range of values ​​for the vertical coordinate of the remaining portion of each cluster of wires in the two-dimensional image, denoted as the coordinate range.

[0026] For each cluster of conductors, a straight line is fitted using the two-dimensional coordinates of the point cloud of each cluster of conductors to obtain the fitted two-dimensional approximate straight line equation and its normal vector.

[0027] As a further limitation, based on the two-dimensional approximate straight line equation and its normal vector, and using the integer values ​​of the two-dimensional image ordinates of each point as indices, the two-dimensional diameter of each cluster of conductors at all two-dimensional image ordinates is calculated, including:

[0028] For any traverse, calculate the foot coordinates of the perpendiculars from the two-dimensional coordinates of all points on the traverse to the two-dimensional approximate line of the traverse;

[0029] For any integer ordinate of a two-dimensional image that falls within the range of coordinate values, points whose ordinates differ from the ordinate of the perpendicular by less than n are taken as points in the neighborhood of the ordinate of the two-dimensional image, where n is half the size of the neighborhood.

[0030] For any integer ordinate of the two-dimensional image within the range of coordinate values, take any two points in the neighborhood to form a two-dimensional vector, calculate the projection length of the two-dimensional vector onto the normal vector of the two-dimensional approximate straight line of the conductor, and traverse all two-dimensional vectors to find the maximum projection length, which is the two-dimensional diameter of the conductor at the ordinate of this two-dimensional image.

[0031] The mean filter is used to smooth the two-dimensional line diameter at each point, resulting in the final two-dimensional line diameter at the ordinate of the two-dimensional image.

[0032] As a further limitation, the approximate vector of each cluster of wires in three-dimensional space and the normal plane of that vector at the origin are calculated, including:

[0033] For any cluster of conductors, calculate the coordinates of the three-dimensional approximate centroid of the two endpoints of the conductor;

[0034] The line connecting the two ends of the conductor and their approximate centroids in three dimensions is taken as the approximate vector of the conductor in three-dimensional space. Based on the approximate vector of the conductor in three-dimensional space, the equation of the normal plane of the current cluster of conductors at the origin is obtained.

[0035] As a further limitation, the coordinates of the three-dimensional approximate centroids of the two endpoints of the conductor are calculated, including:

[0036] Calculate the Y-coordinate of the conductor in the two-dimensional image. min and Y max The coordinates of the approximate three-dimensional centroids of the two endpoints are obtained by taking the points within the neighborhood of the endpoint and calculating the average of the three-dimensional coordinates of the points in the two neighborhoods respectively, where Y... min and Y max These are the minimum and maximum endpoint values ​​of the coordinate range, respectively.

[0037] As a further limitation, based on the approximate vector of each cluster of conductors in three-dimensional space and the normal plane of that vector at the origin, using the integer values ​​of the ordinates of the two-dimensional image at each point as indices, the three-dimensional diameter of each cluster of conductors at all ordinates of the two-dimensional image is calculated, and the average value is taken as the overall three-dimensional diameter of each cluster of conductors, including:

[0038] For any integer ordinate of a two-dimensional image that falls within the range of coordinate values, calculate the points of the conductor within the neighborhood of this ordinate of the two-dimensional image.

[0039] For any integer ordinate of a two-dimensional image within the range of coordinate values, take any two points in the neighborhood to form a three-dimensional vector, and calculate the projection length of the three-dimensional vector onto the normal plane of the three-dimensional approximation vector of the conductor.

[0040] The maximum value of the projected length obtained by traversing all three-dimensional vectors is the three-dimensional diameter of the conductor at the vertical coordinate of this two-dimensional image.

[0041] The average value of the three-dimensional line diameter at the vertical coordinate of all two-dimensional images is taken as the overall three-dimensional line diameter of the conductor. All integer values ​​at the vertical coordinate of the two-dimensional images are within the range of coordinate values.

[0042] As an optional implementation, the scale of each cluster of conductors at the ordinate of all two-dimensional images is calculated, and the scale equation is obtained after fitting, including:

[0043] Using the integer values ​​of the vertical coordinates of the two-dimensional image as indices, calculate the ratio of the three-dimensional wire diameter of each cluster of wires to the two-dimensional wire diameter at all vertical coordinates of the two-dimensional image, thus obtaining the scale of each cluster of wires at the vertical coordinates of the two-dimensional image.

[0044] For a certain cluster of conductors, the vertical coordinate of a two-dimensional image is used as the independent variable, and the scale at the vertical coordinate of the two-dimensional image is used as the dependent variable. The least squares method is used to fit the cubic equation of the dependent and independent variables, and the scale equation is obtained after fitting.

[0045] A second aspect of the present invention provides an amplitude quantization system for video analysis of power transmission line galloping.

[0046] An amplitude quantization system for video analysis of power transmission line galloping includes:

[0047] The scale equation generation module is configured to generate scale equations.

[0048] The amplitude conversion module is configured to: take the traverse closest to the tracking point as the traverse to which the tracking point belongs, substitute the ordinate of the tracking point into the scale equation of the traverse to which the tracking point belongs, obtain the scale of the traverse to which the tracking point belongs at the tracking point, and convert the two-dimensional pixel amplitude into the three-dimensional real amplitude according to the scale.

[0049] The generation of the scale equation includes:

[0050] Acquire 3D point cloud data of the power transmission line scene after processing, where each point includes 3D world coordinates and 2D image coordinates;

[0051] The point cloud is clustered based on the number of conductor clusters in the current scene, and each point is assigned to the corresponding conductor cluster.

[0052] Calculate the two-dimensional approximate straight line equation and its normal vector for each cluster of wires in the image;

[0053] Based on the two-dimensional approximate straight line equation and its normal vector, and using the integer values ​​of the two-dimensional image ordinates of each point as indices, calculate the two-dimensional line diameter of each cluster of wires at all two-dimensional image ordinates.

[0054] Calculate the approximate vector of each cluster of wires in three-dimensional space and the normal plane of that vector at the origin;

[0055] Based on the approximate vector of each cluster of conductors in three-dimensional space and the normal plane of that vector at the origin, the integer values ​​of the ordinates of the two-dimensional images of each point are used as indices to calculate the three-dimensional diameter of each cluster of conductors at all ordinates of the two-dimensional images, and the average value is taken as the overall three-dimensional diameter of each cluster of conductors.

[0056] Calculate the scale of each cluster of conductors at the vertical coordinate of all two-dimensional images, and obtain the scale equation after fitting.

[0057] As an optional implementation, the scale of each cluster of conductors at the ordinate of all two-dimensional images is calculated, and the scale equation is obtained after fitting, including:

[0058] Using the integer values ​​of the vertical coordinates of the two-dimensional image as indices, calculate the ratio of the three-dimensional wire diameter of each cluster of wires to the two-dimensional wire diameter at all vertical coordinates of the two-dimensional image, thus obtaining the scale of each cluster of wires at the vertical coordinates of the two-dimensional image.

[0059] For a certain cluster of conductors, the vertical coordinate of a two-dimensional image is used as the independent variable, and the scale at the vertical coordinate of the two-dimensional image is used as the dependent variable. The least squares method is used to fit the cubic equation of the dependent and independent variables, and the scale equation is obtained after fitting.

[0060] Compared with the prior art, the beneficial effects of the present invention are:

[0061] 1. The amplitude quantization method and system for transmission line galloping video analysis described in this invention combines three-dimensional point cloud data and monocular camera video. By calculating the three-dimensional diameter and continuous two-dimensional diameter of a single-phase conductor, the scale of each part of the conductor is determined. This realizes the function of quantizing the pixel amplitude obtained from conductor galloping video analysis into the real physical amplitude, solving the problem of difficult distance measurement and annotation by monocular cameras in this field.

[0062] 2. The amplitude quantization method and system for transmission line galloping video analysis described in this invention obtains a continuous scale on any conductor through calculation, so that conductor galloping analysis and amplitude quantization are no longer limited to manually selected key points, and can analyze more comprehensive galloping information in practical applications. Attached Figure Description

[0063] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0064] Figure 1 This is a flowchart illustrating the amplitude quantization method for video analysis of power transmission line galloping provided in Embodiment 1 of the present invention.

[0065] Figure 2 This is a schematic diagram of the two-dimensional line diameter changing with Y2 before and after smoothing, as provided in Embodiment 1 of the present invention.

[0066] Figure 3 This is a schematic diagram of the calculation results of the three-dimensional wire diameter of the conductor provided in Embodiment 1 of the present invention;

[0067] Figure 4 This is a schematic diagram of the curve showing the change of the guide scale with Y2 provided in Embodiment 1 of the present invention;

[0068] Figure 5 This is a schematic diagram of the amplitude quantization system for power transmission line galloping video analysis provided in Embodiment 2 of the present invention. Detailed Implementation

[0069] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0070] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0071] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0072] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0073] Example 1:

[0074] like Figure 1 As shown, Embodiment 1 of the present invention provides an amplitude quantization method for video analysis of power transmission line galloping, comprising the following steps:

[0075] A1: Take the processed 3D point cloud data of a power transmission line scene as the input data.

[0076] The "processed 3D point cloud data" refers to point cloud data that retains only the guide lines in the scene. It is obtained from the original 3D point cloud data and video images through calibration, calculation, and filtering. Its format is shown in Table 1. The first row contains a single number representing the number of point cloud data points. Each subsequent row contains five numbers describing the 3D world coordinates and 2D image coordinates of a point, in the order X3, Y3, Z3, X2, Y2. Here, (X3, Y3, Z3) represents the 3D world coordinates of the point in meters, and (X2, Y2) represents the 2D image coordinates of the point in pixels. Specifically, in this embodiment, the 2D image coordinate X2 represents the horizontal coordinate of the 2D image, and the 2D image coordinate Y2 represents the vertical coordinate of the 2D image.

[0077] Table 1: Processed 3D point cloud data

[0078]

[0079] In this embodiment, a cluster of conductors refers to the entirety of a single-phase conductor. Multiple strands of conductors on a multi-split spacer bar constitute a cluster of conductors, as do single strands of conductors without spacers.

[0080] A2: The point cloud is clustered based on the number of conductor clusters in the current scene, assigning each point to a conductor cluster. The clustering uses the three-dimensional coordinates of the points as basic data elements. In this embodiment, the clustering method uses the agglomerative clustering algorithm.

[0081] A3: Calculate the two-dimensional approximate straight line equation and its normal vector for each cluster of wires in the image.

[0082] A3.1: Based on the Y2 coordinate, remove a portion of each cluster of wires at the bottom of the image according to a predetermined proportion, and calculate the range of Y2 coordinate values ​​for the remaining portion of the wires, denoted as [Y...]. min ,Y max In this embodiment, the bottom 15% (i.e., the set ratio is 15%) is removed. It is understood that in other embodiments, those skilled in the art can also select the set ratio according to the specific working conditions, such as 10%, 20%, etc., which will not be elaborated here.

[0083] A3.2: For each cluster of traverse lines, a straight line is fitted using the two-dimensional coordinates of the traverse point cloud. The fitting method is the least squares method. The equation of the fitted line is of the form Y² = k²X² + ​​b². Then, find a normal vector of this line. It can be represented as (-k2,1).

[0084] A4: Using the integer values ​​of the Y2 coordinates as indices, calculate the two-dimensional wire diameter of each cluster of wires at all Y2 coordinates.

[0085] A4.1: Calculate the foot coordinates of the perpendiculars from the two-dimensional coordinates of all points on the traverse to the two-dimensional approximate line of the traverse.

[0086] A4.2: For any integer Y², ∈ [Y min ,Y max Points whose y-coordinates differ from Y2 by less than n are taken as points in the neighborhood of Y2. Here, n is defined as half the size of the neighborhood; in this embodiment, n is set to 5.

[0087] A4.3: For any integer Y², ∈ [Y min ,Y max ], take any two points in the neighborhood to form a two-dimensional vector. calculate The projection length onto the normal vector of the two-dimensional approximate straight line of the conductor (note: this normal vector is obtained in A3.2); traversing all two-dimensional vectors The maximum value of the projected length is the two-dimensional wire diameter of the conductor at this Y2 coordinate.

[0088] A4.4: Mean filtering is used to smooth the 2D line diameter at various points. In this embodiment, the filter window size is 11, and the curves showing the 2D line diameter before and after smoothing as a function of Y2 are as follows: Figure 2 As shown, the horizontal axis represents the Y2 coordinate value of the conductor, and the vertical axis represents the two-dimensional diameter of the conductor, in pixels; the two curves are the two-dimensional diameter curves of the conductor before and after smoothing.

[0089] A5: Calculate the approximate vector of each cluster of wires in three-dimensional space and the normal plane of that vector at the origin.

[0090] A5.1: Calculate the coordinates of the three-dimensional approximate centroids of the two endpoints of the conductor, denoted as P and Q.

[0091] In this embodiment, the method for calculating the approximate centroid of the two endpoints of the conductor is as follows: using the method described in A4.2, calculate the centroid of the conductor at points Y2 equal to Y... min and Y max The coordinates of the three-dimensional approximate centroids of the two endpoints are obtained by finding the points in the neighborhood of the two endpoints and calculating the average of the three-dimensional coordinates of the points in the neighborhood of the two endpoints.

[0092] A5.2: Denote vectors Let (A3, B3, C3) be the approximate vector of the conductor in three-dimensional space. Then the equation of its normal plane at the origin is A3x + B3y + C3z + D3 = 0, where D3 = 0.

[0093] A6: Using the integer values ​​of the Y2 coordinates as indices, calculate the three-dimensional diameter of each cluster of conductors at all Y2 coordinates, and take the average value as the overall three-dimensional diameter of each cluster of conductors.

[0094] A6.1: For any integer Y², ∈ [Y min ,Y max Using the method described in A4.2, calculate the points of the conductor in the Y2 neighborhood.

[0095] A6.2: For any integer Y2, ∈ [Y min ,Y max Take any two points in the neighborhood and form a three-dimensional vector, then calculate... The projection length onto the normal plane of the three-dimensional approximation vector of the conductor (note: this normal plane is obtained in A5.2); traversing all three-dimensional vectors The maximum value of the projected length is the three-dimensional diameter of the conductor at this Y2 coordinate.

[0096] A6.3: Take all integers Y2 belonging to [Y min ,Y max The average three-dimensional wire diameter at the specified location is taken as the overall three-dimensional wire diameter of the conductor. The calculation results of the conductor's three-dimensional wire diameter are as follows: Figure 3 As shown, the horizontal axis represents the Y2 coordinate value of the conductor, and the vertical axis represents the three-dimensional diameter of the conductor, in meters; the curve in the figure is the curve of the three-dimensional diameter of the conductor changing with Y2, and the straight line is the overall diameter result after taking the average value.

[0097] A7: Calculate the scale of each cluster of conductors at various locations and store it as a cubic curve equation.

[0098] A7.1: Using the integer values ​​of the Y2 coordinates as indices, calculate the ratio of the three-dimensional wire diameter of each cluster of conductors to the two-dimensional wire diameter at all Y2 coordinates. This ratio is the scale of each cluster of conductors at Y2.

[0099] A7.2: Using Y2 as the independent variable and the scale S at Y2 as the dependent variable, fit a cubic equation of dependent and independent variables using the least squares method, expressed as S = aY2. 3 +bY2 2 +cY2+d, storing the range of values ​​for coefficients a, b, c, d, and Y2 [Y min ,Y max The fitted equation is described by [a formula]; simultaneously, the coefficients k2 and b of the two-dimensional approximate straight line equation obtained from A3.2 are stored to describe the two-dimensional approximate straight line of the traverse. The curve showing the traverse scale changing with Y2 is as follows: Figure 4 As shown, the horizontal axis represents the Y2 coordinate value of the traverse, and the vertical axis represents the scale of the transformation from two-dimensional to three-dimensional dimensions, in meters per pixel; the two curves are the traverse scale curves before and after fitting, respectively.

[0100] A8: When it is necessary to convert the two-dimensional pixel amplitude of a tracking point on a conductor into a three-dimensional true amplitude, first calculate the distance from the tracking point to each cluster of conductors, take the conductor with the closest distance as the conductor to which the tracking point belongs, and then use the y coordinate of the tracking point to substitute into the conductor scale equation obtained in A7.2 to obtain the scale at that point on the conductor, and then the two-dimensional pixel amplitude can be converted into a three-dimensional true amplitude.

[0101] Example 2:

[0102] like Figure 5 As shown, Embodiment 2 of the present invention provides an amplitude quantization system for video analysis of transmission line galloping, comprising:

[0103] The scale equation generation module is configured to generate scale equations.

[0104] The amplitude conversion module is configured to: take the traverse closest to the tracking point as the traverse to which the tracking point belongs, substitute the ordinate of the tracking point into the scale equation of the traverse to which the tracking point belongs, and obtain the scale of the traverse to which the tracking point belongs at the tracking point. The amplitude conversion module is configured to: convert the two-dimensional pixel amplitude into the three-dimensional real amplitude according to the scale.

[0105] Generating the scale equation includes the following steps:

[0106] B1: Take the processed 3D point cloud data of a power transmission line scene as input data.

[0107] The "processed 3D point cloud data" refers to point cloud data that retains only the guide lines in the scene. It is obtained from the original 3D point cloud data and video images through calibration, calculation, and filtering. Its format is shown in Table 1 of Example 1. The first row contains a single number representing the number of point cloud data points. Each subsequent row contains five numbers describing the 3D world coordinates and 2D image coordinates of a point, in the order X3, Y3, Z3, X2, Y2. Here, (X3, Y3, Z3) represents the 3D world coordinates of the point in meters, and (X2, Y2) represents the 2D image coordinates of the point in pixels. Specifically, in this embodiment, the 2D image coordinate X2 represents the horizontal coordinate of the 2D image, and the 2D image coordinate Y2 represents the vertical coordinate of the 2D image.

[0108] In this embodiment, a cluster of conductors refers to the entirety of a single-phase conductor. Multiple strands of conductors on a multi-split spacer bar constitute a cluster of conductors, as do single strands of conductors without spacers.

[0109] B2: Cluster the point cloud based on the number of conductor clusters in the current scene, assigning each point to a conductor cluster. The clustering uses the three-dimensional coordinates of the points as basic data elements. In this embodiment, the clustering method uses the agglomerative clustering algorithm.

[0110] B3: Calculate the two-dimensional approximate straight line equation and its normal vector for each cluster of wires in the image.

[0111] B3.1: Based on the Y2 coordinate, remove a portion of each cluster of conductors at the bottom of the image according to a predetermined proportion, and calculate the range of Y2 coordinate values ​​for the remaining portion of the conductors, denoted as [Y...]. min ,Y max In this embodiment, the bottom 15% (i.e., the set ratio is 15%) is removed. It is understood that in other embodiments, those skilled in the art can also select the set ratio according to the specific working conditions, such as 10%, 20%, etc., which will not be elaborated here.

[0112] B3.2: For each cluster of traverse lines, perform a straight line fitting using the two-dimensional coordinates of the traverse point cloud. The fitting method is the least squares method. The equation of the fitted line is of the form Y² = k²X² + ​​b². Then, find a normal vector of this line. It can be represented as (-k2,1).

[0113] B4: Using the integer values ​​of the Y2 coordinates as indices, calculate the two-dimensional wire diameter of each cluster of wires at all Y2 coordinates.

[0114] B4.1: Calculate the foot coordinates of the perpendiculars from the two-dimensional coordinates of all points on the traverse to the two-dimensional approximate line of the traverse.

[0115] B4.2: For any integer Y2, ∈ [Y min ,Y max Points whose y-coordinates differ from Y2 by less than n are taken as points in the neighborhood of Y2. Here, n is defined as half the size of the neighborhood; in this embodiment, n is set to 5.

[0116] B4.3: For any integer Y2, ∈ [Y min ,Y max ], take any two points in the neighborhood to form a two-dimensional vector. calculate The projection length onto the normal vector of the two-dimensional approximate straight line of the conductor (note: this normal vector is obtained in B3.2); traversing all two-dimensional vectors The maximum value of the projected length is the two-dimensional diameter of the conductor at this Y2 coordinate.

[0117] B4.4: Use mean filtering to smooth the 2D line diameter at various points. In this embodiment, the filter window size is 11, and the curves showing the change of the 2D line diameter with Y2 before and after smoothing are as shown in Embodiment 1. Figure 2 As shown, the horizontal axis represents the Y2 coordinate value of the conductor, and the vertical axis represents the two-dimensional diameter of the conductor, in pixels; the two curves are the two-dimensional diameter curves of the conductor before and after smoothing.

[0118] B5: Calculate the approximate vector of each cluster of wires in three-dimensional space and the normal plane of that vector at the origin.

[0119] B5.1: Calculate the coordinates of the three-dimensional approximate centroids of the two endpoints of the conductor, denoted as P and Q.

[0120] In this embodiment, the method for calculating the approximate centroid of the two endpoints of the conductor is as follows: using the method described in B4.2, calculate the centroid of the conductor at points Y2 equal to Y... min and Y max The coordinates of the three-dimensional approximate centroids of the two endpoints are obtained by finding the points in the neighborhood of the two endpoints and calculating the average of the three-dimensional coordinates of the points in the neighborhood of the two endpoints.

[0121] B5.2, Let vectors be... Let (A3, B3, C3) be the approximate vector of the conductor in three-dimensional space. Then the equation of its normal plane at the origin is A3x + B3y + C3z + D3 = 0, where D3 = 0.

[0122] B6: Using the integer values ​​of the Y2 coordinates as indices, calculate the three-dimensional diameter of each cluster of conductors at all Y2 coordinates, and take the average value as the overall three-dimensional diameter of each cluster of conductors.

[0123] B6.1: For any integer Y2, [Y...] min ,Y max Using the method described in B4.2, calculate the points of the conductor in the Y2 neighborhood.

[0124] B6.2: For any integer Y2, ∈ [Y min ,Y max Take any two points in the neighborhood and form a three-dimensional vector, then calculate... The projection length onto the normal plane of the 3D approximate vector of the conductor (note: this normal plane is obtained in B5.2); traversing all 3D vectors The maximum value of the projected length is the three-dimensional diameter of the conductor at this Y2 coordinate.

[0125] B6.3: Take all integers Y2 belonging to [Y min ,Y max The average value of the three-dimensional wire diameter at the specified location is taken as the overall three-dimensional wire diameter of the conductor. The calculation results of the conductor's three-dimensional wire diameter are as shown in Example 1. Figure 3As shown, the horizontal axis represents the Y2 coordinate value of the conductor, and the vertical axis represents the three-dimensional diameter of the conductor, in meters; the curve in the figure is the curve of the three-dimensional diameter of the conductor changing with Y2, and the straight line is the overall diameter result after taking the average value.

[0126] B7: Calculate the scale of each cluster of conductors at various locations and store it as a cubic curve equation.

[0127] B7.1: Using the integer values ​​of the Y2 coordinates as indices, calculate the ratio of the three-dimensional wire diameter of each cluster of conductors to the two-dimensional wire diameter at all Y2 coordinates. This ratio is the scale of each cluster of conductors at Y2.

[0128] B7.2: Using Y2 as the independent variable and the scale S at Y2 as the dependent variable, fit a cubic equation of dependent and independent variables using the least squares method, expressed as S = aY2. 3 +bY2 2 +cY2+d, storing the range of values ​​for coefficients a, b, c, d, and Y2 [Y min ,Y max The fitted equation is described by [the following]; simultaneously, the coefficients k2 and b of the two-dimensional approximate straight line equation obtained from B3.2 are stored to describe the two-dimensional approximate straight line of the traverse. The curve of the traverse scale changing with Y2 is as shown in Example 1. Figure 4 As shown, the horizontal axis represents the Y2 coordinate value of the traverse, and the vertical axis represents the scale of the transformation from two-dimensional to three-dimensional dimensions, in meters per pixel; the two curves are the traverse scale curves before and after fitting, respectively.

[0129] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0130] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0131] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0132] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0133] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0134] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An amplitude quantization method for video analysis of power transmission line galloping, characterized in that: Includes the following processes: Generate the scale equation; The traverse closest to the tracking point is taken as the traverse to which the tracking point belongs. The ordinate of the tracking point is substituted into the scale equation of the traverse to which the tracking point belongs to obtain the scale of the traverse to which the tracking point belongs at the tracking point. Based on the scale, the two-dimensional pixel amplitude is converted into the three-dimensional real amplitude. The generation of the scale equation includes: Acquire 3D point cloud data of the power transmission line scene after processing, where each point includes 3D world coordinates and 2D image coordinates; The point cloud is clustered based on the number of conductor clusters in the current scene, and each point is assigned to the corresponding conductor cluster. Calculate the two-dimensional approximate straight line equation and its normal vector for each cluster of wires in the image; Based on the two-dimensional approximate straight line equation and its normal vector, and using the integer values ​​of the two-dimensional image ordinates of each point as indices, calculate the two-dimensional line diameter of each cluster of wires at all two-dimensional image ordinates. Calculate the approximate vector of each cluster of wires in three-dimensional space and the normal plane of that vector at the origin; Based on the approximate vector of each cluster of conductors in three-dimensional space and the normal plane of that vector at the origin, the integer values ​​of the ordinates of the two-dimensional images of each point are used as indices to calculate the three-dimensional diameter of each cluster of conductors at all ordinates of the two-dimensional images, and the average value is taken as the overall three-dimensional diameter of each cluster of conductors. Calculate the scale of each cluster of conductors at the vertical coordinate of all two-dimensional images, and obtain the scale equation after fitting.

2. The amplitude quantization method for video analysis of transmission line galloping as described in claim 1, characterized in that: The processed 3D point cloud data includes: Point cloud data containing only the guide wires in the scene is obtained by calibrating, calculating, and filtering the original 3D point cloud data and video images.

3. The amplitude quantization method for video analysis of transmission line galloping as described in claim 1, characterized in that: Calculate the two-dimensional approximate straight line equation and its normal vector for each cluster of wires in the image, including: Based on the vertical coordinate of the two-dimensional image, remove the portion of each cluster of wires at the bottom of the image with a predetermined proportion, and calculate the range of values ​​for the vertical coordinate of the remaining portion of each cluster of wires in the two-dimensional image, denoted as the coordinate range. For each cluster of conductors, a straight line is fitted using the two-dimensional coordinates of the point cloud of each cluster of conductors to obtain the fitted two-dimensional approximate straight line equation and its normal vector.

4. The amplitude quantization method for transmission line galloping video analysis as described in claim 3, characterized in that: Based on the two-dimensional approximate straight line equation and its normal vector, and using the integer values ​​of the two-dimensional image ordinates of each point as indices, calculate the two-dimensional diameter of each cluster of conductors at all two-dimensional image ordinates, including: For any traverse, calculate the foot coordinates of the perpendiculars from the two-dimensional coordinates of all points on the traverse to the two-dimensional approximate line of the traverse; For any integer ordinate of a two-dimensional image that falls within the range of coordinate values, points whose ordinates differ from the ordinate of the perpendicular by less than n are taken as points in the neighborhood of the ordinate of the two-dimensional image, where n is half the size of the neighborhood. For any integer ordinate of the two-dimensional image within the range of coordinate values, take any two points in the neighborhood to form a two-dimensional vector, calculate the projection length of the two-dimensional vector onto the normal vector of the two-dimensional approximate straight line of the conductor, and traverse all two-dimensional vectors to find the maximum projection length, which is the two-dimensional diameter of the conductor at the ordinate of this two-dimensional image. The mean filter is used to smooth the two-dimensional line diameter at each point, resulting in the final two-dimensional line diameter at the ordinate of the two-dimensional image.

5. The amplitude quantization method for transmission line galloping video analysis as described in claim 3, characterized in that: Calculate the approximate vector of each cluster of wires in three-dimensional space and the normal plane of that vector at the origin, including: For any cluster of conductors, calculate the coordinates of the three-dimensional approximate centroid of the two endpoints of the conductor; The line connecting the two ends of the conductor and their approximate centroids in three dimensions is taken as the approximate vector of the conductor in three-dimensional space. Based on the approximate vector of the conductor in three-dimensional space, the equation of the normal plane of the current cluster of conductors at the origin is obtained.

6. The amplitude quantization method for transmission line galloping video analysis as described in claim 5, characterized in that: Calculate the coordinates of the three-dimensional approximate centroids at both ends of the conductor, including: Calculate the Y-coordinate of the conductor in the two-dimensional image. min and Y max The coordinates of the approximate three-dimensional centroids of the two endpoints are obtained by taking the points within the neighborhood of the endpoint and calculating the average of the three-dimensional coordinates of the points in the two neighborhoods respectively, where Y... min and Y max These are the minimum and maximum endpoint values ​​of the coordinate range, respectively.

7. The amplitude quantization method for video analysis of transmission line galloping as described in claim 5, characterized in that: Based on the approximate vector of each conductor cluster in three-dimensional space and the normal plane of that vector at the origin, using the integer values ​​of the ordinates of the two-dimensional image at each point as indices, the three-dimensional diameter of each conductor cluster at all ordinates of the two-dimensional image is calculated, and the average value is taken as the overall three-dimensional diameter of each conductor cluster, including: For any integer ordinate of a two-dimensional image that falls within the range of coordinate values, calculate the points of the conductor within the neighborhood of this ordinate of the two-dimensional image. For any integer ordinate of a two-dimensional image that falls within the range of coordinate values, take any two points in the neighborhood to form a three-dimensional vector, and calculate the projection length of the three-dimensional vector onto the normal plane of the three-dimensional approximate vector of the conductor. The maximum value of the projected length obtained by traversing all three-dimensional vectors is the three-dimensional diameter of the conductor at the vertical coordinate of this two-dimensional image. The average value of the three-dimensional line diameter at the vertical coordinate of all two-dimensional images is taken as the overall three-dimensional line diameter of the conductor. All integer values ​​at the vertical coordinate of the two-dimensional images are within the range of coordinate values.

8. The amplitude quantization method for video analysis of transmission line galloping as described in any one of claims 1-7, characterized in that: Calculate the scale of each cluster of conductors at the ordinate of all two-dimensional images, and obtain the scale equation after fitting, including: Using the integer values ​​of the vertical coordinates of the two-dimensional image as indices, calculate the ratio of the three-dimensional wire diameter of each cluster of wires to the two-dimensional wire diameter at all vertical coordinates of the two-dimensional image, and obtain the scale of each cluster of wires at the vertical coordinates of the two-dimensional image. For a certain cluster of conductors, using the ordinate of a two-dimensional image as the independent variable and the scale at the ordinate of the two-dimensional image as the dependent variable, the least squares method is used to fit a cubic equation of the dependent and independent variables to obtain the scale equation.

9. An amplitude quantization system for video analysis of transmission line galloping, characterized in that: include: The scale equation generation module is configured to generate scale equations. The amplitude conversion module is configured to: take the traverse closest to the tracking point as the traverse to which the tracking point belongs, substitute the ordinate of the tracking point into the scale equation of the traverse to which the tracking point belongs, obtain the scale of the traverse to which the tracking point belongs at the tracking point, and convert the two-dimensional pixel amplitude into the three-dimensional real amplitude according to the scale. The generation of the scale equation includes: Acquire 3D point cloud data of the power transmission line scene after processing, where each point includes 3D world coordinates and 2D image coordinates; The point cloud is clustered based on the number of conductor clusters in the current scene, and each point is assigned to the corresponding conductor cluster. Calculate the two-dimensional approximate straight line equation and its normal vector for each cluster of wires in the image; Based on the two-dimensional approximate straight line equation and its normal vector, and using the integer values ​​of the two-dimensional image ordinates of each point as indices, calculate the two-dimensional line diameter of each cluster of wires at all two-dimensional image ordinates. Calculate the approximate vector of each cluster of wires in three-dimensional space and the normal plane of that vector at the origin; Based on the approximate vector of each cluster of conductors in three-dimensional space and the normal plane of that vector at the origin, the integer values ​​of the ordinates of the two-dimensional images of each point are used as indices to calculate the three-dimensional diameter of each cluster of conductors at all ordinates of the two-dimensional images, and the average value is taken as the overall three-dimensional diameter of each cluster of conductors. Calculate the scale of each cluster of conductors at the vertical coordinate of all two-dimensional images, and obtain the scale equation after fitting.

10. The amplitude quantization system for video analysis of transmission line galloping as described in claim 9, characterized in that: Calculate the scale of each cluster of conductors at the ordinate of all two-dimensional images, and obtain the scale equation after fitting, including: Using the integer values ​​of the vertical coordinates of the two-dimensional image as indices, calculate the ratio of the three-dimensional wire diameter of each cluster of wires to the two-dimensional wire diameter at all vertical coordinates of the two-dimensional image, and obtain the scale of each cluster of wires at the vertical coordinates of the two-dimensional image. For a certain cluster of conductors, using the ordinate of a two-dimensional image as the independent variable and the scale at the ordinate of the two-dimensional image as the dependent variable, the least squares method is used to fit a cubic equation of the dependent and independent variables to obtain the scale equation.

Images