A method for detecting defects in AC / DC hybrid power lines based on image data

Abstract

The invention discloses an AC / DC hybrid power line defect detection method based on image data, and relates to the technical field of power line defect detection. The technical key points of the method are: taking the same time interval as a sampling period, sampling the AC / DC hybrid power line at the same sampling coordinates and sampling angles, presetting a depth threshold, dividing a depth map into a background depth map and a foreground depth map, calculating a foreground depth deviation value between a foreground depth map to be tested and a sample foreground depth map, and calculating a background depth deviation value between the background depth map to be tested and the sample background depth map. The invention can accurately distinguish between the background and the AC / DC hybrid power line part in the depth image, judge the defects of the AC / DC hybrid power line itself and the surrounding defects, has low requirements on lighting conditions, improves the efficiency of defect detection, can timely discover changes in the surrounding environment, and helps prevent potential safety risks and accidents.

Description

Technical Field

[0001] The present invention relates to the technical field of power line defect detection, and in particular to an AC / DC hybrid power line defect detection method based on image data. Background Art

[0002] With the continuous development of power systems, AC / DC hybrid power lines are increasingly used in power systems. However, due to the complex operating environment of the lines, defect detection has always been a difficult point in power line maintenance. Existing defect detection methods are often based on specific physical or chemical quantities for detection, and cannot fully cover all defects of AC / DC hybrid power lines. Therefore, it is of great significance to develop a method that can comprehensively and real-time detect defects in AC / DC hybrid power lines.

[0003] In a Chinese invention application with application publication number CN109523529B, a method for identifying defects in a power transmission line based on a SURF algorithm is disclosed, comprising the following steps: edge detection is performed on a power transmission line image, and then the edge detection image is grayed, and the power transmission line in the power transmission line image is obtained by using a Hough transform, and the Rect function in OpenCV is called to obtain the coordinates of the four vertices of the minimum rectangular area where the power transmission line is located in the power transmission line image, thereby obtaining the minimum rectangular area containing the power transmission line in the original image, namely, the region of interest ROI, and completing the process of extracting the region of interest of the power transmission line image; according to the characteristics of the power transmission line image, the SURF algorithm is used to extract the features of the power transmission line image, and the power transmission line image is represented as a numerical vector in combination with the BOW algorithm; and the defect type of the power transmission line in the power transmission line image is determined by using the SVM algorithm.

[0004] Combined with the existing technology, there are the following deficiencies:

[0005] 1. In the images of power lines, due to changes in shooting angle, distance, lighting and shadow conditions, the geometric features of the transmission equipment may change significantly in the image, thus causing inaccurate and unstable defect detection results;

[0006] 2. No detection of defects in the surrounding environment. Existing technologies often only focus on the defect detection of the power transmission equipment itself, while ignoring the detection of its surrounding environment. However, the environmental conditions around the power transmission equipment are equally important, because any environmental defects may affect the normal operation of the equipment;

[0007] 3. Lack of comparative detection based on historical data. The current defect detection methods for power transmission equipment often do not make full use of historical data for comparison and analysis. Historical data contains a lot of information, and some defects and problems can be discovered through comparison and analysis. Summary of the invention

[0008] 1. Technical issues to be resolved

[0009] In view of the shortcomings of the prior art, the present invention provides an AC / DC hybrid power line defect detection method based on image data, which uses the same time interval as the sampling period, samples the AC / DC hybrid power line with the same sampling coordinates and sampling angles, pre-sets the depth threshold, divides the depth map into a background depth map and a foreground depth map, and calculates the foreground depth deviation value of the foreground depth map to be tested and the sample foreground depth map. , calculate the background depth deviation value Dev between the background depth map to be tested and the sample background depth map, judge the defects of the AC / DC hybrid power line itself and the surrounding defects, and solve the problems mentioned in the background technology.

[0010] (II) Technical solution

[0011] To achieve the above objectives, the present invention is implemented through the following technical solutions: A method for detecting defects in AC / DC hybrid power lines based on image data, comprising the following steps:

[0012] Collect high-definition images and depth data of AC / DC hybrid power lines to generate depth images with depth information;

[0013] The AC / DC hybrid power lines are sampled at the same time interval as the sampling period, the same sampling coordinates and sampling angles, and the collected depth images are sorted into sample data sets, where each sample includes the depth image itself, sampling coordinates, sampling angles and acquisition time;

[0014] Preset the depth threshold, judge the pixel points with depth value greater than the depth threshold as background, and judge the pixel points with depth value less than or equal to the threshold as AC / DC hybrid power line, i.e. foreground, split into corresponding background depth map and foreground depth map, calculate the foreground depth deviation value of the foreground depth map to be tested and the sample foreground depth map , pre-set the foreground depth deviation threshold and compare it with the foreground depth deviation value Perform a comparison and make a corresponding judgment based on the comparison results;

[0015] Calculate the background depth deviation value Dev between the background depth map to be tested and the sample background depth map, pre-set the background depth deviation threshold, and compare it with the background depth deviation value Dev, and make corresponding judgments according to the comparison results;

[0016] Obtain the judgment results. When there are defects in the judgment results, issue a warning message and display the specific information, including detailed coordinates, high-definition images of the AC / DC hybrid power lines, and depth images.

[0017] Furthermore, drones or ground-based cameras are used to capture images of AC / DC hybrid power lines to obtain high-definition images of the power lines, including insulators, conductors, and poles. LiDAR is used to collect depth data of the AC / DC hybrid power lines to obtain three-dimensional coordinate data of the power lines, including the shape, height, and location information of the lines. Based on the depth data of the AC / DC hybrid power lines, a depth image with depth information is generated.

[0018] Furthermore, the depth image of the AC / DC hybrid power line is collected, and the sampling coordinates, sampling angle and sampling time of each sample are recorded;

[0019] The AC / DC hybrid power lines are sampled at the same time interval as the sampling period, at the same sampling coordinates and sampling angles. The time interval can be one week or one month.

[0020] The collected depth images are organized into sample data sets, each sample including the depth image itself, sampling coordinates, sampling angle, and acquisition time;

[0021] Encrypt the sample data and transmit the encrypted data to the storage end.

[0022] Furthermore, a depth threshold is pre-set, and pixels with depth values ​​greater than the depth threshold are determined as background, and pixels with depth values ​​less than or equal to the threshold are determined as AC / DC hybrid power lines, i.e., foreground, and are segmented into corresponding background depth maps and foreground depth maps;

[0023] Get the depth information of all pixels in the foreground depth map, and sum the depth values ​​of all pixels in the foreground depth map to get the sum of the depth values ​​of the foreground depth map. The calculation formula is as follows:

[0024] ;

[0025] Wherein, h(x,y) represents the depth value of the pixel with abscissa x and ordinate y in the depth image, m represents the total number of abscissas, and n represents the total number of ordinates.

[0026] Further, a foreground depth map to be tested and a sample foreground depth map of the AC / DC hybrid power line are obtained, and depth data of all pixels are extracted;

[0027] Calculate the foreground depth deviation value between the foreground depth map to be tested and the sample foreground depth map , and the sum of the foreground deviation values ​​Dp, the calculation formula is as follows:

[0028] ;

[0029] in, It is represented by the sum of the depth values ​​of the sample foreground depth map, i is represented by the label of the sample foreground depth map, It is represented as the sum of the depth values ​​of the foreground depth map to be tested, and N is represented as the number of sample foreground depth maps.

[0030] Furthermore, a foreground depth deviation threshold is preset. When it is greater than or equal to the foreground depth deviation threshold, and the sum of the foreground deviation values ​​Dp is a positive number, it indicates that there are foreign objects hanging around the AC / DC hybrid power line, and the judgment result is output;

[0031] Current foreground depth deviation value When it is greater than or equal to the foreground depth deviation threshold and the sum of the foreground deviation values ​​Dp is a negative number, it indicates that the AC / DC hybrid power line is damaged or detached, and the judgment result is output.

[0032] Further, obtaining the background depth map to be tested and the sample background depth map, and extracting the depth data of all pixels;

[0033] The sum of the depth values ​​of the background depth map to be tested and the sum of the depth values ​​of the sample background depth map are calculated, and the background depth deviation value Dev between the background depth map to be tested and the sample background depth map, as well as the sum of the background deviation values ​​Sd are calculated using the following formula:

[0034] ;

[0035] in, It is represented as the sum of the depth values ​​of the background depth map to be tested. represents the sum of the depth values ​​of the sample background depth map, j represents the label of the sample background depth map, and M represents the number of sample background depth maps.

[0036] Furthermore, a background depth deviation threshold is set in advance. When the background depth deviation value Dev is greater than or equal to the background depth deviation threshold, and the sum of the background deviation values ​​Sd is a positive number, it indicates that the surrounding environment of the AC / DC hybrid power line has changed and the relative position to the AC / DC hybrid power line has become closer, and the judgment result is output.

[0037] Furthermore, the judgment results are obtained and classified according to the judgment results; when there are defects in the judgment results, a warning message is issued to prompt that repairs are needed; and specific information is displayed, including detailed coordinates, high-definition images of the AC / DC hybrid power lines, and depth images.

[0038] (III) Beneficial effects

[0039] The present invention provides an AC / DC hybrid power line defect detection method based on image data, which has the following beneficial effects:

[0040] (1) Since the sampling coordinates and angles remain consistent, the depth images in the sample data set can be highly consistent, which helps to eliminate data differences caused by different sampling conditions, making subsequent analysis results more accurate and reliable. By sampling at the same time interval, a large number of depth image samples can be obtained, thereby enriching the sample data set. The enriched data set helps to improve the efficiency and accuracy of power line inspection.

[0041] (2) By setting the depth threshold, the background and AC / DC hybrid power line parts in the depth image can be accurately distinguished, and the lighting conditions are low, so that a more accurate foreground depth map containing the power line part can be extracted, which helps to improve the precision and accuracy of subsequent defect detection. By comparing with the pre-set foreground depth deviation threshold, the defect situation can be directly determined according to the comparison result, and the defects of the AC / DC hybrid power line itself and the surrounding defects can be judged without complex calculations and model training, which helps to improve the efficiency of defect detection and reduce the consumption of computing resources and time.

[0042] (3) By analyzing the background depth image, changes in the surrounding environment can be discovered in a timely manner, which helps prevent potential safety risks and accidents. For example, fallen trees may cause damage to power lines. By discovering this in advance and taking corresponding measures, the possibility of potential risks and accidents can be reduced.

[0043] (4) Through defect detection and early warning, line failures and equipment damage caused by further deterioration of defects can be avoided, thereby reducing repair costs and maintenance difficulties. At the same time, through the display of high-definition images and depth images, the specific situation of the defects can be more intuitively understood, providing a more accurate reference for repair and reducing repair costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 The present invention is a flow chart of the AC / DC hybrid power line defect detection method. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] See also Figure 1 The present invention provides an AC / DC hybrid power line defect detection method based on image data, comprising the following steps:

[0047] Step 1: Collect high-definition images and depth data of AC / DC hybrid power lines to generate depth images with depth information;

[0048] The step 1 includes the following contents:

[0049] Step 101: Use a drone or ground camera to collect images of the AC / DC hybrid power line to obtain a high-definition image of the power line, including insulators, conductors, and towers;

[0050] Step 102: Use a laser radar to collect depth data of the AC / DC hybrid power line to obtain three-dimensional coordinate data of the power line, including information such as the shape, height, and position of the line;

[0051] Step 103: Generate a depth image with depth information according to the depth data of the AC / DC hybrid power line.

[0052] It should be noted that converting depth data into a depth image, that is, converting the point cloud data on the surface of the object into pixel values ​​according to the distance information in the depth data, to generate a depth image similar to a grayscale image. In this process, different parameters can be set as needed, such as pixel resolution, threshold, etc., to control the quality and accuracy of the depth image.

[0053] When using, combine the contents of step 101 to step 103:

[0054] By processing and analyzing the collected images and depth data, features related to defects can be extracted, and defect detection and classification can be performed using machine learning and computer vision techniques. These detection results can provide important reference for the maintenance and inspection of power lines.

[0055] Step 2: Sample the AC / DC hybrid power line with the same time interval as the sampling period, the same sampling coordinates and sampling angles, and organize the collected depth images into a sample data set, where each sample includes the depth image itself, sampling coordinates, sampling angles and collection time;

[0056] The step 2 includes the following contents:

[0057] Step 201: Acquire a depth image of the AC / DC hybrid power line and record the sampling coordinates, sampling angle and sampling time of each sample;

[0058] Step 202: sampling the AC / DC hybrid power line with the same time interval as the sampling period, the same sampling coordinates and sampling angles. The time interval may be one week or one month.

[0059] It should be noted that factors such as different shooting positions and shooting angles may affect the depth image. Therefore, when sampling AC / DC hybrid power lines with the same sampling coordinates and sampling angles, it is necessary to pay attention to these factors and take corresponding measures to reduce their impact and ensure the accuracy and consistency of the sampling.

[0060] Step 203: Arrange the acquired depth images into a sample data set, where each sample includes metadata such as the depth image itself, sampling coordinates, sampling angle, and acquisition time;

[0061] It should be noted that when establishing a sample data set, it is necessary to collect depth images of the AC / DC hybrid power lines when there are no defects, to ensure that the collected data is not affected by any defects, and to provide a reliable basis for subsequent data analysis and processing.

[0062] Step 204: Encrypt the sample data and transmit the encrypted data to the storage end.

[0063] When used, combine the contents of step 201 to step 204:

[0064] Since the sampling coordinates and angles remain consistent, the depth images in the sample data set can be ensured to be highly consistent, which helps to eliminate data differences caused by different sampling conditions, making subsequent analysis results more accurate and reliable. By sampling at the same time interval, a large number of depth image samples can be obtained, thereby enriching the sample data set. The enriched data set helps to improve the efficiency and accuracy of power line inspections.

[0065] Step 3: Preset the depth threshold, judge the pixels with depth values ​​greater than the depth threshold as the background, and judge the pixels with depth values ​​less than or equal to the threshold as the AC / DC hybrid power line, i.e. the foreground, and divide them into the corresponding background depth map and foreground depth map, and calculate the foreground depth deviation value of the foreground depth map to be tested and the sample foreground depth map , pre-set the foreground depth deviation threshold and compare it with the foreground depth deviation value Perform a comparison and make a corresponding judgment based on the comparison results;

[0066] The step three includes the following contents:

[0067] Step 301: pre-set a depth threshold, determine the pixel points with a depth value greater than the depth threshold as the background, and determine the pixel points with a depth value less than or equal to the threshold as the AC / DC hybrid power line, i.e., the foreground, and segment them into corresponding background depth map and foreground depth map;

[0068] It should be noted that for the depth map, the depth information can be used to segment the foreground and background in the image. Generally, the larger the depth value, the farther the object is from the camera. Therefore, the pixels with larger depth values ​​can be judged as the background, and the pixels with smaller depth values ​​can be judged as the photographed object.

[0069] Step 302: Obtain depth information of all pixels in the foreground depth map, and perform a sum operation on the depth values ​​of all pixels in the foreground depth map to obtain the sum Sum of the depth values ​​of the foreground depth map. The calculation formula is as follows:

[0070] ;

[0071] Wherein, h(x,y) represents the depth value of the pixel with horizontal coordinate x and vertical coordinate y in the depth image, m represents the total number of horizontal coordinates, and n represents the total number of vertical coordinates;

[0072] Step 303: obtaining a foreground depth map to be tested and a sample foreground depth map of the AC / DC hybrid power line, and extracting depth data of all pixels;

[0073] Step 304: Calculate the foreground depth deviation between the foreground depth map to be tested and the sample foreground depth map , and the sum of the foreground deviation values ​​Dp, the calculation formula is as follows:

[0074] ;

[0075] in, It is represented by the sum of the depth values ​​of the sample foreground depth map, i is represented by the label of the sample foreground depth map, It is represented as the sum of the depth values ​​of the foreground depth map to be tested, and N is represented as the number of sample foreground depth maps;

[0076] Step 305: Preset the foreground depth deviation threshold. When it is greater than or equal to the foreground depth deviation threshold, and the sum of the foreground deviation values ​​Dp is a positive number, it indicates that there are foreign objects hanging around the AC / DC hybrid power line, such as kites, ribbons, balloons, etc., and the judgment result is output;

[0077] Current foreground depth deviation value When it is greater than or equal to the foreground depth deviation threshold and the sum of the foreground deviation values ​​Dp is a negative number, it means that the AC / DC hybrid power line is damaged or detached, such as insulator cracking, detaching, conductor breakage, wear, etc., and the judgment result is output.

[0078] It should be noted that when there are hanging foreign objects around the AC / DC hybrid power lines, the depth data of the foreign objects will be increased, which will cause the depth value of the entire foreground depth map to increase. Conversely, when the AC / DC hybrid power lines are damaged or detached, the depth value of the entire foreground depth map will decrease. Therefore, the foreground depth deviation value can be used to calculate the depth of the foreground depth map. And the sum of the foreground deviation value Dp is used for judgment.

[0079] When used, combine the contents of step 301 to step 305:

[0080] By setting the depth threshold, the background and AC / DC hybrid power line parts in the depth image can be accurately distinguished, so as to extract a more accurate foreground depth map containing the power line part, which helps to improve the precision and accuracy of subsequent defect detection. By comparing with the pre-set foreground depth deviation threshold, the defect situation can be directly determined according to the comparison result, and the defects of the AC / DC hybrid power line itself and the surrounding defects can be judged without complex calculations and model training, which helps to improve the efficiency of defect detection and reduce the consumption of computing resources and time.

[0081] Step 4: Calculate the background depth deviation value Dev between the background depth map to be tested and the sample background depth map, pre-set the background depth deviation threshold, and compare it with the background depth deviation value Dev, and make a corresponding judgment based on the comparison result;

[0082] The step 4 includes the following contents:

[0083] Step 401: obtaining a background depth map to be tested and a sample background depth map, and extracting depth data of all pixels;

[0084] Step 402: Calculate the sum of the depth values ​​of the background depth map to be tested and the sum of the depth values ​​of the sample background depth map, and use this to calculate the background depth deviation value Dev between the background depth map to be tested and the sample background depth map, as well as the sum of the background deviation values ​​Sd, and the calculation formula is as follows:

[0085] ;

[0086] in, It is represented as the sum of the depth values ​​of the background depth map to be tested. represents the sum of the depth values ​​of the sample background depth map, j represents the label of the sample background depth map, and M represents the number of sample background depth maps;

[0087] Step 403: pre-set a background depth deviation threshold. When the background depth deviation value Dev is greater than or equal to the background depth deviation threshold, and the sum of the background deviation values ​​Sd is a positive number, it indicates that the surrounding environment of the AC / DC hybrid power line has changed and the relative position to the AC / DC hybrid power line has become closer, such as illegal buildings, fallen trees, etc., and the judgment result is output.

[0088] It should be noted that when the environment around the AC / DC hybrid power line changes and some objects become closer to the AC / DC hybrid power line, the sum of the depth values ​​of the background depth map will also increase, and vice versa. When these objects exist in the surrounding environment of the AC / DC hybrid power line, they may pose a threat to the power line or affect its normal operation. Therefore, it is necessary to determine whether these threats exist by judging whether the depth value of the background depth map increases, so as to avoid threats to the power line or affecting its normal operation.

[0089] When used, combine the contents of step 401 to step 403:

[0090] By analyzing the background depth image, changes in the surrounding environment can be discovered in a timely manner, which helps prevent potential safety risks and accidents. For example, fallen trees may cause damage to power lines. By discovering them in advance and taking corresponding measures, the possibility of potential risks and accidents can be reduced.

[0091] Step 5: Obtain the judgment result. When there is a defect in the judgment result, issue a warning message and display the specific information, including detailed coordinates, high-definition images of the AC / DC hybrid power lines, and depth images.

[0092] The step five includes the following contents:

[0093] Step 501: Obtaining the determination result and classifying according to the determination result;

[0094] Step 502: When the result of the determination is defective, a warning message is issued to indicate that repair is required;

[0095] Step 503: Display specific information, including detailed coordinates, high-definition images of the AC / DC hybrid power lines, and depth images, etc. Relevant personnel can use this information to understand the specific situation of the defect, including the location, type, degree, etc. of the defect, and then repair it.

[0096] When used, combine the contents of step 501 to step 503:

[0097] Through defect detection and early warning, we can avoid line failure and equipment damage caused by further deterioration of defects, thereby reducing repair costs and maintenance difficulties. At the same time, through the display of high-definition images and depth images, we can more intuitively understand the specific situation of the defects, provide a more accurate reference for repair, and reduce repair costs.

[0098] In the application, the several formulas involved are all calculated by removing the dimensions and taking their numerical values. The formula is a formula obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The formula is set by technical personnel in this field according to actual conditions.

[0099] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. A person of ordinary skill in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0100] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0101] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be covered by the protection scope of the present application.

Claims

1. A method for detecting defects in AC / DC hybrid power lines based on image data, characterized in that: The following steps are involved: Collect high-definition images and depth data of AC / DC hybrid power lines to generate depth images with depth information; The AC / DC hybrid power lines are sampled at the same time interval as the sampling period, the same sampling coordinates and sampling angles, and the collected depth images are sorted into sample data sets, where each sample includes the depth image itself, sampling coordinates, sampling angles and acquisition time; Preset the depth threshold, judge the pixel points with depth value greater than the depth threshold as background, and judge the pixel points with depth value less than or equal to the threshold as AC / DC hybrid power line, i.e. foreground, split into corresponding background depth map and foreground depth map, calculate the foreground depth deviation value of the foreground depth map to be tested and the sample foreground depth map , pre-set the foreground depth deviation threshold and compare it with the foreground depth deviation value Perform a comparison and make a corresponding judgment based on the comparison results; Calculate the background depth deviation value Dev between the background depth map to be tested and the sample background depth map, pre-set the background depth deviation threshold, and compare it with the background depth deviation value Dev, and make corresponding judgments according to the comparison results; Obtain the judgment results. When there are defects in the judgment results, issue a warning message and display the specific information, including detailed coordinates, high-definition images of the AC / DC hybrid power lines, and depth images.

2. The AC / DC hybrid power line defect detection method based on image data according to claim 1 is characterized in that: Use drones or ground-based cameras to collect images of AC / DC hybrid power lines to obtain high-definition images of power lines, including insulators, conductors, and towers; use lidar to collect depth data of AC / DC hybrid power lines to obtain three-dimensional coordinate data of power lines, including line shape, height, and location information; A depth image with depth information is generated based on the depth data of the AC / DC hybrid power line.

3. The AC / DC hybrid power line defect detection method based on image data according to claim 1 is characterized in that: Perform depth image acquisition on the AC / DC hybrid power line and record the sampling coordinates, sampling angle and sampling time of each sample; use the same time interval as the sampling period, and sample the AC / DC hybrid power line with the same sampling coordinates and sampling angle; The collected depth images are organized into sample data sets, each sample including the depth image itself, sampling coordinates, sampling angle, and acquisition time; Encrypt the sample data and transmit the encrypted data to the storage end.

4. The AC / DC hybrid power line defect detection method based on image data according to claim 1, characterized in that: Pixels with depth values ​​greater than the depth threshold are judged as the background, and pixels with depth values ​​less than or equal to the threshold are judged as AC / DC hybrid power lines, i.e., the foreground, and are divided into corresponding background depth maps and foreground depth maps; the depth information of all pixels in the foreground depth map is obtained, and the depth values ​​of all pixels in the foreground depth map are summed to obtain the sum of the depth values ​​of the foreground depth map Sum, and the calculation formula is as follows: ； Wherein, h(x,y) represents the depth value of the pixel with abscissa x and ordinate y in the depth image, m represents the total number of abscissas, and n represents the total number of ordinates.

5. The AC / DC hybrid power line defect detection method based on image data according to claim 4 is characterized in that: Obtain a foreground depth map to be tested and a sample foreground depth map of an AC / DC hybrid power line, and extract depth data of all pixels; Calculate the foreground depth deviation value between the foreground depth map to be tested and the sample foreground depth map , and the sum of the foreground deviation values ​​Dp, the calculation formula is as follows: ； in, It is represented as the sum of the depth values ​​of the sample foreground depth map, i is represented as the label of the sample foreground depth map, It is represented as the sum of the depth values ​​of the foreground depth map to be tested, and N is represented as the number of sample foreground depth maps.

6. The method for detecting defects in AC / DC hybrid power lines based on image data according to claim 5, characterized in that: Current foreground depth deviation value When it is greater than or equal to the foreground depth deviation threshold, and the sum of the foreground deviation values ​​Dp is a positive number, it indicates that there are foreign objects hanging around the AC / DC hybrid power line, and the judgment result is output; Current foreground depth deviation value When it is greater than or equal to the foreground depth deviation threshold and the sum of the foreground deviation values ​​Dp is a negative number, it indicates that the AC / DC hybrid power line is damaged or detached, and the judgment result is output.

7. The method for detecting defects in AC / DC hybrid power lines based on image data according to claim 1, characterized in that: Obtain the background depth map to be tested and the sample background depth map, and extract the depth data of all pixels; Calculate the sum of the depth values ​​of the background depth map to be measured The sum of the depth values ​​of the sample background depth map , and use this to calculate the background depth deviation value Dev between the background depth map to be tested and the sample background depth map, as well as the sum of the background deviation values ​​Sd. The calculation formula is as follows: ； Among them, j represents the label of the sample background depth map, and M represents the number of sample background depth maps.

8. The method for detecting defects in AC / DC hybrid power lines based on image data according to claim 7, characterized in that: When the background depth deviation value Dev is greater than or equal to the background depth deviation threshold, and the sum of the background deviation values ​​Sd is a positive number, it indicates that the surrounding environment of the AC / DC hybrid power line has changed and the relative position to the AC / DC hybrid power line has become closer, and the judgment result is output.

9. The method for detecting defects in AC / DC hybrid power lines based on image data according to claim 8, characterized in that: Obtain the judgment results and classify them according to the judgment results; when there are defects in the judgment results, issue a warning message to prompt the need for repair, and display specific information, including detailed coordinates, high-definition images of the AC / DC hybrid power lines, and depth images.

Images