Line sag monitoring method based on defogging algorithm

Abstract

The application discloses a line sag monitoring method based on a defogging algorithm and relates to the technical field of power line construction. Fog weather has an influence on the sag measurement of the erected ground wire; the method comprises the following steps: obtaining the three-dimensional reconstruction structure of a scene according to images shot from multiple times, multiple angles and multiple positions of a line, and calculating the three-dimensional coordinates of each point of the actual line and the lowest point of the sag according to the relationship between the world coordinates and the pixel coordinates; obtaining a fog weather sag diagram, and completing the conversion between the pixel coordinates and the actual three-dimensional coordinates; combining temperature and humidity to statistically analyze the sag data, training the statistical sag, and obtaining a defogging algorithm model; when the line is constructed in fog weather, the difference between the sag in the current state and the sag in the reference state is calculated through the defogging algorithm model according to the temperature and the humidity, so that the sag of the line during construction is adjusted, and the position of the lowest point of the sag is determined. The technical scheme eliminates the construction influence of fog weather and improves the installation precision of the line in fog weather.

Description

Technical Field

[0001] This invention relates to the field of power line construction technology, and in particular to a method for monitoring line sag based on a defogging algorithm. Background Technology

[0002] In foggy weather, after the construction unit completes the installation of conductors and ground wires, how can it determine whether the design conditions are met under natural conditions? Measuring sag in foggy weather often results in problems such as the lowest point of sag not being the lowest point under natural weather conditions, and the sag value not being accurate. If the construction unit completes the installation of conductors and ground wires under these conditions, it will inevitably be inconsistent with the design, thus bringing hidden dangers. Summary of the Invention

[0003] The technical problem to be solved and the technical task proposed by this invention is to improve and refine existing technical solutions, and to provide a line sag monitoring method based on a defogging algorithm, so as to reduce the impact of fog on construction. To this end, this invention adopts the following technical solution.

[0004] A method for monitoring line sag based on a defogging algorithm includes the following steps:

[0005] 1) Establish a dehazing algorithm model

[0006] 101) Establish a vision system, which includes a vision sensor, an image acquisition module, and an image processing module. The vision sensor includes a camera. Determine the position of the camera based on the horizontal plane and the height of the line so that the camera can accurately capture the sag position of the line during the shooting process. Take photos at multiple positions and angles at different times and record local weather conditions including temperature and humidity.

[0007] 102) Dehaze the images taken in foggy conditions and improve the atmospheric light value to obtain the dehazed image, which enhances the contrast of the image and makes the edges clearer;

[0008] 103) Perform image preprocessing on the dehazed image, use edge detection to locate the line position, and ensure the integrity of the line edge information;

[0009] 104) Based on images taken multiple times from multiple angles and locations along the route, a three-dimensional reconstruction of the scene is obtained. Based on the relationship between world coordinates and pixel coordinates, the three-dimensional coordinates of each point on the actual route and the lowest point of the sag are calculated. A foggy sag map is obtained, and the conversion between pixel coordinates and actual three-dimensional coordinates is completed.

[0010] 105) After the fog dissipates, repeat steps 101), 103), and 104) to form a sag diagram under natural climate conditions. Combine the relevant sag data with temperature and humidity to perform statistics, and train the statistically collected sag data to obtain the defogging algorithm model.

[0011] 2) When the line is being constructed in foggy weather, the difference between the sag under the current state and the sag under the reference state is calculated based on the temperature and humidity using a defogging algorithm model, thereby adjusting the sag of the line during construction and determining the position of the lowest sag point.

[0012] By taking multiple, multi-angle, and multi-position photos of the line through camera movement, a three-dimensional reconstruction of the scene is obtained, thereby solving the problem of detecting the sag of the line between two towers. Due to scene limitations, it is impossible to use a calibration object, and placing the calibration object near the line by climbing or other means would lose the significance of non-contact measurement.

[0013] This technical solution combines defogging algorithm image processing and pixel 3D technology to generate foggy day measurement graphics under various temperature and humidity conditions. Based on the defogging algorithm model, the difference between the sag under the current state and the sag under the reference state is obtained. The line sag is corrected based on the difference, so that even when construction is carried out in foggy weather, the lowest point of the line sag under natural climate conditions can be consistent with the design, eliminating the impact of foggy day construction, improving the installation accuracy of the line in foggy weather, and providing a reference selection graphic for sag measurement in subsequent similar projects.

[0014] As a preferred technical means: In step 101), when taking photos of the installed wiring in foggy weather, the camera is positioned near the sag. This ensures that a relatively clear image of the lowest point of the sag can be obtained even in foggy weather.

[0015] As a preferred technical approach, in step 102), dehazing is performed based on the dark channel principle. This allows for the acquisition of more edge information and improves the real-time performance of the image.

[0016] As a preferred technical approach: In step 103), preprocessing includes image grayscale conversion and filtering / denoising; after preprocessing, a fusion algorithm of Canny detection operator and eight-neighbor edge tracking is used, along with dual thresholds in the Canny operator, to accurately locate the line position, ensuring the integrity of the line edge information. The fusion algorithm of Canny detection operator and eight-neighbor edge tracking avoids the loss of edge information, thereby ensuring the integrity of the line edge information.

[0017] As a preferred technical approach: In step 105, during the statistical analysis, temperature and humidity are set as baseline values, and the statistically analyzed sag is trained using machine learning based on the TensorFlow framework. This automatic learning through machine learning achieves artificial intelligence.

[0018] Beneficial effects: This technical solution is based on machine vision recognition technology and combines it with the control of sag installation accuracy during line construction in foggy weather, thereby controlling the stress index of the tower, enabling the construction to meet higher design requirements, achieving the goal of lean construction, mastering the high-precision control of construction technology during foggy weather, and ensuring the safe operation of the line project. It has the advantages of high precision, fast speed and wide monitoring coverage. Attached Figure Description

[0019] Figure 1 This is a flowchart of the present invention.

[0020] Figure 2 This is a block diagram of the vision system of the present invention.

[0021] Figure 3 This is a schematic diagram of the on-site structure of the present invention. Detailed Implementation

[0022] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings.

[0023] like Figure 1 As shown, the present invention includes step 1 of establishing a defogging algorithm model and step 2 of adjusting the sag during construction.

[0024] Step 1 in establishing the dehazing algorithm model includes:

[0025] Step 101: Establish a computer vision system (e.g., Figure 2 Based on the horizontal plane and the height of the line, ensure that the camera can accurately capture the sag position of the line during the shooting process. In foggy weather, set up a camera near the sag of the already installed line and take multiple photos from multiple angles at different times and locations, and record local weather conditions such as temperature and humidity;

[0026] Step 102: Dehaze the captured images. Image dehazing methods generally fall into three categories: image enhancement, image fusion, and image restoration. Different needs correspond to different dehazing methods. In this embodiment, to obtain more edge information and ensure real-time image quality, a dehazing method based on the dark channel principle is adopted, and atmospheric light values ​​are improved to obtain the dehazed image, enhancing image contrast and making edges clearer.

[0027] Step 103: Perform image preprocessing on the dehazed image. This includes image grayscale conversion, filtering, and noise reduction. Then, analyze commonly used edge detection operators, such as the Sobel operator, LOG operator, Roberts operator, and Canny operator. First-order differential operators like Roberts and Sobel offer high positioning accuracy and good real-time performance, but they often lose some edge information and sometimes produce false edges. The LOG operator is an improvement on the Laplace operator, addressing the noise issue, but it still smooths out some edge information. After comparing with other detection operators, the Canny edge detection operator, which performs better, was selected. Here, a fusion algorithm of the Canny detection operator and eight-neighbor edge tracking is used, and the double threshold setting in the Canny operator is improved to accurately locate the line position and ensure the integrity of the line edge information.

[0028] Step 104: This technical solution requires detecting the sag of the line between two towers. However, due to scene limitations, it is impossible to use a calibration object. Placing the calibration object near the line by climbing or other means would defeat the purpose of non-contact measurement. Therefore, this embodiment adopts a self-calibration method based on an SFM camera. That is, a single camera is used to take multiple, multi-angle, and multi-position photos of the line. The three-dimensional reconstructed structure of the scene is obtained from the captured images. Based on the relationship between world coordinates and pixel coordinates, the three-dimensional coordinates of each point on the actual line and the lowest point of sag are calculated.

[0029] Step 105: After the fog dissipates, repeat steps 101, 103, and 104 to generate foggy sag maps and sag maps under natural weather conditions. Perform statistical analysis on the sag data (set a baseline value for temperature and humidity), and use artificial intelligence methods based on the TensorFlow framework to train the statistically analyzed sag data. In this way, the difference between the sag and the baseline state can be calculated using conditions such as temperature and humidity.

[0030] Sag adjustment steps during construction:

[0031] When the line is being constructed in foggy weather, the difference between the sag under the current condition and the sag under the reference condition is calculated based on the temperature and humidity using a defogging algorithm model. This allows for the adjustment of the line's sag during construction and the determination of the lowest sag point.

[0032] This technical solution is based on machine vision recognition technology, combined with the control of sag installation accuracy during line construction in foggy weather, thereby controlling the stress index of the tower, enabling the construction to meet higher design requirements, achieving the goal of lean construction, mastering the high-precision control of construction technology in foggy weather, and ensuring the safe operation of the line project.

[0033] The line sag monitoring method based on the defogging algorithm shown above is a specific embodiment of the present invention, which has demonstrated the substantial features and progress of the present invention. According to actual use needs, equivalent modifications in shape, structure, etc. can be made to it under the guidance of the present invention, all of which are within the protection scope of this solution.

Claims

1. A method for monitoring line sag based on a defogging algorithm, characterized in that... Includes the following steps: 1) Establish a dehazing algorithm model 101) Establish a vision system, which includes a vision sensor, an image acquisition module, and an image processing module. The vision sensor includes a camera. Determine the position of the camera based on the horizontal plane and the height of the line so that the camera can accurately capture the sag position of the line during the shooting process. Take photos at multiple positions and angles at different times and record local weather conditions including temperature and humidity. 102) Dehaze the images taken in foggy conditions and improve the atmospheric light value to obtain the dehazed image, which enhances the contrast of the image and makes the edges clearer; 103) Perform image preprocessing on the dehazed image, use edge detection to locate the line position, and ensure the integrity of the line edge information; 104) Based on images taken multiple times from multiple angles and locations along the route, a three-dimensional reconstruction of the scene is obtained. Based on the relationship between world coordinates and pixel coordinates, the three-dimensional coordinates of each point on the actual route and the lowest point of the sag are calculated. A foggy sag map is obtained, and the conversion between pixel coordinates and actual three-dimensional coordinates is completed. 105) After the fog dissipates, repeat steps 101), 103), and 104 to form a sag map under natural climate conditions. Combine the relevant sag data with temperature and humidity to perform statistics, and train the statistically collected sag data to obtain the defogging algorithm model. 2) When the line is being constructed in foggy weather, the difference between the sag under the current state and the sag under the reference state is calculated based on the temperature and humidity using a defogging algorithm model. This allows for the adjustment of the line's sag during construction and the determination of the lowest sag point.

2. The method for monitoring line sag based on a defogging algorithm according to claim 1, characterized in that: In step 101), when taking photos of the installed wires on foggy days, the camera is positioned near the sag.

3. The method for monitoring line sag based on a defogging algorithm according to claim 2, characterized in that: In step 102), dehazing is performed using the dark channel principle.

4. The method for monitoring line sag based on a defogging algorithm according to claim 3, characterized in that: In step 103), the preprocessing includes image grayscale conversion and filtering for noise reduction. After preprocessing, a fusion algorithm of Canny detection operator and eight-neighbor edge tracking is used, along with the double threshold in the Canny operator, to accurately locate the line position and ensure the integrity of the line edge information.

5. The method for monitoring line sag based on a defogging algorithm according to claim 4, characterized in that: In step 105, during the statistical analysis, temperature and humidity are set as baseline values, and the statistically analyzed sag is trained using machine learning based on the TensorFlow framework.

Images