A method for monitoring and early warning of transmission tower deformation based on image feature analysis

Abstract

The present invention discloses a method for monitoring and warning the deformation of a transmission tower based on image feature analysis, which relates to the field of image processing technology. The method comprises: monitoring the transmission tower to obtain a monitoring image of the transmission tower; performing time series analysis to obtain time series image features; performing fusion to generate multiple fused tower monitoring images; performing feature extraction to obtain multiple structural reference points; generating multiple three-dimensional structural models of the transmission tower, performing deformation analysis to obtain deformation data and temperature change data; when the temperature change data is greater than a preset temperature threshold, positioning is performed based on the corresponding relationship to obtain abnormal structural points; performing deformation trend analysis on the abnormal structural points in combination with the deformation data to obtain the deformation amount, and generating an early warning signal when the deformation amount is greater than the preset deformation early warning threshold. The method solves the technical problem of low accuracy in monitoring the deformation of transmission towers in complex environments, and achieves the technical effect of improving the accuracy of monitoring the deformation of transmission towers in complex environments.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a method for monitoring and early warning of transmission tower deformation based on image feature analysis. Background Art

[0002] In today's rapidly developing power system, transmission towers, as critical infrastructure for power transmission, are crucial for ensuring the reliability and continuity of power supply. However, with the expansion of the power grid and the impact of natural environmental factors, transmission towers face numerous challenges, such as structural deformation caused by extreme weather conditions, material fatigue from long-term load accumulation, and overheating of electrical connection components. If these issues are not promptly identified and effective measures are not taken, they may cause serious power accidents and affect the safe and stable operation of the power grid. Traditional monitoring methods often focus on monitoring a single parameter, but in complex and changing environments, it is difficult to fully reflect the actual operating status of the towers, resulting in low monitoring accuracy. Summary of the Invention

[0003] The present application provides a transmission tower deformation monitoring and early warning method based on image feature analysis, which solves the technical problem of low accuracy in transmission tower deformation monitoring in complex environments.

[0004] In view of the above problems, the present application provides a transmission tower deformation monitoring and early warning method based on image feature analysis.

[0005] This application provides a method for monitoring and early warning of transmission tower deformation based on image feature analysis, the method comprising:

[0006] Deploy multiple sensor clusters in the target area where the transmission tower is located. Based on the sensor clusters, the transmission tower is monitored to obtain multiple groups of transmission tower monitoring images at different time points, including standard tower monitoring images and infrared tower monitoring images. Time series analysis is performed on the multiple groups of transmission tower monitoring images at different time points to obtain time series image features. Based on the time series image features, the standard tower monitoring images and the infrared tower monitoring images are fused to generate multiple fused tower monitoring images, each of which has a temperature mark. Feature extraction is performed on each of the multiple fused tower monitoring images. , obtain multiple structural reference points of multiple groups of transmission towers; based on the structural reference points, use three-dimensional reconstruction technology to generate multiple three-dimensional structural models of the transmission towers, and perform deformation analysis on the multiple three-dimensional structural models of the transmission towers to obtain deformation data and temperature change data of the transmission towers, wherein the deformation data and the temperature change data have a one-to-one correspondence; when the temperature change data is greater than a preset temperature threshold, locate the abnormal structure point based on the correspondence and obtain the abnormal structure point; combine the deformation data to perform deformation trend analysis on the abnormal structure point to obtain the deformation amount, and generate a warning signal when the deformation amount is greater than a preset deformation warning threshold.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] First, multiple sensor clusters are deployed in the target area where transmission towers are located. Based on these sensor clusters, the transmission towers are monitored, acquiring multiple sets of transmission tower monitoring images at different time points. These images include standard tower monitoring images and infrared tower monitoring images. Next, time series analysis is performed on these multiple sets of transmission tower monitoring images at different time points to obtain temporal image features. Furthermore, based on these temporal image features, the standard tower monitoring images and infrared tower monitoring images are fused to generate multiple fused tower monitoring images. These fused tower monitoring images are then temperature-labeled. Feature extraction is then performed on each of these fused tower monitoring images to obtain multiple sets of structural reference points for the transmission towers. Based on these structural reference points, 3D reconstruction technology is used to generate multiple 3D structural models of the transmission towers. Deformation analysis is then performed on these 3D structural models to obtain deformation data and temperature change data for the transmission towers, with the deformation data and temperature change data having a one-to-one correspondence. When the temperature change data exceeds a preset temperature threshold, the system locates the abnormal structure point based on the corresponding relationship. Finally, the deformation trend of the abnormal structure point is analyzed based on the deformation data to obtain the deformation amount. When the deformation amount exceeds the preset deformation warning threshold, an early warning signal is generated. This solves the technical problem of low accuracy in deformation monitoring of transmission towers in complex environments, achieving the technical effect of improving the accuracy of deformation monitoring of transmission towers in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 A schematic flow chart of a method for monitoring and early warning of transmission tower deformation based on image feature analysis provided in an embodiment of the present application;

[0011] Figure 2 A schematic diagram of the process of layered fusion in a method for monitoring and early warning of transmission tower deformation based on image feature analysis provided in an embodiment of the present application. DETAILED DESCRIPTION

[0012] The present application solves the technical problem of low accuracy in deformation monitoring of transmission towers in complex environments by providing a transmission tower deformation monitoring and early warning method based on image feature analysis.

[0013] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only some of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0014] It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices.

[0015] Examples, such as Figure 1 As shown, the embodiment of the present application provides a method for monitoring and early warning of transmission tower deformation based on image feature analysis, wherein the method includes:

[0016] Multiple sensor clusters are deployed in the target area where the transmission towers are located. Based on the sensor clusters, the transmission towers are monitored to obtain multiple sets of transmission tower monitoring images at different time points. The transmission tower monitoring images include standard tower monitoring images and infrared tower monitoring images.

[0017] In the target area where the transmission tower is located, multiple sensor clusters are deployed. That is, the deployment location of the sensor cluster is reasonably selected according to the specific location, height, surrounding environment, etc. of the transmission tower. Usually, the sensors should be installed in a position that can fully cover all key parts of the tower and is not easily obstructed. The sensor cluster includes various types of sensors, such as visual sensors (used to obtain standard tower monitoring images) and infrared sensors (used to obtain infrared tower monitoring images). The sensor cluster is responsible for continuously monitoring the status of the transmission tower to obtain multiple sets of transmission tower monitoring images at different time points. The transmission tower monitoring images include standard tower monitoring images and infrared tower monitoring images. Among them, the standard tower monitoring image refers to the appearance image of the transmission tower, which contains intuitive information such as the structural details and surface condition of the tower; the infrared tower monitoring image reflects the temperature changes of the tower caused by current, environment or internal faults, as well as possible tower defects.

[0018] Furthermore, multiple sensor clusters are deployed in the target area where the transmission tower is located. Based on the sensor clusters, the transmission tower is monitored to obtain multiple sets of transmission tower monitoring images at different time points. The method includes:

[0019] Based on the target area where the transmission tower is located, multiple deployment points are determined; at the multiple deployment points, a sensor cluster consisting of visual sensors and infrared sensors is deployed; according to a preset startup interval, the sensor cluster is started to monitor the transmission tower and obtain multiple groups of transmission tower monitoring images at different time points.

[0020] Specifically, a detailed geographical, climatic and environmental analysis is conducted on the target area where the transmission towers are located to determine which areas are the focus of monitoring. Based on the analysis results, multiple deployment points are selected that can fully cover the key parts of the transmission towers and are not easily obstructed. At each deployment point, a sensor cluster consisting of visual sensors (such as high-definition cameras) and infrared sensors (such as infrared thermal imagers) is deployed. These two sensors are used to obtain standard monitoring images and infrared monitoring images of the transmission towers, respectively. According to the monitoring needs and system resources, the startup interval of the sensor cluster is preset. This interval can be a fixed time interval (such as once an hour) or can be triggered based on specific conditions (such as weather changes, grid load changes, etc.). According to the preset startup interval, the sensor cluster is started to monitor the transmission towers. During the monitoring process, the visual sensors and infrared sensors obtain standard monitoring images and infrared monitoring images of the transmission towers, respectively.

[0021] Furthermore, based on the target area where the transmission tower is located, multiple deployment points are determined, and the method includes:

[0022] Acquire geographic information data of the target area; perform fault frequency analysis on historical deformation monitoring data of the transmission tower to determine multiple risk areas; and determine multiple deployment points based on the geographic information data and the risk areas.

[0023] Specifically, geographic information data of the target area are collected using geographic information systems (GIS) and remote sensing technology. These data may include topography, soil type, vegetation cover, meteorological conditions, hydrological conditions, etc. The collected geographic information data are sorted and analyzed to extract key information related to the stability of transmission towers, such as geological structure, terrain slope, groundwater level, etc. The historical deformation monitoring data of transmission towers are obtained from the database, and the collected historical deformation monitoring data are statistically analyzed to identify areas with frequent deformation or high failure rates. These areas may be more prone to deformation or failure due to geological conditions, environmental factors or design defects. Based on the results of the failure frequency analysis, the target area where the transmission towers are located is divided into multiple risk areas. The geographic information data is combined with the analysis results of the risk areas, and a comprehensive assessment is conducted by an expert team. Based on the comprehensive assessment results, multiple deployment points are selected. These points can fully cover the key parts and potential risk areas of the transmission towers while avoiding blind spots.

[0024] Time series analysis is performed on the multiple groups of transmission tower monitoring images at different time points to obtain time series image features.

[0025] Time series analysis is performed on multiple sets of transmission tower monitoring images at different time points, aiming to extract time-varying features from the image sequence, namely time-series image features. Time-series image features can reflect the state changes of transmission towers at different time points and provide an important basis for subsequent monitoring, evaluation and early warning.

[0026] Furthermore, performing time series analysis on the plurality of groups of transmission tower monitoring images at different time points to obtain time series image features includes:

[0027] Image registration is performed on the multiple groups of transmission tower monitoring images at different time points to obtain multiple groups of aligned transmission tower monitoring images; time series analysis is performed on the multiple groups of aligned transmission tower monitoring images to extract time series image features of the transmission tower deformation trend.

[0028] Preferably, feature points are detected in each monitoring image. These feature points are usually easily identifiable and stable points in the image, such as corner points, edge intersections, etc.; the detected feature points are described to generate feature descriptors so as to match similar feature points between different images; the feature descriptors are used to find matching feature point pairs between monitoring images at different time points; based on the matched feature point pairs, the transformation matrix between the images (such as affine transformation, perspective transformation, etc.) is calculated, which describes the spatial transformation relationship from one image to another; the transformation matrix is applied to transform all monitoring images into a unified coordinate system to obtain multiple sets of aligned transmission tower monitoring images; the difference image between the aligned images at adjacent time points is calculated to highlight the changes in the transmission towers at different time points; features are extracted from the difference image, which are temporal image features, which may include the shape, size, direction, texture changes, etc. of the deformed area.

[0029] According to the temporal image features, the standard pole tower monitoring image and the infrared pole tower monitoring image are fused to generate a plurality of fused pole tower monitoring images, wherein the fused pole tower monitoring images have temperature marks.

[0030] According to the temporal image characteristics, the standard tower monitoring images and the infrared tower monitoring images are preprocessed to remove useless features in the images. Image fusion algorithms, such as weighted averaging method and multi-resolution fusion method (such as pyramid fusion and wavelet transform fusion), are used to fuse the preprocessed standard tower monitoring images with the infrared tower monitoring images to generate a fused tower monitoring image with temperature identification. The fused tower monitoring image not only contains the appearance information of the tower in the standard monitoring image, but also contains the temperature information and defect information in the infrared image.

[0031] Furthermore, based on the temporal image features, the standard tower monitoring image and the infrared tower monitoring image are fused to generate a fused tower monitoring image, wherein the fused tower monitoring image has a temperature mark. The method includes:

[0032] Based on the temporal image features, structural change features and temperature features are extracted from the standard pole tower monitoring image and the infrared pole tower monitoring image; feature matching is performed on the structural change features and the temperature features, and based on the feature matching results, the standard pole tower monitoring image and the infrared pole tower monitoring image are layered fused to generate a fused pole tower monitoring image.

[0033] Specifically, based on the temporal image features, the structural features of the pole tower and its surrounding environment, such as edges, corners, textures, etc., are extracted from the standard pole tower monitoring image, and the temperature features are extracted from the infrared pole tower monitoring image; the extracted structural change features and temperature features are matched, and the matching can be based on spatial position, shape similarity or other advanced features (such as features obtained through machine learning training). During the matching process, the temporal image features can be used to help determine which structural changes are associated with temperature changes; based on the results of feature matching, the standard pole tower monitoring image and the infrared pole tower monitoring image are layered and fused. During the fusion process, different weights can be given to different features or image layers to highlight important information. A variety of fusion techniques can be used, such as multi-resolution fusion, region-based fusion or pixel-based fusion, to ensure the quality and accuracy of the fused image, and finally generate a fused pole tower monitoring image.

[0034] Furthermore, if Figure 2 As shown, based on the feature matching result, the standard tower monitoring image and the infrared tower monitoring image are layered-fused to generate a fused tower monitoring image. The method includes:

[0035] According to the structural change characteristics and the temperature characteristics, the standard tower monitoring image and the infrared tower monitoring image are divided into multiple feature layers; based on the feature layers and the feature matching results, the corresponding feature layers of the standard tower monitoring image and the infrared tower monitoring image are mapped to obtain mapping results; based on the mapping results, the corresponding feature layers of the standard tower monitoring image and the infrared tower monitoring image are weightedly fused, and the fused image is reconstructed to generate the fused tower monitoring image.

[0036] Specifically, based on the extracted structural change features and temperature features, the standard tower monitoring image and the infrared tower monitoring image are each divided into multiple feature layers. These feature layers can be divided based on different feature types (such as edges, textures, temperature distribution, etc.) or spatial regions (such as different parts of the tower or different areas of the surrounding environment). Based on the feature matching results, the corresponding feature layers of the standard tower monitoring image and the infrared tower monitoring image are mapped to find the spatially and feature-corresponding parts of the two images. The mapping result is a correspondence relationship, which indicates which feature layer in the standard image corresponds to which feature layer in the infrared image. For each pair of mapped feature layers, an appropriate fusion strategy is selected for weighted fusion based on the needs and purpose of fusion to effectively combine the information in the two images while retaining their respective important features. The weighting coefficient can be determined based on the importance, reliability, or user preference of the feature. For example, the standard image can be given a higher weight for areas with rich structural details, while the infrared image can be given a higher weight for areas with significant temperature changes. All fused feature layers are recombined according to the structure of the original images to form a fused tower monitoring image.

[0037] Feature extraction is performed on each of the multiple fused tower monitoring images to obtain multiple structural reference points of multiple groups of transmission towers.

[0038] Each fused tower monitoring image is preprocessed as necessary, including denoising, contrast enhancement, brightness adjustment, etc., to improve image quality and make features more obvious; according to the structural characteristics of the transmission tower and the characteristics of the fused image, a suitable feature extraction strategy is selected, including shape-based feature extraction (such as edge detection, corner detection), texture-based feature extraction, template matching-based method or deep learning-based feature extraction; the selected feature extraction strategy is applied to each fused tower monitoring image to extract features, with the goal of identifying multiple structural reference points on the tower; edge detection algorithms can be used to identify the edges and contours of the tower, and then locate intersections or support points; the extracted structural reference points are verified to ensure that they are accurate, reliable, and consistent with the actual structure of the transmission tower.

[0039] Based on the structural reference points, a plurality of three-dimensional structural models of transmission towers are generated using three-dimensional reconstruction technology, and deformation analysis is performed on the plurality of three-dimensional structural models of transmission towers to obtain deformation data and temperature change data of the transmission towers, wherein the deformation data and the temperature change data have a one-to-one correspondence.

[0040] Using 3D reconstruction technology, such as methods based on multi-view geometry, the extracted structural reference points are converted into point clouds in three-dimensional space, and the surface of the point cloud is reconstructed to generate a three-dimensional structural model of the transmission tower. This model can accurately reflect the shape, size and structural characteristics of the tower; obtain the three-dimensional structural model of the same transmission tower at different time points; compare the three-dimensional models at different time points, and measure the deformation of the tower in various directions, which can be obtained by calculating the distance change, angle change or volume change between corresponding points on the model; the deformation data should include parameters such as displacement, rotation, strain, etc., and can accurately reflect the deformation of the tower in different time periods; extract temperature information from the fused image, and this data corresponds to the structural reference points or areas in the three-dimensional structural model. Match the deformation data with the temperature change data to ensure that they have a one-to-one correspondence in time and space.

[0041] Furthermore, the deformation analysis is performed on the three-dimensional structural models of the plurality of transmission towers, and the method includes:

[0042] Deviation analysis is performed on the three-dimensional structural models of the multiple transmission towers to obtain temperature deviation data and displacement deviation data; when the temperature deviation data is greater than a temperature deviation threshold, the temperature deviation data is added to the temperature change data; when the displacement deviation data is greater than a displacement deviation threshold, the displacement deviation data is added to the deformation data.

[0043] Specifically, a baseline 3D structural model is selected as a reference. The baseline model is typically acquired under normal conditions (e.g., standard temperature, no external forces), or a representative model selected at the beginning of the analysis. For each 3D structural model of the transmission tower to be analyzed, it is compared with the baseline model, and the temperature deviation and displacement deviation of each corresponding point or region are calculated. The temperature deviation can be obtained by comparing the temperature values in the infrared monitoring data, while the displacement deviation can be obtained by calculating the spatial distance difference between corresponding points in the 3D model. Temperature deviation thresholds and displacement deviation thresholds are set based on the design specifications, historical data, or expert experience of the transmission tower. For each temperature deviation data point, if it is greater than the set temperature deviation threshold, it is considered a significant temperature change and needs to be added to the temperature change data. For each displacement deviation data point, if it is greater than the set displacement deviation threshold, it is considered a significant displacement change, i.e., the tower has undergone significant deformation. In this case, the displacement deviation data needs to be added to the deformation data. Through the above steps, in-depth deformation analysis can be performed on the 3D structural models of multiple transmission towers, and valuable temperature change and displacement change data can be obtained, providing strong support for the safe operation of the transmission towers.

[0044] When the temperature change data is greater than a preset temperature threshold, positioning is performed based on the corresponding relationship to obtain abnormal structure points.

[0045] When performing deformation analysis on a three-dimensional transmission tower model, if temperature variation data is detected to be greater than a preset temperature threshold, this typically indicates that a certain area on the tower may have been subjected to abnormal thermal stress or other temperature-related influences. Based on the correspondence between this temperature variation and the three-dimensional structural model, the abnormal structural point can be located and retrieved. Specifically, when the temperature variation of a certain area or point exceeds a preset threshold, it is marked as an abnormal temperature variation area. Using the previously established correspondence between the three-dimensional structural model and the temperature monitoring data, the abnormal temperature variation area is mapped to the three-dimensional structural model. After locating the abnormal temperature variation area on the three-dimensional structural model, the structural characteristics of the area are further analyzed to determine which structural points or components may have been affected, thereby retrieving the abnormal structural point.

[0046] The deformation trend of the abnormal structural point is analyzed in combination with the deformation data to obtain the deformation amount. When the deformation amount is greater than a preset deformation warning threshold, a warning signal is generated.

[0047] Collect and integrate deformation data from the three-dimensional structural model of the transmission tower and information on abnormal structural points previously identified through temperature monitoring. This data includes displacement measurements of each abnormal structural point at different time points, such as horizontal displacement, vertical displacement, or angle change. Arrange the deformation data in chronological order and observe its changing trend over time. Based on the smoothed data, select an appropriate mathematical model (such as a linear model, polynomial model, or exponential model) to fit the deformation trend. Based on the results of the trend analysis, calculate the total deformation of the abnormal structural point at the current time point. This deformation can be the cumulative displacement relative to the initial state, or the average deformation rate or maximum deformation within a specific time period. Set a reasonable deformation warning threshold based on the design specifications, historical data, and expert experience of the transmission tower. Compare the calculated deformation with the preset deformation warning threshold. If the deformation is greater than the warning threshold, generate a warning signal.

[0048] Furthermore, after generating the warning signal, the method further includes:

[0049] Defect identification is performed based on the fused tower monitoring image to obtain a defect identification result; a defect analysis model is constructed to quantitatively analyze the defect identification result to generate a defect assessment index; when the defect assessment index is greater than a defect warning threshold, the early warning signal is generated.

[0050] Preferably, image processing technology and machine learning algorithms (such as convolutional neural networks (CNN), support vector machines (SVM), etc.) are used to identify defects in the fused image. Defect identification may include but is not limited to wire breakage or loosening, tower tilt or cracks, etc.; a machine learning model is trained based on historical defect data, and a defect analysis model is obtained after training. The defect analysis model can evaluate the severity of the defect, the impact on the safety of the tower structure, and the possible consequences; the defect assessment index is compared with a preset defect warning threshold. The preset defect warning threshold is set according to the design specifications, historical data, expert experience, and safety standards of the transmission tower, and is used to determine whether the defect has reached a level that requires an early warning. If the defect assessment index is greater than the preset defect warning threshold, an early warning signal is generated.

[0051] In summary, the embodiments of the present application have at least the following technical effects:

[0052] First, multiple sensor clusters are deployed in the target area where transmission towers are located. Based on these sensor clusters, the transmission towers are monitored, acquiring multiple sets of transmission tower monitoring images at different time points. These images include standard tower monitoring images and infrared tower monitoring images. Next, time series analysis is performed on these multiple sets of transmission tower monitoring images at different time points to obtain temporal image features. Furthermore, based on these temporal image features, the standard tower monitoring images and infrared tower monitoring images are fused to generate multiple fused tower monitoring images. These fused tower monitoring images are then temperature-labeled. Feature extraction is then performed on each of these fused tower monitoring images to obtain multiple sets of structural reference points for the transmission towers. Based on these structural reference points, 3D reconstruction technology is used to generate multiple 3D structural models of the transmission towers. Deformation analysis is then performed on these 3D structural models to obtain deformation data and temperature change data for the transmission towers, with the deformation data and temperature change data having a one-to-one correspondence. When the temperature change data exceeds a preset temperature threshold, the system locates the abnormal structure point based on the corresponding relationship. Finally, the deformation trend of the abnormal structure point is analyzed based on the deformation data to obtain the deformation amount. When the deformation amount exceeds the preset deformation warning threshold, an early warning signal is generated. This solves the technical problem of low accuracy in deformation monitoring of transmission towers in complex environments, achieving the technical effect of improving the accuracy of deformation monitoring of transmission towers in complex environments.

[0053] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0054] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

[0055] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. A method for monitoring and early warning of transmission tower deformation based on image feature analysis, characterized in that: The method comprises: Deploy multiple sensor clusters in the target area where the transmission towers are located. Based on the sensor clusters, the transmission towers are monitored to obtain multiple sets of transmission tower monitoring images at different time points. The transmission tower monitoring images include standard tower monitoring images and infrared tower monitoring images. Performing time series analysis on the multiple groups of transmission tower monitoring images at different time points to obtain time series image features; According to the temporal image features, the standard tower monitoring image and the infrared tower monitoring image are fused to generate a plurality of fused tower monitoring images, wherein the fused tower monitoring images have temperature marks; Performing feature extraction on the multiple fused tower monitoring images respectively to obtain multiple structural reference points of multiple groups of transmission towers; Based on the structural reference points, a plurality of three-dimensional structural models of transmission towers are generated using a three-dimensional reconstruction technology, and deformation analysis is performed on the plurality of three-dimensional structural models of transmission towers to obtain deformation data and temperature change data of the transmission towers, wherein the deformation data and the temperature change data have a one-to-one correspondence; When the temperature change data is greater than a preset temperature threshold, positioning is performed based on the corresponding relationship to obtain an abnormal structure point; The deformation trend of the abnormal structural point is analyzed in combination with the deformation data to obtain the deformation amount. When the deformation amount is greater than a preset deformation warning threshold, a warning signal is generated.

2. The method for monitoring and early warning of transmission tower deformation based on image feature analysis according to claim 1, characterized in that: Deploy multiple sensor clusters in a target area where a transmission tower is located, monitor the transmission tower based on the sensor clusters, and obtain multiple sets of transmission tower monitoring images at different time points. The method includes: Determine multiple deployment points based on the target area where the transmission towers are located; Deploying a sensor cluster consisting of visual sensors and infrared sensors at the plurality of deployment points; According to a preset start-up interval, the sensor cluster is started to monitor the transmission tower, and multiple groups of transmission tower monitoring images at different time points are obtained.

3. The method for monitoring and early warning of transmission tower deformation based on image feature analysis according to claim 2, characterized in that: Based on the target area where the transmission tower is located, a plurality of deployment points are determined, and the method includes: Acquiring geographic information data of the target area; Performing fault frequency analysis on historical deformation monitoring data of the transmission tower to identify multiple risk areas; Based on the geographic information data and the risk area, multiple deployment points are determined.

4. The method for monitoring and early warning of transmission tower deformation based on image feature analysis according to claim 1, characterized in that: Performing time series analysis on the multiple groups of transmission tower monitoring images at different time points to obtain time series image features, the method comprising: Performing image registration on the multiple groups of transmission tower monitoring images at different time points to obtain multiple groups of aligned transmission tower monitoring images; Time series analysis is performed on the multiple groups of aligned transmission tower monitoring images to extract time series image features of the transmission tower deformation trend.

5. The method for monitoring and early warning of transmission tower deformation based on image feature analysis according to claim 4, characterized in that: According to the temporal image features, the standard tower monitoring image and the infrared tower monitoring image are fused to generate a fused tower monitoring image, wherein the fused tower monitoring image has a temperature mark. The method includes: Extracting structural change features and temperature features from the standard tower monitoring image and the infrared tower monitoring image based on the temporal image features; Feature matching is performed on the structural change feature and the temperature feature, and based on the feature matching result, the standard tower monitoring image and the infrared tower monitoring image are layered-fused to generate a fused tower monitoring image.

6. The method for monitoring and early warning of transmission tower deformation based on image feature analysis according to claim 5, characterized in that: Based on the feature matching result, the standard tower monitoring image and the infrared tower monitoring image are layered-fused to generate a fused tower monitoring image. The method includes: Dividing the standard pole tower monitoring image and the infrared pole tower monitoring image into a plurality of feature layers according to the structural change characteristics and the temperature characteristics; Based on the feature layer and the feature matching result, mapping the corresponding feature layers of the standard tower monitoring image and the infrared tower monitoring image to obtain a mapping result; Based on the mapping result, weighted fusion is performed on corresponding feature layers of the standard pole tower monitoring image and the infrared pole tower monitoring image, and the fused image is reconstructed to generate the fused pole tower monitoring image.

7. The method for monitoring and early warning of transmission tower deformation based on image feature analysis according to claim 1, characterized in that: Performing deformation analysis on the three-dimensional structural models of the plurality of transmission towers, the method comprising: Performing deviation analysis on the three-dimensional structural models of the plurality of transmission towers to obtain temperature deviation data and displacement deviation data; When the temperature deviation data is greater than a temperature deviation threshold, adding the temperature deviation data to the temperature change data; When the displacement deviation data is greater than a displacement deviation threshold, the displacement deviation data is added to the deformation data.

8. The method for monitoring and early warning of transmission tower deformation based on image feature analysis according to claim 1, characterized in that: After generating the warning signal, the method further includes: Perform defect recognition based on the fused tower monitoring image to obtain a defect recognition result; Constructing a defect analysis model to quantitatively analyze the defect identification results and generate defect assessment indicators; When the defect assessment index is greater than the defect warning threshold, the early warning signal is generated.

Images