Iron tower structure defect detection method and system

Abstract

The invention provides an iron tower structure defect detection method and system, and is suitable for the technical field of data processing, and the method comprises the steps: generating to-be-detected iron tower image feature information and iron tower defect virtual image feature information according to to-be-detected iron tower image information and iron tower defect virtual image information; performing registration processing according to the to-be-detected iron tower image feature information and the iron tower defect virtual image feature information to obtain to-be-detected iron tower image registration information and iron tower defect virtual image registration information; performing fusion processing on the to-be-detected iron tower image registration information and the iron tower defect virtual image registration information according to the to-be-detected iron tower sensor measurement information to obtain to-be-detected iron tower image fusion information; and generating iron tower structure defect detection information according to the to-be-detected iron tower sensor measurement information and the to-be-detected iron tower image fusion information. The method effectively improves the detection rate and recognition precision of the defects of the communication iron tower structure, and achieves the comprehensive evaluation of the health state of the iron tower structure.

Description

A method and system for detecting structural defects in iron towers Technical Field

[0001] This application belongs to the field of data processing technology, and in particular relates to a method and system for detecting defects in iron tower structures. Background Technology

[0002] With the widespread coverage of communication networks, the number of communication towers continues to surge. Their structural safety is directly related to the stable operation of communication networks, making the detection of structural defects in towers a core industry need.

[0003] In existing technologies, drones are typically used to take aerial photos of communication towers, and machine learning or deep learning models are combined to identify defects in the tower structure.

[0004] However, existing technologies are not sensitive enough to identify small structural defects, which can easily lead to missed detections, and the overall accuracy of defect detection is greatly affected by environmental factors. Summary of the Invention

[0005] In view of this, the present application provides a method and system for detecting defects in iron tower structures, aiming to solve the problems of low sensitivity in identifying small defects and the fact that the detection accuracy is greatly affected by the environment in the prior art.

[0006] The first aspect of this application provides a method for detecting defects in iron tower structures, including:

[0007] Acquire image information of multiple towers to be inspected and sensor measurement information of multiple towers to be inspected;

[0008] Generate virtual image information of defects in multiple iron towers;

[0009] Based on the preset tower image feature extraction model, feature extraction and parsing calculations are performed on the multiple tower images to be detected and the multiple tower defect virtual image information to generate multiple tower image feature information and multiple tower defect virtual image feature information.

[0010] Based on the feature information of the multiple tower images to be detected and the feature information of the multiple tower defect virtual images, registration processing is performed to obtain the tower image registration information and the tower defect virtual image registration information.

[0011] Based on the measurement information of the multiple tower sensors to be detected, the image registration information of the tower to be detected and the virtual image registration information of the tower defects are fused to obtain the fused information of multiple tower images to be detected.

[0012] Based on the measurement information from the multiple tower sensors and the fusion information from the multiple tower images, multiple tower structural defect detection information is generated.

[0013] A second aspect of this application provides a tower structure defect detection system, comprising:

[0014] The information acquisition module is used to acquire image information of multiple towers to be detected and sensor measurement information of multiple towers to be detected.

[0015] The tower defect virtual image information generation module is used to generate multiple tower defect virtual image information.

[0016] The image feature information generation module is used to perform feature extraction and analysis calculation based on the preset tower image feature extraction model, according to the multiple tower images to be detected and the multiple virtual images of tower defects, to generate multiple tower image feature information and multiple virtual image feature information of tower defects.

[0017] The image registration information generation module is used to perform registration processing based on the feature information of the multiple tower images to be detected and the feature information of the multiple tower defect virtual images to obtain the tower image registration information and the tower defect virtual image registration information.

[0018] The image fusion information generation module for the tower to be detected is used to fuse the image registration information of the tower to be detected and the virtual image registration information of the tower defects based on the measurement information of the multiple tower sensors to be detected, so as to obtain the image fusion information of multiple towers to be detected.

[0019] The tower structure defect detection information generation module is used to generate multiple tower structure defect detection information based on the measurement information of the multiple tower sensors and the image fusion information of the multiple towers to be detected.

[0020] A third aspect of this application provides a terminal device, which includes a memory and a processor. The memory stores a computer program that can run on the processor. When the processor executes the computer program, it implements the steps of the tower structure defect detection method described in the first aspect above.

[0021] A fourth aspect of this application provides a computer-readable storage medium, comprising: storing a computer program, wherein when executed by a processor, the computer program implements the steps of the tower structure defect detection method described in the first aspect above.

[0022] The beneficial effects of this application embodiment compared with the prior art are: this application enriches the basic data dimensions of tower structure detection, supplements the scarce defect samples in real detection, realizes the accurate characterization of tower structure features and defect features, and achieves deep complementarity between visual features and quantitative parameters, thereby significantly improving the accuracy and comprehensiveness of communication tower structure defect detection, and effectively solving the problems of defect omission and misjudgment in traditional detection methods. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 is a schematic diagram of the implementation process of the tower structure defect detection method provided in Embodiment 1 of this application;

[0025] Figure 2 is a schematic diagram of the implementation process of the tower structure defect detection method provided in Embodiment 2 of this application;

[0026] Figure 3 is a schematic diagram of the implementation process of the tower structure defect detection method provided in Embodiment 3 of this application;

[0027] Figure 4 is a schematic diagram of the implementation process of the tower structure defect detection method provided in Embodiment 4 of this application;

[0028] Figure 5 is a schematic diagram of the implementation process of the tower structure defect detection method provided in Embodiment 5 of this application;

[0029] Figure 6 is a schematic diagram of the implementation process of the tower structure defect detection method provided in Embodiment 6 of this application;

[0030] Figure 7 is a schematic diagram of the implementation process of the tower structure defect detection method provided in Embodiment 7 of this application;

[0031] Figure 8 is a schematic diagram of the structure of the tower structural defect detection system provided in an embodiment of this application;

[0032] Figure 9 is a schematic diagram of the terminal device provided in an embodiment of this application. Detailed Implementation

[0033] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0034] To illustrate the technical solution described in this application, specific embodiments are provided below.

[0035] Figure 1 shows a flowchart of the tower structure defect detection method provided in Embodiment 1 of this application, which is described in detail below:

[0036] Step S101: Acquire image information of multiple towers to be detected and sensor measurement information of multiple towers to be detected.

[0037] In this embodiment, the multiple tower images to be inspected can be obtained by taking multi-angle, multi-height photos of the communication tower using a drone equipped with a high-definition camera, by continuously capturing images of key parts of the tower using a fixed monitoring device, or by manually capturing images of difficult-to-detect areas of the tower using a handheld camera. The sensor measurement information of the multiple towers to be inspected can include stress data of tower components collected by stress sensors, tilt angle information of the tower body measured by tilt sensors, and vibration frequency data of the tower structure recorded by vibration sensors. Furthermore, the multiple tower images and sensor measurement information can be aligned using timestamp synchronization technology to ensure consistency between the two types of data in the time dimension.

[0038] Step S102: Generate virtual image information of multiple tower defects.

[0039] In this embodiment, a standard 3D model of a communication tower can be constructed using professional 3D modeling software. It can also be based on common defect types of communication towers, such as component corrosion, loose bolts, weld cracks, and tower material deformation. Geometric features and appearance attributes of various defects can be artificially added to the standard 3D model. Then, the rendering engine can be used to set parameters such as lighting, texture, and viewpoint on the 3D model with defect features, thereby generating multiple virtual image information of tower defects with different defect types, different defect degrees, and different shooting angles.

[0040] Step S103: Based on the preset tower image feature extraction model, feature extraction and analysis calculation are performed on the multiple tower image information to be detected and the multiple tower defect virtual image information to generate multiple tower image feature information to be detected and multiple tower defect virtual image feature information.

[0041] In this embodiment, the preset tower image feature extraction model can be pre-set manually. This model can be built based on deep learning algorithms and can include network structures such as convolutional layers, pooling layers, and fully connected layers. Multiple tower images to be detected and multiple virtual images of tower defects can be input into the model. Then, shallow features such as edge features, texture features, and shape features in the image are extracted through the model's convolution operation. Subsequently, the shallow features are reduced in dimensionality through pooling operation. Finally, the processed features are analyzed and calculated through a fully connected layer, thereby generating multiple tower image feature information that can characterize the structural attributes of the tower to be detected and multiple virtual image feature information that can characterize the essential features of the defects.

[0042] Step S104: Perform registration processing based on the feature information of the multiple tower images to be detected and the feature information of the multiple tower defect virtual images to obtain the registration information of the tower images to be detected and the registration information of the tower defect virtual images.

[0043] In this embodiment, the similarity between the feature information of multiple tower images to be detected and the feature information of multiple virtual images of tower defects can be calculated. The matching feature point pair between the two types of feature information can be selected by using a feature point matching algorithm. Then, the virtual image feature information of tower defects can be transformed into spatial coordinates based on the standard structural features of the tower to be detected. Then, the position, angle, scale and other parameters of the virtual image features can be adjusted to make the two types of feature information accurately aligned in the spatial dimension, thereby obtaining the registration information of the tower image to be detected and the registration information of the virtual image of tower defects after spatial calibration.

[0044] Step S105: Based on the measurement information of the multiple tower sensors to be detected, the image registration information of the tower to be detected and the virtual image registration information of the tower defects are fused to obtain the fused information of multiple tower images to be detected.

[0045] In this embodiment, the stress data, tilt angle data, vibration frequency data, and other quantitative parameters from the sensor measurement information of multiple towers to be tested can be used as constraints. Then, the image registration information of the tower to be tested and the virtual image registration information of the tower defects can be weighted and fused. Then, the features of corresponding regions in the two types of image information after registration can be complementaryly verified, thereby enhancing the identification of defect features in the image and weakening the interference caused by environmental noise, shooting angle, and other factors. This generates multiple tower image fusion information that fuses visual features and quantitative parameter features.

[0046] Step S106: Generate multiple tower structural defect detection information based on the measurement information of the multiple tower sensors and the image fusion information of the multiple towers to be detected.

[0047] In this embodiment, the measurement information of multiple tower sensors to be tested can be used as the quantitative basis for defect judgment, and the image fusion information of multiple towers to be tested can be used as the visual basis for defect judgment. Then, the two types of data are jointly analyzed. Then, by comparing the actual data of the tower to be tested with the standard threshold, the tower component area with abnormality can be identified. Then, by combining the defect features in the image fusion information, the key information such as the defect type, defect location, and defect severity of the area can be determined, thereby generating multiple tower structure defect detection information containing complete defect attributes.

[0048] The tower structure defect detection method provided in this application enriches the basic data dimensions of tower structure detection, supplements the scarce defect samples in real-world detection, and achieves accurate characterization of tower structure features and defect features, as well as deep complementarity between visual features and quantitative parameters. This significantly improves the accuracy and comprehensiveness of communication tower structure defect detection and effectively solves the problems of missed and misjudged defects in traditional detection methods.

[0049] Figure 2 shows a flowchart of the tower structure defect detection method provided in Embodiment 2 of this application. The difference between this method and Embodiment 1 is that:

[0050] The preset tower image feature extraction model includes a preset first-scale tower image feature extraction sub-model and a preset second-scale tower image feature extraction sub-model;

[0051] Step S103 specifically includes:

[0052] Step S201: Generate a pixel matrix of the tower to be detected and a virtual pixel matrix of tower defects based on the multiple tower image information and the multiple tower defect virtual image information.

[0053] In this embodiment, multiple images of towers to be detected and multiple virtual images of tower defects can be read using professional image processing tools. These images can then be converted into a standard pixel matrix format, and the converted matrices can be used as the pixel matrix of the towers to be detected and the virtual pixel matrix of tower defects, respectively.

[0054] Step S202: Based on the preset first-scale tower image feature extraction sub-model and the preset second-scale tower image feature extraction sub-model, analyze and calculate the tower pixel matrix to be detected and the tower defect virtual pixel matrix to generate the first-scale tower pixel matrix variable, the second-scale tower pixel matrix variable, the first-scale tower defect virtual pixel matrix variable, and the second-scale tower defect virtual pixel matrix variable.

[0055] In this embodiment, both the preset first-scale tower image feature extraction sub-model and the preset second-scale tower image feature extraction sub-model can be pre-defined by humans. They can be constructed based on a deep learning convolutional neural network structure or designed based on traditional image processing feature extraction algorithms. The feature extraction accuracy corresponding to the first scale can be higher than that of the second scale, or the feature extraction range corresponding to the second scale can be larger than that of the first scale. It is understood that the descriptions of the first scale and the second scale are only used as examples to illustrate that tower image feature extraction sub-models with different processing scales analyze and calculate the pixel matrix of the tower to be detected and the virtual pixel matrix of tower defects. In actual applications, there can also be a third-scale tower image feature extraction sub-model, a fourth-scale tower image feature extraction sub-model, and so on. The pixel matrix of the tower to be detected and the virtual pixel matrix of the tower defects can be used as input data for the preset first-scale tower image feature extraction sub-model and the preset second-scale tower image feature extraction sub-model, respectively. After analysis and calculation by the preset first-scale tower image feature extraction sub-model and the preset second-scale tower image feature extraction sub-model, the calculation results are used as the first-scale tower pixel matrix variable, the second-scale tower pixel matrix variable, the first-scale tower defect virtual pixel matrix variable, and the second-scale tower defect virtual pixel matrix variable, respectively.

[0056] Step S203: Based on the first-scale tower to be detected pixel matrix variable, the second-scale tower to be detected pixel matrix variable, the first-scale tower defect virtual pixel matrix variable, the second-scale tower defect virtual pixel matrix variable, and the preset tower feature extraction radius information, generate multiple tower image feature information and multiple tower defect virtual image feature information.

[0057] In this embodiment, the preset tower feature extraction radius information can be manually preset, and its value can be set according to the resolution of the tower image and the distribution density of feature points. It can be achieved by first fusing the pixel matrix variables of the tower to be detected at the first scale and the second scale, and then fusing the virtual pixel matrix variables of tower defects at the first scale and the second scale. Then, using the preset tower feature extraction radius information as a range, feature point information is extracted from the fused matrix variables. Finally, the extracted feature point information is format-converted and quantized to generate multiple tower image feature information and multiple virtual tower defect image feature information.

[0058] The tower structure defect detection method provided in this application realizes multi-scale analysis and calculation of multiple tower images and multiple virtual images of tower defects, so as to simultaneously capture the fine structural features and macroscopic overall features in the tower images. Combined with the preset tower feature extraction radius information, the feature range is accurately extracted, making the generated tower image feature information and tower defect virtual image feature information more comprehensive and accurate, thereby improving the accuracy of subsequent image feature registration processing.

[0059] Figure 3 shows a flowchart of the tower structure defect detection method provided in Embodiment 3 of this application. The difference between this method and Embodiment 2 is that step S203 specifically includes:

[0060] Step S301: Perform a difference operation on the first-scale tower pixel matrix variable to be detected and the second-scale tower pixel matrix variable to be detected to obtain the scale-difference tower pixel matrix variable to be detected.

[0061] In this embodiment, when the value of the first-scale tower pixel matrix variable to be detected is greater than the value of the second-scale tower pixel matrix variable to be detected, a difference operation is performed by subtracting the second-scale tower pixel matrix variable from the first-scale tower pixel matrix variable to obtain the scale-difference tower pixel matrix variable to be detected; when the value of the second-scale tower pixel matrix variable to be detected is greater than the value of the first-scale tower pixel matrix variable to be detected, a difference operation is performed by subtracting the first-scale tower pixel matrix variable from the second-scale tower pixel matrix variable to obtain the scale-difference tower pixel matrix variable to be detected.

[0062] Step S302: Perform a difference operation on the virtual pixel matrix variable of the first-scale tower defect and the virtual pixel matrix variable of the second-scale tower defect to obtain the scale-difference virtual pixel matrix variable of the tower defect.

[0063] In this embodiment, when the value of the virtual pixel matrix variable for tower defects at the first scale is greater than the value of the virtual pixel matrix variable for tower defects at the second scale, a difference operation is performed by subtracting the virtual pixel matrix variable for tower defects at the second scale from the first scale to obtain the scale-difference virtual pixel matrix variable for tower defects; when the value of the virtual pixel matrix variable for tower defects at the second scale is greater than the value of the virtual pixel matrix variable for tower defects at the first scale, a difference operation is performed by subtracting the virtual pixel matrix variable for tower defects at the second scale from the first scale to obtain the scale-difference virtual pixel matrix variable for tower defects.

[0064] Step S303: Extract the extreme point coordinates of the scale difference tower pixel matrix variable to be detected and the extreme point coordinates of the scale difference tower defect virtual pixel matrix variable to obtain the feature center information of the tower image to be detected and the feature center information of the tower defect virtual image.

[0065] In this embodiment, it can be understood that both the scale-difference tower pixel matrix variable to be detected and the scale-difference tower defect virtual pixel matrix variable are in matrix form. It can be determined by traversing all elements in the matrix, identifying the maximum value element and the minimum value element, and then using the matrix coordinates corresponding to these extreme value elements as extreme value point coordinates. These extreme value point coordinates are then used as the feature center information of the tower image to be detected and the feature center information of the tower defect virtual image, respectively.

[0066] Step S304: Based on the preset tower feature extraction radius information, multiple tower image feature information and multiple tower defect virtual image feature information are generated with the center information of the tower image feature and the center information of the tower defect virtual image feature as the center.

[0067] In this embodiment, the preset tower feature extraction radius information can be manually preset, and its value can be adjusted according to the density of feature points in the tower image and the size of the defect features. It can be that the feature center information of the tower image to be detected and the feature center information of the virtual image of the tower defect are respectively used as the center, and the preset tower feature extraction radius information is used as the radius to delineate the feature extraction region. Then, the feature data of all pixels within this region are extracted, and the extracted feature data is integrated and encoded to generate multiple tower image feature information and multiple virtual image feature information of the tower defect.

[0068] The tower structure defect detection method provided in this application highlights the differences in tower image features and virtual defect image features at different scales by performing differential operations on pixel matrix variables at different scales, thereby enhancing feature recognition. By extracting the extreme point coordinates of the differential matrix, the feature center is accurately located, and the extraction range is defined by combining the preset tower feature extraction radius information. This makes the generated multiple tower image feature information to be detected and multiple virtual defect image feature information more targeted and representative, effectively improving the accuracy and reliability of feature extraction, providing high-quality feature data for subsequent registration and fusion processing, and thus improving the overall accuracy of tower structure defect detection.

[0069] Figure 4 shows a flowchart of the tower structure defect detection method provided in Embodiment 4 of this application. The difference between this method and Embodiment 1 is that step S104 specifically includes:

[0070] Step S401: Based on the preset tower image feature aggregation window, perform neighborhood feature aggregation processing on the feature information of the multiple tower images to be detected and the feature information of the multiple tower defects virtual images to be detected, respectively, to obtain the feature aggregation information of the multiple towers to be detected and the feature aggregation information of the multiple tower defects virtual images.

[0071] In this embodiment, the preset tower image feature aggregation window can be manually preset, and its size can be set according to the distribution density of the tower image feature information to be detected and the virtual image feature information of tower defects. The preset tower image feature aggregation window can be slid to sequentially cover the feature regions of multiple tower image feature information to be detected and multiple virtual image feature information of tower defects. Then, the mean or summation of the feature information within the coverage area of ​​each window is calculated, and the calculation result is used as the aggregation feature of the corresponding region, thereby generating multiple aggregated tower feature information to be detected and multiple aggregated virtual feature information of tower defects.

[0072] Step S402: Calculate the cosine similarity between the aggregated feature information of the multiple towers to be detected and the aggregated virtual feature information of the multiple tower defects, and obtain the aggregated similarity information of the multiple tower image features.

[0073] In this embodiment, each tower feature aggregation information to be detected can be paired with each tower defect virtual feature aggregation information, and then the cosine similarity calculation method can be used to measure the feature similarity between each pair of feature aggregation information. Then, the calculation result of each pair of feature aggregation information can be used as the corresponding tower image feature aggregation similarity information, thereby obtaining multiple tower image feature aggregation similarity information.

[0074] Step S403: Determine whether the similarity information of the aggregated features of the tower image is greater than the preset similarity threshold of tower feature registration; if yes, proceed to step S404; if no, proceed to step S405.

[0075] In this embodiment, the preset tower feature registration similarity threshold can be manually set, and its value can be adjusted according to the accuracy requirements of tower structural defect detection. It can be achieved by comparing the aggregated similarity information of each tower image feature with the preset tower feature registration similarity threshold one by one, thereby determining whether the aggregated feature information of the tower to be detected corresponding to the aggregated similarity information of the tower image feature meets the registration conditions with the virtual feature aggregation information of the tower defect.

[0076] Step S404: Based on the similarity information of the tower image features, the spatial coordinates of the tower to be detected and the virtual feature aggregation information of the tower defects are calculated to obtain the registration information of the tower image to be detected and the registration information of the virtual image of the tower defects.

[0077] In this embodiment, the spatial coordinates corresponding to the aggregated feature information of the tower to be detected can be used as a reference to calculate the spatial offset of the virtual feature aggregated information of the tower defect relative to the aggregated feature information of the tower to be detected. Then, the spatial coordinates of the virtual feature aggregated information of the tower defect can be adjusted according to the spatial offset to achieve spatial alignment between the two. Thus, the aligned aggregated feature information of the tower to be detected can be used as the registration information of the tower image to be detected, and the aligned virtual feature aggregated information of the tower defect can be used as the registration information of the virtual image of the tower defect.

[0078] Step S405: Skip the tower feature aggregation information to be detected and the tower defect virtual feature aggregation information corresponding to the tower image feature aggregation similarity information.

[0079] In this embodiment, when the similarity information of the tower image feature aggregation is not greater than the preset tower feature registration similarity threshold, it indicates that the similarity between the corresponding tower feature aggregation information to be detected and the tower defect virtual feature aggregation information is low and does not meet the registration requirements. Therefore, this pair of feature aggregation information is skipped directly, and subsequent spatial coordinate alignment calculation is not performed, thereby avoiding invalid features from interfering with the subsequent registration results.

[0080] The tower structural defect detection method provided in this application enhances the robustness of feature information, reduces the impact of local noise on registration, accurately measures the similarity of feature aggregation information, combines a preset tower feature registration similarity threshold to screen effective feature pairs, avoids invalid features from interfering with the registration results, and then performs spatial coordinate alignment calculations on the feature pairs that meet the requirements to generate registration information, thereby improving the accuracy and reliability of feature registration, and thus improving the accuracy and comprehensiveness of tower structural defect detection.

[0081] Figure 5 shows a flowchart of the tower structure defect detection method provided in Embodiment 5 of this application. The difference between this method and Embodiment 1 is that step S105 specifically includes:

[0082] Step S501: Extract the three-dimensional spatial coordinate range of the registration information of the tower image to be detected and the three-dimensional spatial coordinate range of the registration information of the virtual image of the tower defect, to obtain the registration coordinate domain of the tower to be detected and the virtual registration coordinate domain of the tower defect.

[0083] In this embodiment, all spatial coordinates contained in the registration information of multiple tower images to be detected can be traversed and statistically analyzed to determine the maximum and minimum values ​​of the horizontal, vertical, and vertical coordinates. Then, the three-dimensional spatial range can be delineated using these extreme values ​​as the registration coordinate domain of the tower to be detected. Similarly, the spatial coordinates of the registration information of multiple tower defect virtual images can be traversed and statistically analyzed to delineate the corresponding three-dimensional spatial range, thereby obtaining the virtual registration coordinate domain of the tower defect.

[0084] Step S502: Based on the registration coordinate domain of the tower to be detected, the virtual registration coordinate domain of the tower defect is scaled and spatially mapped proportionally to obtain the virtual registration mapping coordinate domain of the tower defect.

[0085] In this embodiment, the size ratio of the registration coordinate domain of the tower to be detected to the virtual registration coordinate domain of the tower defect in each coordinate axis direction can be calculated first. Then, the virtual registration coordinate domain of the tower defect can be scaled proportionally according to the ratio so that the size of the scaled virtual registration coordinate domain of the tower defect matches the size of the registration coordinate domain of the tower to be detected. Then, the scaled virtual registration coordinate domain of the tower defect is mapped to the corresponding spatial position of the registration coordinate domain of the tower to be detected through a spatial mapping algorithm, thereby obtaining the virtual registration mapping coordinate domain of the tower defect.

[0086] Step S503: Based on the registration coordinate domain of the tower to be detected, the virtual registration mapping coordinate domain of the tower defect, and the measurement information of multiple tower sensors to be detected, the image fusion information of multiple towers to be detected is obtained.

[0087] In this embodiment, the registration information of the tower image to be detected corresponding to the registration coordinate domain of the tower to be detected and the registration information of the virtual image of the tower defect corresponding to the virtual registration mapping coordinate domain of the tower defect can be associated and matched with the sensor measurement information of multiple towers to be detected. Then, the quantized data in the sensor measurement information is used as a weighting factor to perform weighted fusion calculation on the two types of registration information. The fusion calculation result is then used as the fusion feature of the corresponding coordinate region to generate multiple tower image fusion information to be detected.

[0088] The tower structure defect detection method provided in this application realizes accurate matching of multimodal data in the spatial dimension, achieves deep integration of visual registration information and quantified sensor data, enhances the comprehensiveness and reliability of fused information, and thus provides a high-quality fused data foundation for accurately generating tower structure defect detection information.

[0089] Figure 6 shows a flowchart of the implementation of the tower structure defect detection method provided in Embodiment Six of this application. The difference between it and Embodiment Five above is that:

[0090] The sensor measurement information of the tower to be tested includes the measurement information of the tower stress sensor, the measurement information of the tower tilt angle sensor, the measurement information of the tower displacement sensor, and the measurement information of the tower ambient temperature sensor.

[0091] Step S503 specifically includes:

[0092] Step S601: Generate structural deformation matrix information of the tower to be tested based on the measurement information of the multiple tower stress sensors, the multiple tower tilt sensors, and the multiple tower displacement sensors.

[0093] In this embodiment, the stress data corresponding to the stress sensor measurement information of multiple towers to be tested, the tilt data corresponding to the tilt sensor measurement information of multiple towers to be tested, and the displacement data corresponding to the displacement sensor measurement information of multiple towers to be tested can be correlated and mapped. Then, using the spatial coordinates of the registration coordinate domain of the tower to be tested as an index, the measurement data of various sensors are filled into the corresponding coordinate positions to form a matrix, thereby generating the structural deformation matrix information of the tower to be tested that can characterize the deformation state of the tower at different positions.

[0094] Step S602: Based on the registration coordinate domain of the tower to be detected, the virtual registration mapping coordinate domain of the tower defect, and the structural deformation matrix information of the tower to be detected, a fusion calculation is performed to obtain the structural deformation detection information of the tower to be detected.

[0095] In this embodiment, the registration information of the tower image corresponding to the registration coordinate domain of the tower to be detected and the registration information of the virtual image of the tower defect corresponding to the virtual registration mapping coordinate domain of the tower defect can be first aligned at the pixel level. Then, the deformation data in the deformation matrix information of the tower structure to be detected is used as weight to perform weighted fusion calculation on the two types of registration information after alignment. Finally, the fusion calculation result is associated with the corresponding deformation data to obtain the deformation detection information of the tower structure to be detected.

[0096] Step S603: Based on the measurement information of the ambient temperature sensors of the multiple towers to be tested, error compensation calculation is performed on the structural deformation detection information of the towers to be tested to obtain the image fusion information of the multiple towers to be tested.

[0097] In this embodiment, a temperature error compensation model can be established by analyzing the correlation between the measurement information of multiple temperature sensors on the tower under test and the structural deformation of the tower. Then, the measurement information of the temperature sensors on the tower under test is input into the model to obtain the error compensation amount. Finally, the error correction is completed by subtracting the corresponding error compensation amount from the structural deformation detection information of the tower under test, thereby generating image fusion information of multiple towers under test.

[0098] The tower structure defect detection method provided in this application accurately captures the deformation state of the tower structure, deeply fuses the deformation data with two types of registration information to obtain deformation detection information, and then combines the measurement information of the ambient temperature sensor for error compensation, effectively eliminating the interference of temperature factors on deformation detection, so that the fusion information of multiple tower images to be detected is more consistent with the actual structural state of the tower, thereby improving the accuracy and effectiveness of tower structure defect detection.

[0099] Figure 7 shows a flowchart of the tower structure defect detection method provided in Embodiment 7 of this application. The difference between this method and Embodiment 1 is that step S106 specifically includes:

[0100] Step S701: Based on the measurement information of the multiple tower sensors to be detected and the multiple preset tower sensor measurement threshold information, obtain multiple tower sensor measurement abnormality information.

[0101] In this embodiment, the preset tower sensor measurement threshold information can be manually preset, and its value can be determined according to the tower structure design standards and safety specifications. Each tower sensor measurement information to be tested can be compared with its corresponding preset tower sensor measurement threshold information. When the measurement information of a tower sensor to be tested exceeds the preset tower sensor measurement threshold information range, the sensor measurement information is marked as abnormal, thereby obtaining multiple abnormal tower sensor measurement information.

[0102] Step S702: Based on the abnormal measurement information of the multiple tower sensors to be detected, perform positioning calculation on the image fusion information of the multiple towers to be detected to obtain abnormal positioning information of the multiple towers to be detected.

[0103] In this embodiment, the sensor installation location information corresponding to the abnormal measurement information of multiple tower sensors to be detected can be extracted, and then the location information can be matched with the spatial coordinates of the image fusion information of multiple towers to be detected. Then, the specific coordinates of the abnormal location in the image fusion information can be determined by coordinate mapping calculation, thereby obtaining the abnormal location information of multiple towers to be detected.

[0104] Step S703: Based on the preset information on the types and quantities of defects in the tower structure, randomly extract the abnormal location information of the multiple towers to be detected to obtain multiple extracted abnormal location information of the towers.

[0105] In this embodiment, the preset number of types of tower structural defects can be manually preset, and its value can be set according to the number of common defect types in communication towers. The number of samples can be determined according to the preset number of types of tower structural defects, and then a random sampling algorithm can be used to extract the corresponding number of location information from multiple tower anomaly location information to be detected, thereby obtaining multiple extracted tower anomaly location information.

[0106] Step S704: Based on the multiple abnormal location information of the multiple towers to be detected and the multiple abnormal location information of the extracted towers, obtain multiple abnormal location information of the remaining towers.

[0107] In this embodiment, multiple abnormal tower location information to be detected can be taken as a whole set, multiple extracted abnormal tower location information can be taken as a subset, and then the elements in the subset can be removed from the whole set, and the remaining elements can be taken as multiple remaining abnormal tower location information.

[0108] Step S705: Calculate the Euclidean distance between the multiple extracted abnormal tower location information and the multiple remaining abnormal tower location information to obtain multiple abnormal tower location distance information.

[0109] In this embodiment, each extracted tower abnormal location information can be paired with each remaining tower abnormal location information, and then the Euclidean distance between each pair of location information can be calculated. The distance calculation result of each pair of location information can then be used as the corresponding tower abnormal location distance information, thereby obtaining multiple tower abnormal location distance information.

[0110] Step S706: Determine whether the abnormal tower positioning distance information is less than or equal to the preset abnormal tower positioning distance threshold information; if yes, proceed to step S707; if no, proceed to step S708.

[0111] In this embodiment, the preset tower anomaly location distance threshold information can be manually preset, and its value can be adjusted according to the component size of the tower structure and the defect detection accuracy requirements. It can be that each tower anomaly location distance information is compared with the preset tower anomaly location distance threshold information one by one, thereby determining whether the corresponding extracted tower anomaly location information and the remaining tower anomaly location information belong to the same defect area.

[0112] Step S707: Generate tower structural defect detection category information based on the extracted tower abnormal positioning information and the remaining tower abnormal positioning information corresponding to the tower abnormal positioning distance information.

[0113] In this embodiment, the extracted abnormal tower location information that meets the distance condition can be classified into the same defect category as the remaining abnormal tower location information. Then, combined with the preset defect category classification standard, the category is assigned a corresponding category identifier, thereby generating tower structure defect detection category information.

[0114] Step S708: Generate tower structural defect detection category information based on the extracted tower abnormal positioning information corresponding to the tower abnormal positioning distance information.

[0115] In this embodiment, when the abnormal tower location distance information is greater than the preset abnormal tower location distance threshold information, it indicates that the corresponding remaining abnormal tower location information and the extracted abnormal tower location information do not belong to the same defect area. Therefore, based solely on the extracted abnormal tower location information and combined with the preset defect category classification standard, a category identifier is assigned to it, thereby generating tower structure defect detection category information.

[0116] Step S709: Calculate the median of the multiple tower structure defect detection category information to obtain the median information of multiple tower structure defect detection categories.

[0117] In this embodiment, the information on multiple tower structure defect detection categories corresponding to the same defect category can be sorted and arranged, and then the median value after sorting can be selected as the median value of the category, thereby obtaining the median value information of multiple tower structure defect detection categories.

[0118] Step S710: Calculate the median information of the multiple tower structural defect detection categories and the Euclidean distance of the multiple extracted tower anomaly location information to obtain the median distance information of the multiple tower structural defect detection categories.

[0119] In this embodiment, the median information of each tower structural defect detection category can be paired with the corresponding multiple extracted tower anomaly location information, and then the Euclidean distance between each pair of information can be calculated. The calculation result is then used as the median distance information of multiple tower structural defect detection categories.

[0120] Step S711: Determine whether the median distance information of the tower structure defect detection category is less than the preset median distance threshold of the tower structure defect detection category; if yes, proceed to step S712; if no, proceed to step S713.

[0121] In this embodiment, the preset median distance threshold for tower structural defect detection categories can be manually set, and its value is used to determine the degree of deviation between the extracted tower anomaly location information and the corresponding category median. This can be achieved by comparing the median distance information for each tower structural defect detection category with the preset threshold one by one to determine whether the corresponding defect category classification is reasonable.

[0122] Step S712: Generate multiple tower structure defect detection information based on the multiple tower structure defect detection category information.

[0123] In this embodiment, when the median distance information of the tower structure defect detection category is less than the preset threshold, it indicates that the defect category division is reasonable. Then, the tower structure defect detection category information of each category, the corresponding abnormal location information, and the sensor measurement information are integrated to generate multiple tower structure defect detection information containing complete information such as defect category, location, and severity.

[0124] Step S713: The median information of the multiple tower structure defect detection categories is used as the multiple extracted tower anomaly location information, and the process is returned to step S704.

[0125] In this embodiment, when the median distance information of the tower structure defect detection category is not less than the preset threshold, it indicates that there is a deviation in the defect category classification. Then, the median information of multiple tower structure defect detection categories is used as multiple extracted tower anomaly location information, and the remaining anomaly location information is re-screened and the distance is recalculated, thereby achieving the optimization and adjustment of the defect category classification.

[0126] The tower structure defect detection method provided in this application accurately locates abnormal areas, achieves reasonable classification of defect categories, and forms a closed-loop defect classification optimization mechanism. This effectively improves the accuracy of tower structure defect category identification, making the generated multiple tower structure defect detection information more reliable and valuable for reference, and effectively assisting in the precise operation and maintenance of communication towers.

[0127] Corresponding to the methods in the embodiments above, Figure 8 shows a structural block diagram of the tower structure defect detection system provided in this application embodiment. For ease of explanation, only the parts related to the embodiments of this application are shown. The tower structure defect detection system illustrated in Figure 8 can be the execution subject of the tower structure defect detection method provided in the aforementioned embodiment one.

[0128] Referring to Figure 8, the tower structure defect detection system includes:

[0129] The information acquisition module 810 is used to acquire image information of multiple towers to be detected and sensor measurement information of multiple towers to be detected.

[0130] The tower defect virtual image information generation module 820 is used to generate multiple tower defect virtual image information.

[0131] The image feature information generation module 830 is used to perform feature extraction and analysis calculation based on the multiple tower images to be detected and the multiple tower defect virtual image information, according to the preset tower image feature extraction model, to generate multiple tower image feature information and multiple tower defect virtual image feature information.

[0132] The image registration information generation module 840 is used to perform registration processing based on the feature information of the multiple tower images to be detected and the feature information of the multiple tower defect virtual images to obtain the tower image registration information and the tower defect virtual image registration information.

[0133] The image fusion information generation module 850 for the tower to be detected is used to fuse the image registration information of the tower to be detected and the virtual image registration information of the tower defect based on the measurement information of the multiple tower sensors to be detected, so as to obtain the image fusion information of multiple towers to be detected.

[0134] The tower structure defect detection information generation module 860 is used to generate multiple tower structure defect detection information based on the measurement information of the multiple tower sensors and the image fusion information of the multiple towers to be detected.

[0135] The process of each module in the tower structure defect detection system provided in this application realizing its respective function can be referred to the description of Embodiment 1 shown in Figure 1 above, and will not be repeated here.

[0136] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0137] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0138] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0139] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0140] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only for distinguishing descriptions and should not be construed as indicating or implying relative importance. It should also be understood that although the terms "first," "second," etc., are used in the text to describe various elements in some embodiments of this application, these elements should not be limited by these terms. These terms are merely used to distinguish one element from another. For example, a first table may be named a second table, and similarly, a second table may be named a first table, without departing from the scope of the various described embodiments. Both the first table and the second table are tables, but they are not the same table.

[0141] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0142] The method for detecting defects in the tower structure provided in this application can be applied to terminal devices such as mobile phones, tablets, wearable devices, vehicle-mounted devices, augmented reality / virtual reality devices, and laptops. This application does not impose any restrictions on the specific type of terminal device.

[0143] Figure 9 is a schematic diagram of the structure of a terminal device provided in an embodiment of this application. As shown in Figure 9, the terminal device 9 of this embodiment includes: at least one processor 90 (only one is shown in Figure 9) and a memory 91, wherein the memory 91 stores a computer program 92 that can run on the processor 90. When the processor 90 executes the computer program 92, it implements the steps in the above-described embodiments of the tower structure defect detection method, such as steps S101 to S106 shown in Figure 1. Alternatively, when the processor 90 executes the computer program 92, it implements the functions of each module / unit in the above-described system embodiments, such as the functions of modules 810 to 860 shown in Figure 8.

[0144] The terminal device 9 can be a desktop computer, laptop, handheld computer, cloud server, or other computing device. The terminal device may include, but is not limited to, a processor 90 and a memory 91. Those skilled in the art will understand that Figure 9 is merely an example of the terminal device 9 and does not constitute a limitation on the terminal device 9. It may include more or fewer components than shown, or combine certain components, or use different components. For example, the terminal device may also include input / transmission devices, network access devices, buses, etc.

[0145] The processor 90 may be a central processing unit, or it may be other general-purpose processors, digital signal processors, application-specific integrated circuits, off-the-shelf programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0146] In some embodiments, the memory 91 may be an internal storage unit of the terminal device 9, such as a hard disk or memory of the terminal device 9. The memory 91 may also be an external storage device of the terminal device 9, such as a plug-in hard disk or smart memory card equipped on the terminal device 9. Furthermore, the memory 91 may include both internal and external storage units of the terminal device 9. The memory 91 is used to store operating systems, applications, bootloaders, data, and other programs, such as the program code of computer programs. The memory 91 can also be used to temporarily store data that has been sent or will be sent.

[0147] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0148] This application also provides a terminal device, which includes at least one memory, at least one processor, and a computer program stored in the at least one memory and executable on the at least one processor. When the processor executes the computer program, it causes the terminal device to implement the steps in any of the above method embodiments.

[0149] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0150] This application provides a computer program product that, when run on a terminal device, enables the terminal device to implement the steps described in the various method embodiments above.

[0151] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0152] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0153] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0154] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0155] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for detecting structural defects in iron towers, characterized in that, include: Acquire image information of multiple towers to be inspected and sensor measurement information of multiple towers to be inspected; Generate virtual image information of defects in multiple iron towers; Based on a preset tower image feature extraction model, feature extraction and analysis calculations are performed on multiple tower images to be detected and multiple virtual images of tower defects to generate multiple tower image feature information and multiple virtual image feature information of tower defects. Registration processing is then performed on these multiple tower image feature information and multiple virtual image feature information of tower defects to obtain tower image registration information and tower defect virtual image registration information. Based on the sensor measurement information of the multiple towers to be detected, the tower image registration information and tower defect virtual image registration information are fused to obtain multiple tower image fusion information. Finally, multiple tower structural defect detection information is generated based on the sensor measurement information of the multiple towers to be detected and the multiple tower image fusion information.

2. The method for detecting structural defects in iron towers as described in claim 1, characterized in that, The preset tower image feature extraction model includes a preset first-scale tower image feature extraction sub-model and a preset second-scale tower image feature extraction sub-model. The step of generating multiple tower image feature information and multiple tower defect virtual image feature information based on the preset tower image feature extraction model specifically includes: generating a tower pixel matrix and a tower defect virtual pixel matrix based on the multiple tower image information and multiple tower defect virtual image information; and further generating multiple tower image feature information and multiple tower defect virtual image feature information based on the preset first-scale tower image feature extraction sub-model and the preset second-scale tower image feature extraction sub-model. The two-scale tower image feature extraction sub-model analyzes and calculates the pixel matrix of the tower to be detected and the virtual pixel matrix of tower defects, generating first-scale tower pixel matrix variables, second-scale tower pixel matrix variables, first-scale tower defect virtual pixel matrix variables, and second-scale tower defect virtual pixel matrix variables. Based on the first-scale tower pixel matrix variables, the second-scale tower pixel matrix variables, the first-scale tower defect virtual pixel matrix variables, the second-scale tower defect virtual pixel matrix variables, and the preset tower feature extraction radius information, it generates multiple tower image feature information and multiple tower defect virtual image feature information.

3. The method for detecting defects in iron tower structures as described in claim 2, characterized in that, The step of generating multiple image feature information of towers to be detected and multiple virtual image feature information of tower defects based on the first-scale tower pixel matrix variable, the second-scale tower pixel matrix variable, the first-scale tower defect virtual pixel matrix variable, the second-scale tower defect virtual pixel matrix variable, and the preset tower feature extraction radius information specifically includes: performing a difference operation on the first-scale tower pixel matrix variable and the second-scale tower pixel matrix variable to obtain a scale-difference tower pixel matrix variable; performing a difference operation on the first-scale tower defect virtual pixel matrix variable and the second-scale tower pixel matrix variable to obtain a scale-difference tower pixel matrix variable; and performing a difference operation on the first-scale tower defect virtual pixel matrix variable and the second-scale tower pixel matrix variable to obtain a scale-difference tower pixel matrix variable. The virtual pixel matrix variables of the two-scale tower defects are subjected to a difference operation to obtain the scale-difference virtual pixel matrix variables of the tower defects; the extreme point coordinates of the scale-difference tower pixel matrix variables to be detected and the extreme point coordinates of the virtual pixel matrix variables of the tower defects are extracted to obtain the feature center information of the tower image to be detected and the feature center information of the virtual image of the tower defects; based on the preset tower feature extraction radius information, multiple tower image feature information and multiple virtual image feature information of the tower defects are generated with the feature center information of the tower image to be detected and the feature center information of the virtual image of the tower defects as the center.

4. The method for detecting structural defects in iron towers as described in claim 1, characterized in that, The step of performing registration processing based on the feature information of the multiple tower images to be detected and the feature information of the multiple virtual images of tower defects to obtain the registration information of the tower images to be detected and the registration information of the virtual images of tower defects specifically includes: performing neighborhood feature aggregation processing on the feature information of the multiple tower images to be detected and the feature information of the multiple virtual images of tower defects based on a preset tower image feature aggregation window to obtain the feature aggregation information of the multiple towers to be detected and the virtual feature aggregation information of the multiple tower defects; calculating the feature aggregation information of the multiple towers to be detected and the virtual feature aggregation information of the multiple tower defects. The cosine similarity of the combined information is used to obtain the aggregated similarity information of multiple tower image features; it is then determined whether the aggregated similarity information of the tower image features is greater than a preset tower feature registration similarity threshold; if so, spatial coordinates are calculated based on the aggregated feature information of the tower to be detected and the virtual feature information of the tower defects corresponding to the aggregated similarity information of the tower image features to obtain the registration information of the tower to be detected and the virtual image registration information of the tower defects; if not, the aggregated feature information of the tower to be detected and the virtual feature information of the tower defects corresponding to the aggregated similarity information of the tower image features to be detected are skipped.

5. The method for detecting structural defects in iron towers as described in claim 1, characterized in that, The step of fusing the image registration information of the tower to be detected and the virtual image registration information of the tower defects based on the measurement information of the multiple tower sensors to obtain the fused image information of multiple towers to be detected specifically includes: extracting the three-dimensional spatial coordinate range of the image registration information of the tower to be detected and the three-dimensional spatial coordinate range of the virtual image registration information of the tower defects to obtain the registration coordinate domain of the tower to be detected and the virtual registration coordinate domain of the tower defects; scaling and spatially mapping the virtual registration coordinate domain of the tower defects according to the registration coordinate domain of the tower to be detected to obtain the virtual registration mapping coordinate domain of the tower defects; and fusing the registration coordinate domain of the tower to be detected, the virtual registration mapping coordinate domain of the tower defects, and the measurement information of the multiple tower sensors to obtain the fused image information of multiple towers to be detected.

6. The method for detecting structural defects in iron towers as described in claim 5, characterized in that, The sensor measurement information of the tower to be tested includes the measurement information of the tower stress sensor, the tower tilt angle sensor, the tower displacement sensor, and the tower ambient temperature sensor. The step of fusing the measurement information of the tower registration coordinate domain, the tower defect virtual registration mapping coordinate domain, and the multiple tower sensor measurements to obtain multiple tower image fusion information specifically includes: generating the tower structure deformation matrix information based on the multiple tower stress sensor measurement information, the multiple tower tilt angle sensor measurement information, and the multiple tower displacement sensor measurement information; performing fusion calculation based on the tower registration coordinate domain, the tower defect virtual registration mapping coordinate domain, and the tower structure deformation matrix information to obtain the tower structure deformation detection information; and performing error compensation calculation on the tower structure deformation detection information based on the multiple tower ambient temperature sensor measurements to obtain the multiple tower image fusion information.

7. The method for detecting structural defects in iron towers as described in claim 1, characterized in that, The step of generating multiple tower structural defect detection information based on the measurement information of the multiple tower sensors and the image fusion information of the multiple towers to be detected specifically includes: obtaining multiple tower sensor measurement anomaly information based on the measurement information of the multiple tower sensors and multiple preset tower sensor measurement threshold information; performing location calculation on the image fusion information of the multiple towers to be detected based on the multiple tower sensor measurement anomaly information to obtain multiple tower anomaly location information; randomly sampling the multiple tower anomaly location information based on preset tower structural defect type and quantity information to obtain multiple sampled tower anomaly location information; obtaining multiple remaining tower anomaly location information based on the multiple tower anomaly location information and the multiple sampled tower anomaly location information; calculating the Euclidean distance between the multiple sampled tower anomaly location information and the multiple remaining tower anomaly location information to obtain multiple tower anomaly location distance information; determining whether the tower anomaly location distance information is less than or equal to the preset tower anomaly location distance threshold information; if so, then based on the tower... The abnormal positioning distance information of the extracted towers and the abnormal positioning information of the remaining towers are used to generate tower structural defect detection category information. If not, tower structural defect detection category information is generated based on the abnormal positioning distance information of the extracted towers. The median of the multiple tower structural defect detection category information is calculated to obtain the median information of multiple tower structural defect detection categories. The Euclidean distance between the median information of multiple tower structural defect detection categories and the multiple extracted tower abnormal positioning information is calculated to obtain the median distance information of multiple tower structural defect detection categories. It is determined whether the median distance information of the tower structural defect detection categories is less than a preset median distance threshold for tower structural defect detection categories. If yes, multiple tower structural defect detection information is generated based on the multiple tower structural defect detection category information. If not, the median information of the multiple tower structural defect detection categories is used as multiple extracted tower abnormal positioning information, and the process returns to the step of obtaining multiple remaining tower abnormal positioning information based on the multiple tower abnormal positioning information to be detected and the multiple extracted tower abnormal positioning information.

8. A system for detecting structural defects in iron towers, characterized in that, include: The information acquisition module is used to acquire image information of multiple towers to be detected and sensor measurement information of multiple towers to be detected. The tower defect virtual image information generation module is used to generate multiple tower defect virtual image information. The image feature information generation module is used to perform feature extraction and analysis calculations based on the multiple tower images to be detected and the multiple virtual images of tower defects, according to a preset tower image feature extraction model, to generate multiple tower image feature information and multiple virtual image feature information of tower defects; the image registration information generation module is used to perform registration processing based on the multiple tower image feature information and the multiple virtual image feature information of tower defects, to obtain tower image registration information and virtual image registration information of tower defects. The image fusion information generation module for the tower to be detected is used to fuse the image registration information of the tower to be detected and the virtual image registration information of the tower defects based on the measurement information of the multiple tower sensors to be detected, so as to obtain the image fusion information of multiple towers to be detected. The tower structure defect detection information generation module is used to generate multiple tower structure defect detection information based on the measurement information of the multiple tower sensors and the image fusion information of the multiple towers to be detected.

9. A terminal device, characterized in that, The terminal device includes a memory and a processor. The memory stores a computer program that can run on the processor. When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7.

Images