Bridge cable state evaluation method based on unsupervised single classification analysis

Abstract

The application relates to the technical field of bridge health monitoring, in particular to a bridge cable state evaluation method based on unsupervised single classification analysis. Bridge data of multiple target bridges is collected, all the collected bridge data is preprocessed to obtain multiple preprocessing data, an evaluation model is created, and the preprocessing data is input into the evaluation model; the evaluation model extracts image feature vectors of the target bridges based on the preprocessing data, a single classification model is trained based on the image feature vectors, a trained evaluation model is obtained, and finally, real-time data of a real-time bridge is collected; the real-time data is input into the trained evaluation model, and the trained evaluation model is used to judge the cable state of the real-time bridge. The HOG feature vectors of the target bridges are extracted, single classification analysis learning is carried out based on the HOG feature vectors, the trained evaluation model can quickly judge the cable state according to the real-time data of the real-time bridge, and the detection efficiency of the cable state is improved.

Description

Technical Field

[0001] This invention relates to the field of bridge health monitoring technology, and more specifically, to a method for assessing the condition of bridge cables based on unsupervised single-classification analysis. Background Technology

[0002] The cable system of long-span bridges is the main load-bearing structure of suspension bridges, cable-stayed bridges, and arch-suspended bridges. Because they span rivers, lakes, and seas and bear dynamic and static loads, they are constantly exposed to wind, rain, humidity, and polluted air, making them susceptible to stress corrosion, fatigue corrosion, and metal corrosion. The cables and anchorages are all prone to corrosion damage. Therefore, to ensure the safety of the bridge and cable structure, corrosion protection for the cables is essential.

[0003] Chinese Patent No. CN119043520B discloses a method and apparatus for identifying the health status of a cable. This method involves acquiring cable temperature data and ambient temperature data; determining whether the cable body is damaged based on the cable temperature data; determining the temperature response coefficient and response hysteresis coefficient of the cable based on the cable temperature data and the ambient temperature data; and determining whether the cable clamp is loose based on the temperature response coefficient and the response hysteresis coefficient. However, existing technologies often require consideration of multiple cable states, leading to low efficiency in determining abnormal states and a high risk of errors. Summary of the Invention

[0004] The purpose of this invention is to address the problems existing in the background technology by proposing a bridge cable condition assessment method based on unsupervised single classification analysis.

[0005] The technical solution of this invention:

[0006] On the one hand, this application provides a method for assessing the condition of bridge cables based on unsupervised single classification analysis, including:

[0007] Bridge data from multiple target bridges is collected, and all collected bridge data is preprocessed to obtain multiple preprocessed data sets.

[0008] Create an evaluation model;

[0009] The preprocessed data is input into the evaluation model. The evaluation model extracts the image feature vector of the target bridge based on the preprocessed data, and trains a single classification model based on the image feature vector to obtain the trained evaluation model.

[0010] Real-time data of the bridge is collected and input into a trained evaluation model. The trained evaluation model is then used to determine the cable status of the bridge in real time.

[0011] Preferably, bridge data from multiple target bridges is collected, and all collected bridge data is preprocessed to obtain multiple preprocessed data sets, including:

[0012] Create a bridge database;

[0013] For multiple target bridges, the acquisition parameters for each target bridge are set separately;

[0014] Data is collected for each target bridge based on the acquisition parameters, resulting in multiple bridge data and image data. All collected data is then placed into the bridge database. The bridge data includes cable tension and cable dynamic characteristics.

[0015] Randomly select an image from the bridge database;

[0016] The image data is denoised to obtain a preprocessed image;

[0017] Returns a randomly selected image from the bridge database to obtain multiple preprocessed data.

[0018] Preferably, the preprocessed data is input into the evaluation model, which extracts image feature vectors of the target bridge based on the preprocessed data. A single-classification model is then trained based on these image feature vectors to obtain the trained evaluation model, which includes:

[0019] All preprocessed data are divided into training and test sets according to a random ratio;

[0020] The training set is input into the evaluation model, which extracts the image feature vector of each cable image and constructs a feature matrix based on the image feature vector. The feature matrix is ​​then used to train a single-classification model, thus obtaining the trained evaluation model. The trained evaluation model has the ability to automatically extract image features from the input cable images and determine the cable status of the target bridge based on the image features.

[0021] The test set is input into the trained evaluation model to determine whether the trained evaluation model has completed training.

[0022] Preferably, the training set is input into the evaluation model, the evaluation model extracts the image feature vector of each cable image, and a feature matrix is ​​constructed based on the image feature vector. A single-classification model is then trained using the feature matrix to obtain the trained evaluation model, including:

[0023] Randomly select a cable image from the training set;

[0024] The cable image is converted to grayscale to obtain a grayscale image of the cable.

[0025] Extracting HOG feature vectors from cable grayscale images;

[0026] Return a cable image randomly selected from the training set until all cable images in the training set have been selected, and obtain the image feature vector of each cable image;

[0027] A feature matrix is ​​established based on image feature vectors; each row of the feature matrix is ​​an image feature vector of a cable image.

[0028] Preferably, the HOG feature vector extraction based on the cable grayscale image includes:

[0029] Set the size standard and set the cable grayscale image size to the standard size to obtain the standard grayscale image;

[0030] Obtain the directional gradient number and divide the standard grayscale image into multiple cells;

[0031] Multiple cells are combined into a cell block, and the cell block is normalized to obtain a normalized image;

[0032] Slide the normalized image and calculate the HOG features at each position in the normalized image to obtain multiple HOG features;

[0033] Multiple HOG features are concatenated to obtain the HOG feature vector.

[0034] Preferably, real-time data of the bridge is collected and input into a trained evaluation model. The trained evaluation model is then used to determine the cable status of the real-time bridge, including:

[0035] Collect real-time data of the bridge; the real-time data includes real-time bridge data and real-time image data;

[0036] Real-time data is input into the trained evaluation model. The trained evaluation model performs cluster analysis on the real-time bridges based on the bridge data to obtain the target bridges corresponding to the real-time bridges. The target bridges corresponding to the real-time bridges are recorded as real-time targets.

[0037] Extract the real-time feature vector of the bridge and determine the cable status of the bridge based on the real-time feature vector and the feature matrix of the real-time target.

[0038] Preferably, real-time data is input into a trained evaluation model, which then performs cluster analysis on the real-time bridges based on the bridge data to obtain the target bridges corresponding to the real-time bridges. The target bridges corresponding to the real-time bridges are denoted as real-time targets, including:

[0039] K target bridges are randomly selected and recorded as the initial cluster centers. The remaining target bridges and real-time bridges are randomly assigned to the K initial cluster centers to form K clusters.

[0040] Set an iteration count threshold;

[0041] For each cluster, calculate the distance from each point within the cluster to the cluster center, and record the point corresponding to the average distance as the new cluster center;

[0042] Determine if the number of iterations is greater than or equal to the iteration count threshold;

[0043] If the number of iterations is greater than or equal to the iteration threshold, then the target bridge corresponding to the cluster center obtained in the last iteration is recorded as the real-time target.

[0044] Preferably, the real-time feature vector of the bridge is extracted, and the cable status of the bridge is determined based on the real-time feature vector and the feature matrix of the real-time target, including:

[0045] Set the difference threshold;

[0046] The trained evaluation model obtains the real-time feature vector of the bridge based on real-time image data, and calculates the difference between the real-time feature vector and the feature matrix of the real-time target.

[0047] Determine whether the difference between the real-time feature vector and the feature matrix of the real-time target is greater than or equal to the difference threshold;

[0048] If the difference between the real-time feature vector and the feature matrix of the real-time target is greater than or equal to the difference threshold, then the cable status of the real-time bridge is judged to be abnormal.

[0049] On the other hand, this application also provides a bridge cable condition assessment system based on unsupervised single classification analysis, applied to the bridge cable condition assessment method based on unsupervised single classification analysis as described in any one of the foregoing, including a data acquisition component and an assessment component. The data acquisition component acquires bridge data of the target bridge, and the assessment component is communicatively connected to the data acquisition component. The assessment component extracts image features of the target bridge based on the bridge data and determines the cable condition of the template bridge in combination with the image features.

[0050] Preferably, the acquisition component includes a data acquisition module and an image acquisition module. The data acquisition module acquires bridge data of the target bridge, and the image acquisition module acquires bridge images of the target bridge.

[0051] Compared with the prior art, the above-mentioned technical solution of the present invention has the following beneficial technical effects:

[0052] By collecting bridge data from multiple target bridges, preprocessing all collected bridge data to obtain multiple preprocessed data sets, and then creating an evaluation model, the preprocessed data is input into the evaluation model. The evaluation model extracts image feature vectors of the target bridges based on the preprocessed data, and performs single-classification model training based on the image feature vectors to obtain the trained evaluation model. Finally, real-time data of the bridges is collected and input into the trained evaluation model. The trained evaluation model determines the cable status of the real-time bridges. This application improves the detection efficiency of cable status by extracting HOG feature vectors of the target bridges and performing single-classification analysis and learning based on HOG feature vectors, thereby enabling the trained evaluation model to quickly determine the cable status based on the real-time data of the bridges. Attached Figure Description

[0053] Figure 1 This is a flowchart illustrating a bridge cable condition assessment method based on unsupervised single-classification analysis proposed in this invention.

[0054] Figure 2 This is a schematic diagram of a bridge cable condition assessment system based on unsupervised single-classification analysis proposed in this invention.

[0055] Figure label:

[0056] 100. Acquisition Components; 101. Data Acquisition Module; 102. Image Acquisition Module; 200. Evaluation Components. Detailed Implementation

[0057] Example 1, as Figure 1 As shown, the present invention proposes a bridge cable condition assessment method based on unsupervised single-classification analysis, comprising:

[0058] S100: Collect bridge data from multiple target bridges, preprocess all collected bridge data to obtain multiple preprocessed data.

[0059] Specifically, the bridge data collected in this application are all data under normal working conditions of the bridge cables, so as to facilitate the subsequent single-classification analysis of the evaluation model;

[0060] S200, Create an evaluation model;

[0061] S300: Input the preprocessed data into the evaluation model. The evaluation model extracts the image feature vector of the target bridge based on the preprocessed data and trains a single-classification model based on the image feature vector to obtain the trained evaluation model.

[0062] The S400 collects real-time data of the bridge and inputs it into a trained evaluation model to determine the cable status of the bridge.

[0063] In this invention, bridge data from multiple target bridges is collected, and all collected bridge data is preprocessed to obtain multiple preprocessed data sets. An evaluation model is then created, and the preprocessed data is input into the evaluation model. The evaluation model extracts image feature vectors of the target bridges based on the preprocessed data, and a single-classification model is trained based on these image feature vectors to obtain the trained evaluation model. Finally, real-time data from the bridges is collected and input into the trained evaluation model. The trained evaluation model then determines the cable status of the real-time bridges. This application improves the efficiency of cable status detection by extracting HOG feature vectors from the target bridges and performing single-classification analysis based on these HOG feature vectors.

[0064] In an optional embodiment, S100 includes:

[0065] S110, Create the bridge database;

[0066] S120, for multiple target bridges, set the acquisition parameters for each target bridge separately;

[0067] S130, Data is collected for each target bridge based on the acquisition parameters, resulting in multiple bridge data and image data, and all collected data is placed into the bridge database; the bridge data includes cable tension and cable dynamic characteristics.

[0068] Specifically, the dynamic characteristics of the cable include natural frequency, amplitude, vibration mode, and damping ratio. The reason for collecting dynamic characteristic data is that the cable will vibrate under the action of wind and traffic flow, which will cause fatigue damage to the cable. Moreover, the dynamic characteristics can also reflect the changes in cable force or structural connection state.

[0069] S140, randomly select an image data from the bridge database;

[0070] S150, Denoise the image data to obtain a preprocessed image;

[0071] S160 returns a randomly selected image from the bridge database, resulting in multiple preprocessed data.

[0072] It should be noted that, in the daily monitoring of bridges, regularly inspecting the condition of bridge cables is one of the important means to ensure bridge safety. Therefore, it is necessary to collect bridge data of the target bridge and the corresponding image data at the same time for each bridge data point, so as to complete the assessment of the condition of the bridge cables based on this data.

[0073] After the data collection is completed, in order to ensure the integrity and reliability of the data, the image data is denoised to improve the quality of the image data. The reason for using image data as training samples is that image data is a non-contact detection method that can be applied to high-altitude detection, thereby improving the applicability of this application. In addition, image data can also detect early defects that are difficult for sensors to capture, and promptly detect abnormal cable conditions.

[0074] In an optional embodiment, S300 includes:

[0075] S310, divide all preprocessed data into training and test sets according to a random ratio;

[0076] S320: The training set is input into the evaluation model. The evaluation model extracts the image feature vector of each cable image and constructs a feature matrix based on the image feature vector. The single-classification model is trained in combination with the feature matrix to obtain the trained evaluation model. The trained evaluation model has the ability to automatically extract image features from the input cable images and determine the cable status of the target bridge based on the image features.

[0077] S330 inputs the test set into the trained evaluation model and determines whether the trained evaluation model has completed training.

[0078] It should be noted that when dividing the training set and the test set, the proportion of the training set should be greater than that of the test set to increase the number of training samples in the training set. When verifying whether the trained evaluation model has been successfully trained using the test set, the output accuracy and / or response time of the trained evaluation model can be used as the judgment criteria.

[0079] By sequentially inputting training samples from the training set into the evaluation model, the evaluation model completes the evaluation of the training sample data based on the single classification analysis method. Essentially, it is a data-driven approach to identify anomalies and assess the health of cables. Single classification analysis is a method that builds a model that can describe the "normal state" based on a large number of "normal" samples, and then determines whether a new sample belongs to the normal category. If it does not belong, it is considered to be an abnormal or unknown state. Since single classification analysis is only trained on normal sample data, it does not need to collect abnormal samples, making it suitable for anomaly detection tasks.

[0080] Since bridge cables are in a normal state for most of the time during use, it is difficult to collect data on abnormal states of bridge cables, and the sample size is small. Therefore, a single-classification analysis method is used to train the evaluation model, thereby constructing a benchmark model of cable health status based on normal state data. Finally, by collecting real-time cable data, abnormal states that deviate from the normal state can be identified.

[0081] In an optional embodiment, S320 includes:

[0082] S321, Randomly select a cable image from the training set;

[0083] S322, perform grayscale processing on the cable image to obtain a cable grayscale image;

[0084] S323, Extracting HOG feature vectors from cable grayscale images;

[0085] Specifically, HOG stands for Histogram of Oriented Gradients, which is an image feature description method. This method divides the image into cell units to extract the gradient direction distribution and enhances the adaptability to changes in illumination through block-level normalization.

[0086] S324, return a cable image randomly selected from the training set until all cable images in the training set have been selected, and obtain the image feature vector of each cable image;

[0087] S325, establish a feature matrix based on image feature vectors; each row of the feature matrix is ​​an image feature vector of a cable image.

[0088] It should be noted that by collecting the HOG feature vector of each image data, the characteristics of the bridge cable corresponding to the image data can be reflected through the HOG feature vector. Since the image data are all images of normal bridge cables, each collected HOG feature vector can also represent that the bridge cable is in a normal state. Therefore, when the difference between the HOG feature vector and the feature matrix of the image data of the target bridge is too large, it means that the cable state of the target bridge is abnormal, thereby improving the efficiency of detecting the cable state of the target bridge.

[0089] In an optional embodiment, S323 includes:

[0090] S3231, Set the size standard and set the cable grayscale image size to the standard size to obtain the standard grayscale image;

[0091] S3232, obtains the directional gradient number and divides the standard grayscale image into multiple cells;

[0092] S3233 combines multiple cells into a cell block and normalizes the cell block to obtain a normalized image;

[0093] S3234, slide the normalized image, calculate the HOG features at each position of the normalized image, and obtain multiple HOG features;

[0094] S3235 concatenates multiple HOG features to obtain the HOG feature vector.

[0095] It should be noted that when extracting HOG features from an image, the reason for setting the cable grayscale image to the same standard size is to ensure that the number of cell blocks is consistent in the subsequent division. Before dividing the cell, the gradient of the standard grayscale image in the X and Y directions needs to be calculated using the Sobel operator, and the gradient magnitude and direction of each pixel need to be calculated in order to divide the cell.

[0096] In an optional embodiment, S400 includes:

[0097] S410, Collect real-time data of the bridge; the real-time data includes real-time bridge data and real-time image data;

[0098] S420: Input real-time data into the trained evaluation model. The trained evaluation model performs cluster analysis on the real-time bridges based on the real-time bridge data to obtain the target bridges corresponding to the real-time bridges. The target bridges corresponding to the real-time bridges are recorded as real-time targets.

[0099] S430 extracts the real-time feature vector of the bridge and determines the cable status of the bridge based on the real-time feature vector and the feature matrix of the real-time target.

[0100] It should be noted that by feeding the real-time data of the real-time bridge into the trained evaluation model, the trained evaluation model can then filter out the real-time target that is closest to the real-time bridge from the database based on the real-time bridge data. Since the database stores target bridges in normal state, the real-time target is the target bridge in normal state that is closest to the real-time bridge. Therefore, if the real-time feature vector of the real-time bridge differs significantly from the feature vector of the target bridge in normal state that is closest to it, the cable state of the real-time bridge can be considered abnormal.

[0101] In an optional embodiment, S420 includes:

[0102] S421, randomly select K target bridges, record the selected K target bridges as the initial cluster centers, and randomly assign the remaining target bridges and real-time bridges to the K initial cluster centers to form K clusters;

[0103] S422, Set the threshold for the number of iterations;

[0104] S423, For each cluster, calculate the distance from each point in the cluster to the cluster center, and record the point corresponding to the average distance as the new cluster center;

[0105] S424, determine whether the number of iterations is greater than or equal to the iteration number threshold;

[0106] S425, if the number of iterations is greater than or equal to the iteration number threshold, then the target bridge corresponding to the cluster center obtained in the last iteration is recorded as the real-time target;

[0107] Specifically, if the number of iterations is less than the iteration threshold, return to step S423 until the number of iterations is greater than or equal to the iteration threshold.

[0108] It should be noted that this application uses the K-means clustering method to cluster real-time bridges, thereby selecting the real-time targets that are closest to the real-time bridges. Then, the cable status of the real-time bridges can be determined based on the feature vectors of the real-time targets, thus improving the detection efficiency of the cable status of the target bridges.

[0109] K-means clustering is a widely used clustering algorithm, primarily used to divide data into K predefined clusters. This method iteratively assigns data points to the nearest centroids (i.e., "centroids") and updates the positions of these centroids until a stopping condition is met.

[0110] In an optional embodiment, S430 includes:

[0111] S431, Set the difference threshold;

[0112] S432, the real-time feature vector of the bridge is obtained based on real-time image data through the trained evaluation model, and the difference between the real-time feature vector and the feature matrix of the real-time target is calculated.

[0113] S433, determine whether the difference between the real-time feature vector and the feature matrix of the real-time target is greater than or equal to the difference threshold;

[0114] S434, if the difference between the real-time feature vector and the feature matrix of the real-time target is greater than or equal to the difference threshold, then the cable status of the real-time bridge is judged to be abnormal.

[0115] It should be noted that by inputting real-time image data of the real-time bridge into the trained evaluation model, the real-time feature vector of the real-time bridge can be obtained. Then, the difference between the feature matrix of the real-time target and the real-time feature vector of the real-time bridge is calculated. If the difference is too large, it means that the cable status of the real-time bridge is abnormal. Therefore, the difference threshold is actually a descriptive value for judging whether the data difference between the real-time bridge and the target bridge is normal. So, the larger the difference threshold is set, the more lenient the judgment range of the cable status of the real-time bridge is, and the smaller the difference threshold is set, the more precise the judgment range of the cable status of the real-time bridge is.

[0116] like Figure 2As shown, this application also provides a bridge cable condition assessment system based on unsupervised single classification analysis, applied to the bridge cable condition assessment method based on unsupervised single classification analysis as described in any of the preceding claims. The system includes a data acquisition component and an assessment component. The data acquisition component acquires bridge data of the target bridge. The assessment component is communicatively connected to the data acquisition component. The assessment component extracts image features of the target bridge based on the bridge data and determines the cable condition of the template bridge by combining the image features.

[0117] It should be noted that the data acquisition component collects bridge data and image data of the target bridge, and all the acquired data is input into the evaluation component. The evaluation component combines the acquired data to determine the cable status of the target bridge, thereby promptly identifying problematic cables.

[0118] In an optional embodiment, the acquisition component includes a data acquisition module and an image acquisition module. The data acquisition module acquires bridge data of the target bridge, and the image acquisition module acquires bridge images of the target bridge.

[0119] It should be noted that the data acquisition module collects bridge data of the target bridge. The data acquisition module includes various types of sensors to collect various types of bridge data. The image acquisition module collects bridge images of the target bridge. The image acquisition module can be set as an infrared camera.

[0120] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited thereto. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention.

Claims

1. A bridge cable condition assessment method based on unsupervised single classification analysis, characterized in that, The method comprises the following steps: collecting bridge data of a plurality of target bridges, preprocessing all collected bridge data to obtain a plurality of preprocessed data; creating an evaluation model; inputting the preprocessed data into the evaluation model, extracting an image feature vector of the target bridge based on the preprocessed data through the evaluation model, and training a single classification model based on the image feature vector to obtain a trained evaluation model; collecting real-time data of a real-time bridge, inputting the real-time data into the trained evaluation model, and determining the cable state of the real-time bridge through the trained evaluation model; the step of inputting the preprocessed data into the evaluation model, extracting the image feature vector of the target bridge based on the preprocessed data through the evaluation model, and training the single classification model based on the image feature vector to obtain the trained evaluation model comprises the following steps: dividing all preprocessed data into a training set and a test set according to a random ratio; inputting the training set into the evaluation model, extracting an image feature vector of each cable image through the evaluation model, constructing a feature matrix based on the image feature vector, training a single classification model combined with the feature matrix, and obtaining a trained evaluation model; the trained evaluation model has the ability to automatically extract image features from the input cable image and determine the cable state of the target bridge based on the image features; inputting the test set into the trained evaluation model to determine whether the trained evaluation model is trained; the step of collecting real-time data of a real-time bridge, inputting the real-time data into the trained evaluation model, and determining the cable state of the real-time bridge through the trained evaluation model comprises the following steps: collecting real-time data of a real-time bridge; the real-time data comprises real-time bridge data and real-time image data; inputting the real-time data into the trained evaluation model, performing cluster analysis on the real-time bridge based on the bridge data through the trained evaluation model to obtain a target bridge corresponding to the real-time bridge; the target bridge corresponding to the real-time bridge is recorded as a real-time target; extracting a real-time feature vector of the real-time bridge, and determining the cable state of the real-time bridge based on the real-time feature vector and a feature matrix of the real-time target; the step of extracting a real-time feature vector of the real-time bridge, and determining the cable state of the real-time bridge based on the real-time feature vector and a feature matrix of the real-time target comprises the following steps: setting a difference threshold; obtaining a real-time feature vector of the real-time bridge based on the real-time image data through the trained evaluation model, and calculating a difference between the real-time feature vector and the feature matrix of the real-time target; determining whether the difference between the real-time feature vector and the feature matrix of the real-time target is greater than or equal to the difference threshold; if the difference between the real-time feature vector and the feature matrix of the real-time target is greater than or equal to the difference threshold, determining that the cable state of the real-time bridge is abnormal. The method comprises the following steps: collecting bridge data of a plurality of target bridges, preprocessing all collected bridge data to obtain a plurality of preprocessed data, comprising: creating a bridge database; setting collection parameters for each target bridge; collecting data of each target bridge based on the collection parameters to obtain a plurality of bridge data and image data, and ​ ​ 2. The bridge cable condition assessment method based on unsupervised single classification analysis according to claim 1, ​ ​ ​ ​ ​ Put all the data collected into the bridge database; the bridge data includes cable force and cable dynamic characteristics; Randomly select an image data from the bridge database; Denoise the image data to obtain a preprocessed image; Return to randomly select an image data from the bridge database to obtain a plurality of preprocessed data.

3. The bridge cable condition assessment method based on unsupervised single classification analysis according to claim 2, The method is characterized in that the training set is input into the evaluation model, the image feature vector of each cable image is extracted through the evaluation model, the feature matrix is constructed based on the image feature vector, the single classification model training is combined with the feature matrix, and thus the trained evaluation model is obtained, including: Randomly select a cable image from the training set; Grayscale processing is performed on the cable image to obtain a cable grayscale image; Based on the cable grayscale image, the image HOG feature vector is extracted; Return to randomly select a cable image from the training set until all cable images in the training set are selected to obtain the image feature vector of each cable image; Based on the image feature vector, a feature matrix is established; each row of the feature matrix is an image feature vector of a cable image. Based on the cable grayscale image, the image HOG feature vector is extracted, including: Set the size standard, and set the size of the cable grayscale image to the standard size to obtain a standard grayscale image; 4. The bridge cable condition assessment method based on unsupervised single classification analysis according to claim 3, characterized in that, Obtain the direction gradient number, and divide the standard grayscale image into a plurality of cells; Combine the plurality of cells into a unit block, and perform normalization processing on the unit block to obtain a normalized image; Slide the normalized image to calculate the HOG feature of each position of the normalized image to obtain a plurality of HOG features; Serially connect the plurality of HOG features to obtain a HOG feature vector. The method is characterized in that the real-time data is input into the trained evaluation model, the real-time bridge is analyzed based on the bridge data through the trained evaluation model, and the target bridge corresponding to the real-time bridge is obtained; 5. The bridge cable condition assessment method based on unsupervised single classification analysis according to claim 4, The target bridge corresponding to the real-time bridge is recorded as a real-time target, including: Randomly select K target bridges, record the selected K target bridges as initial cluster centers, and randomly allocate the remaining target bridges and the real-time bridge to the K initial cluster centers to form K clusters; Set an iteration number threshold; For each cluster, the distance from each point in the cluster to the cluster center is calculated, and the point corresponding to the average distance is recorded as a new cluster center; Determine whether the iteration number is greater than or equal to the iteration number threshold; If the iteration number is greater than or equal to the iteration number threshold, the target bridge corresponding to the cluster center obtained in the last iteration is recorded as the real-time target. Including: The acquisition component acquires the bridge data of the target bridge through the acquisition component; The evaluation component is in communication connection with the acquisition component, extracts the image features of the target bridge based on the bridge data through the evaluation component, and judges the cable state of the template bridge in combination with the image features. The acquisition component includes a data acquisition module and an image acquisition module, which acquires the bridge data of the target bridge through the data acquisition module and acquires the bridge image of the target bridge through the image acquisition module. ​ ​ ​ 6. A bridge cable condition assessment system based on unsupervised single classification analysis, which executes the bridge cable condition assessment method based on unsupervised single classification analysis according to any one of claims 1 to 5, characterized by, ​ ​ ​ ​ 7. The bridge cable condition assessment system based on unsupervised single classification analysis according to claim 6, wherein, ​

Images