Bridge joint height difference recognition method and system based on image and vibration data analysis

Abstract

The present invention discloses a bridge joint height difference identification method and system based on image and vibration data analysis. Compared with the existing daily inspection and on-foot detection methods, the patent is more objective in evaluating the height difference of expansion joints. The expansion joints can be quickly evaluated in daily inspections, and the height difference problems of expansion joints can be discovered in time, helping maintenance units to discover expansion joints that may be damaged this morning. Relying on high-frequency inspections and high-precision positioning, the height difference estimation of expansion joints can be tracked for a long time, and it can be evaluated whether the height difference will reappear after the expansion joints are repaired, and whether it will cause damage again, so as to facilitate maintenance personnel to evaluate whether the joint repair measures are reasonable. In addition, the system mentioned in this patent does not require personnel to perform special operations, and inspectors only need to focus on driving. Compared with manual on-foot inspections, the safety risks of inspectors are avoided.

Description

Technical Field

[0001] The present invention relates to the technical field related to road anomaly detection, and in particular to a method and system for identifying height differences of bridge joints based on image and vibration data analysis. Background Art

[0002] Expansion joints are important ancillary facilities of roads and bridges. They are set at the weak parts of the beam ends and directly bear the repeated impacts of wheel loads. They are also exposed to nature for a long time in a relatively harsh environment. Therefore, expansion joints are prone to damage. Usually, due to the inconsistent deformation of the supports on both sides and the uneven settlement of the foundations on both sides of the joints, it is easy to cause a height difference on both sides of the expansion joints, causing vehicles to jump when passing by, affecting driving safety; and the wheels will also bring a large impact load to the joints, causing further deterioration of the joints, and further expansion of the height difference on both sides of the joints, and even large potholes may appear, seriously affecting driving safety; after many expansion joints were obviously damaged, the maintenance personnel only made emergency repairs to fill the potholes, but did not repair the height difference on both sides. After the expansion joints were repaired, because the height difference on both sides still existed, under the influence of vehicle loads and the environment, they soon deteriorated again.

[0003] In the routine inspection of the existing technology, the inspection method mainly relies on manual observation. The inspectors inspect the road conditions in inspection vehicles. The speed of the vehicles is relatively fast during the inspection. The inspectors cannot directly measure the height difference of the expansion joints and can only make corresponding records and repairs for the more serious parts. Therefore, the current bridge joint height difference basically relies on annual special inspections, which are carried out by manual on-foot measurements.

[0004] Manual walking inspection mainly uses tools such as a three-meter ruler to measure the height difference of bridge expansion joints. The measurement speed is slow and the efficiency is low. In order to ensure the safety of the surveyors, the traffic may need to be closed or semi-closed during the measurement operation, which affects the road traffic. In addition, the frequency of walking measurement is low, and the height difference problem at the joint cannot be discovered in time. In addition, since walking measurement relies on a lot of manpower, the measurement cost is high, and large-scale and high-frequency detection cannot be achieved. In remote areas and many low-grade roads, the detection of expansion joint height difference is often ignored. Under the influence of the environment and impact loads, the expansion joints are prone to damage. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for identifying height differences of bridge joints based on image and vibration data analysis to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The bridge joint height difference recognition method based on image and vibration data analysis includes the following steps:

[0008] Collect and generate image information at preset time intervals, and identify and determine the image information through a preset neural network algorithm, and generate a determination result, wherein the identification and determination is used to determine whether a vehicle passes through an expansion joint of a bridge, and when the vehicle passes through the expansion joint, a determination result is generated;

[0009] In response to the determination result, the start time t of the vehicle passing through the expansion joint is recorded. i-strat and the end time t i-end , and obtain the three-axis sensing data of the corresponding time period, the three-axis sensing data includes the z-axis acceleration data acc i ;

[0010] According to the starting time t i-start and the end time t i-end The three-axis sensor data is preprocessed to calculate the vehicle vibration ratio R i , and the vehicle vibration ratio R is determined according to a preset determination threshold. i If the value exceeds the limit, it is determined that the expansion joint has an abnormal height difference and needs to be repaired;

[0011] The relevant data of the abnormal height difference is stored and synchronized through a cloud server.

[0012] As a further solution of the present invention: i-start and the end time t i-end The three-axis sensor data is preprocessed to calculate the vehicle vibration ratio R i The steps include:

[0013] For the starting time t i-strat To the end time t i-end All the z-axis acceleration data acc within the time period i Perform zero mean and get acc i ';

[0014] Calculate the acc i 'Get the average root mean square and obtain rms i , the rms i To characterize the starting time t i-strat To the end time t i-end Average vibration of the vehicle during the process;

[0015] The acci' is processed by a preset window function to obtain an array rms containing ceil((n-w+1) / s) numbers i_list, wherein w is the size of the window, the size of the window w≤50, s is the step size, the step size s<5, and ceil is a rounding function;

[0016] Take the rms i_list The maximum value rms i_w_max , and calculate the vehicle vibration ratio R i =rms i_w_max / rms i , the rms i_w_max To characterize the starting time t i-strat To the end time t i-end The maximum short-term vibration of the vehicle within the time period;

[0017] According to the preset threshold R threshold The vehicle vibration ratio R i Make a judgment, if it exceeds the threshold R threshold , it is determined that there is an abnormal height difference at the expansion joint that needs to be repaired.

[0018] As a further solution of the present invention: the number of sensors corresponding to the three-axis sensor data is a pair, which are respectively used to generate R i_left , R i_right , when the R i_left With the R i_right Any of the above threshold values ​​R threshold When , it is determined that there is an abnormal height difference at the expansion joint that needs to be repaired, and the R i_left With the R i_right They are used to characterize the vibration and bumps experienced by the left and right wheels of a car.

[0019] As a further solution of the present invention: the specific content of the relevant data includes: the starting time t corresponding to the expansion joint i-strat , the end time t i-end , as well as the longitude and latitude information of the vehicle corresponding to the time, the image information collected corresponding to the time, the vehicle's driving speed and heading angle.

[0020] As a further solution of the present invention: the sampling frequency of the three-axis sensing data is ≥250Hz, the accuracy is ≥0.1g, the three-axis sensing data also includes x-axis acceleration data and y-axis acceleration data, the direction of the x-axis acceleration data is consistent with the vehicle's forward direction, and the z-axis acceleration data is acc i Perpendicular to the plane of the vehicle.

[0021] The embodiment of the present invention aims to provide a bridge joint height difference recognition system based on image and vibration data analysis, comprising:

[0022] An image trigger module is used to collect and generate image information at preset time intervals, and to identify and determine the image information through a preset neural network algorithm, and to generate a determination result, wherein the identification and determination is used to determine whether a vehicle passes through an expansion joint of a bridge, and a determination result is generated when the vehicle passes through the expansion joint;

[0023] A data preparation module is used to respond to the determination result and record the start time t of the vehicle passing through the expansion joint. i-start and the end time t i-end , and obtain the three-axis sensing data of the corresponding time period, the three-axis sensing data includes the z-axis acceleration data acc i ;

[0024] An operation and determination module is used to determine the i-start And the i-end The three-axis sensor data is preprocessed to calculate the vehicle vibration ratio R i , and the vehicle vibration ratio R is determined according to a preset determination threshold. i If the value exceeds the limit, it is determined that the expansion joint has a height difference that needs to be repaired;

[0025] The synchronous recording module stores the relevant data of the abnormal height difference and synchronizes it through the cloud server.

[0026] As a further solution of the present invention: the operation determination module includes:

[0027] A preprocessing unit is used to i-strat To the end time t i-end All the z-axis acceleration data acc within the time period i Perform zero mean and get acc i ', calculate the acc i 'Get the average root mean square and obtain rms i , the rms i To characterize the starting time t i-strat To the end time t i-end The average vibration of the vehicle during the process is processed by a preset window function to obtain an array rms containing ceil((n-w+1) / s) numbers. i_list , wherein w is the size of the window, the size of the window w≤50, s is the step size, the step size s<5, and ceil is a rounding function;

[0028] Calculation unit for taking the rms i_list The maximum value rms i_w_max, and calculate the vehicle vibration ratio R i =rms i_w_max / rms i , the rms i_w_max To characterize the starting time t i-strat To the end time t i-end The maximum short-term vibration of the vehicle within the time period;

[0029] The judgment unit is used to judge the threshold The vehicle vibration ratio R i Make a judgment, if it exceeds the threshold R threshold , it is determined that there is an abnormal height difference at the expansion joint that needs to be repaired.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] In terms of execution efficiency, compared with the existing daily inspection and walking detection methods, this patent is more objective in evaluating the height difference of expansion joints. During daily inspections, the height difference of expansion joints can be quickly evaluated, and the height difference problem of expansion joints can be discovered in time, helping maintenance units to find expansion joints that may be damaged this morning; relying on high-frequency inspections and high-precision positioning, the height difference estimation of expansion joints can be tracked for a long time, and it can be evaluated whether the height difference will reappear after the expansion joints are repaired, and whether it will cause damage again, so as to facilitate maintenance personnel to evaluate whether the joint repair measures are reasonable; in addition, the system mentioned in this patent does not require personnel to perform special operations, and the inspectors only need to focus on driving. Compared with manual walking inspections, the safety risks of inspectors are avoided;

[0032] In terms of economic benefits, the method and system mentioned in this patent mainly rely on intelligent algorithms and lightweight sensing equipment, which greatly improves the inspection frequency. At the same time, the equipment cost is invested on the basis of the existing daily inspection, and the height difference of the expansion joint can be evaluated, which can avoid the related costs of special inspections.

[0033] In terms of social benefits, the method and system mentioned in this patent allow maintenance personnel to quickly and timely locate joints that may be damaged, avoid major damage to the joints, and improve driving safety; at the same time, maintenance personnel can discover expansion joints that may have problems as early as possible, and repair the height difference of the expansion joints before obvious damage occurs, avoiding traffic-enclosed maintenance and ensuring the road's traffic capacity. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a diagram showing the implementation of a method for identifying height differences in bridge joints based on image and vibration data analysis.

[0035] Figure 2This is a diagram showing the principle implementation of the bridge joint height difference identification method based on image and vibration data analysis. DETAILED DESCRIPTION

[0036] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0037] The specific implementation of the present invention is described in detail below in conjunction with specific embodiments.

[0038] A bridge joint height difference recognition method based on image and vibration data analysis provided in one embodiment of the present invention comprises the following steps:

[0039] Image information is collected and generated at preset time intervals, and the image information is identified and judged by a preset neural network algorithm to generate a judgment result. The identification and judgment is used to determine whether a vehicle passes through an expansion joint of a bridge. When the vehicle passes through the expansion joint, a judgment result is generated.

[0040] In response to the determination result, the start time t of the vehicle passing through the expansion joint is recorded. i-strat and the end time t i-end , and obtain the three-axis sensing data of the corresponding time period, the three-axis sensing data includes the z-axis acceleration data acc i .

[0041] According to the starting time t i-start and the end time t i-end The three-axis sensor data is preprocessed to calculate the vehicle vibration ratio R i , and the vehicle vibration ratio R is determined according to a preset determination threshold. i A determination is made, and if exceeded, it is determined that the expansion joint has an abnormal height difference that requires height difference repair.

[0042] The relevant data of the abnormal height difference is stored and synchronized through a cloud server.

[0043] In this embodiment, the image information and the three-axis sensing data are collected by the vehicle-mounted camera and the three-axis acceleration sensor respectively, and the data segment preprocessing and calculation judgment are implemented by the central industrial control computer set on the vehicle (such as Figure 1) When in use, after the system is started, the camera collects images while the vehicle is moving forward, and the deep convolutional neural network in the central industrial control computer analyzes the collected images to determine whether the image contains a bridge expansion joint, record the start time and end time of the vehicle passing through the bridge expansion joint, extract the z-axis acceleration data within this time range, and calculate the abnormal bump index of the z-axis acceleration data. When the abnormal bump index exceeds a certain threshold, it is determined that there is an obvious height difference in the bridge expansion joint and repair is required. Further, for the collection of image information, there are two forms of adjusting the image collection frequency: one is that the collection software obtains the real-time speed of the vehicle through the positioning module, and dynamically adjusts the image collection frequency according to the speed. Secondly, the central industrial computer is connected to the rotary encoder installed on the wheel. Every time the vehicle moves forward a certain distance, the rotary encoder sends a command to the on-board camera to collect images. By combining image and vibration data for analysis, it is possible to quickly, accurately and in real time perceive whether there is an abnormal height difference in the expansion joint of the bridge. In daily inspections, the inspection vehicle only needs to start the system and evaluate the height difference of the expansion joint passed under the condition of driving at a higher speed. When the vehicle is jolted beyond a certain threshold, it is determined that there is a certain height difference in the expansion joint, and the before and after images, before and after time, before and after longitude and latitude, vehicle speed, heading angle, vehicle vibration ratio and other data of passing through the expansion joint are directly recorded locally and uploaded to the cloud system.

[0044] As another preferred embodiment of the present invention, the starting time t i-start and the end time t i-end The three-axis sensor data is preprocessed to calculate the vehicle vibration ratio R i The steps include:

[0045] For the starting time t i-strat To the end time t i-end All the z-axis acceleration data acc within the time period i Perform zero mean and get acc i '.

[0046] Calculate the acc i 'Get the average root mean square and obtain rms i , the rms i To characterize the starting time t i-strat To the end time t i-end Average vibration of the vehicle during the process.

[0047] The acci' is processed by a preset window function to obtain an array rms containing ceil((n-w+1) / s) numbers i_list, wherein w is the size of the window, the window size w≤50, s is the step size, the step size s<5, and ceil is the upward rounding function.

[0048] Take the rms i_list The maximum value rms i_w_max , and calculate the vehicle vibration ratio R i =rms i_w_max / rms i , the rms i_w_max To characterize the starting time t i-strat To the end time t i-end The maximum short-term vibration of the vehicle within a time period.

[0049] According to the preset threshold R threshold The vehicle vibration ratio R i Make a judgment, if it exceeds the threshold R threshold , it is determined that there is an abnormal height difference at the expansion joint that needs to be repaired.

[0050] In this embodiment, Figure 2 As shown in the figure, the data preprocessing, calculation and final judgment level are described in detail. It should be added that for the start time t i-strat and end time t i-end Generation method: Run the target detection deep convolutional neural network on the central industrial computer to analyze each collected image. When a bridge expansion joint is identified in the image, the image acquisition time t is recorded. i， And according to t i， Extrapolate forward and backward a certain time Δt to get t i-strat and t i-end , where Δt can generally be set to a fixed value based on experience, such as 1.5 seconds; it can also be dynamically adjusted according to the actual vehicle speed. The greater the vehicle speed, the smaller Δt. Here, acc i is an array containing n numbers. Generally, filtering is required first to remove the outliers. i It can be uncorrected acceleration data or acceleration data corrected by vehicle speed (the method of vehicle speed correction is to convert acc i *speed 0 / speed i , where speed 0 For the specified driving speed, such as 30km / h, speed i is the start time t i-strat and end time t i-endThe average speed between; w is generally set by experience, generally set to 50, and the step size s is set according to experience, generally set to 1.

[0051] As another preferred embodiment of the present invention, the number of sensors corresponding to the three-axis sensor data is a pair, which are respectively used to generate R i_left , R i_right , when the R i_left With the R i_right Any of the above threshold values ​​R threshold When , it is determined that there is an abnormal height difference at the expansion joint that needs to be repaired, and the R i_left With the R i_right They are used to characterize the vibration and bumps experienced by the left and right wheels of a car.

[0052] Furthermore, the relevant data specifically includes: the start time t corresponding to the expansion joint i-strat , the end time t i-end , as well as the longitude and latitude information of the vehicle corresponding to the time, the image information collected corresponding to the time, the vehicle's driving speed and heading angle.

[0053] Furthermore, the sampling frequency of the three-axis sensor data is ≥250Hz, and the accuracy is ≥0.1g. The three-axis sensor data also includes x-axis acceleration data and y-axis acceleration data. The direction of the x-axis acceleration data is consistent with the vehicle's forward direction, and the z-axis acceleration data is consistent with the vehicle's forward direction. i Perpendicular to the plane of the vehicle.

[0054] In this embodiment, the relevant content is further supplemented and limited here, so that the method can have higher detection accuracy and detection coverage. In order to cope with network fluctuations during vehicle inspection, the relevant data is uploaded by breakpoint resumption. After the data is uploaded to the cloud, it will be recorded in the database and pushed to the corresponding visualization platform and business system. The uploading step is implemented by the central industrial control computer set in the vehicle, which includes a GPU processing unit. The neural network program can achieve prediction acceleration with the help of the GPU processing unit. The central industrial control computer is powered by the vehicle and is equipped with a 4G / 5G mobile Internet module for uploading data and a high-precision positioning module for obtaining the longitude and latitude information of the vehicle.

[0055] The present invention also provides a bridge joint height difference identification system based on image and vibration data analysis, which includes:

[0056] The image trigger module is used to collect and generate image information at preset time intervals, and to identify and determine the image information through a preset neural network algorithm to generate a determination result. The identification and determination is used to determine whether a vehicle passes through the expansion joint of the bridge. When the vehicle passes through the expansion joint, a determination result is generated.

[0057] A data preparation module is used to respond to the determination result and record the start time t of the vehicle passing through the expansion joint. i-start and the end time t i-end , and obtain the three-axis sensing data of the corresponding time period, the three-axis sensing data includes the z-axis acceleration data acc i .

[0058] An operation and determination module is used to determine the i-start And the i-end The three-axis sensor data is preprocessed to calculate the vehicle vibration ratio R i , and the vehicle vibration ratio R is determined according to a preset determination threshold. i A determination is made, and if exceeded, it is determined that the expansion joint has a height difference that requires height difference repair.

[0059] The synchronous recording module stores the relevant data of the abnormal height difference and synchronizes it through the cloud server.

[0060] As another preferred embodiment of the present invention, the operation determination module includes:

[0061] A preprocessing unit is used to i-strat To the end time t i-end All the z-axis acceleration data acc within the time period i Perform zero mean and get acc i ', calculate the acc i 'Get the average root mean square and obtain rms i , the rms i To characterize the starting time t i-strat To the end time t i-end The average vibration of the vehicle during the process is processed by a preset window function to obtain an array rms containing ceil((n-w+1) / s) numbers. i_list , wherein w is the size of the window, the window size w≤50, s is the step size, the step size s<5, and ceil is the upward rounding function.

[0062] Calculation unit for taking the rms i_list The maximum value rms i_w_max , and calculate the vehicle vibration ratio Ri =rms i_w_max / rms i , the rms i_w_max To characterize the starting time t i-strat To the end time t i-end The maximum short-term vibration of the vehicle within a time period.

[0063] The judgment unit is used to judge the threshold The vehicle vibration ratio R i Make a judgment, if it exceeds the threshold R threshold , it is determined that there is an abnormal height difference at the expansion joint that needs to be repaired.

[0064] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0065] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the disclosure in the specification and examples. This application is intended to cover any variations, uses or adaptations of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the claims.

[0066] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A bridge joint height difference recognition method based on image and vibration data analysis is characterized in that: The following steps are involved: Collect and generate image information at preset time intervals, and identify and determine the image information through a preset neural network algorithm, and generate a determination result, wherein the identification and determination is used to determine whether a vehicle passes through an expansion joint of a bridge, and when the vehicle passes through the expansion joint, a determination result is generated; In response to the determination result, the start time t of the vehicle passing through the expansion joint is recorded. i-strat and the end time t i-end , and obtain the three-axis sensing data of the corresponding time period, the three-axis sensing data includes the z-axis acceleration data acc i ; According to the starting time t i-start and the end time t i-end The three-axis sensor data is preprocessed to calculate the vehicle vibration ratio R i , and the vehicle vibration ratio R is determined according to a preset determination threshold. i If the value exceeds the limit, it is determined that the expansion joint has an abnormal height difference and needs to be repaired. The relevant data of the abnormal height difference is stored and synchronized through a cloud server; According to the starting time t i-start and the end time t i-end The three-axis sensor data is preprocessed to calculate the vehicle vibration ratio R i The steps include: For the starting time t i-strat To the end time t i-end All the z-axis acceleration data acc within the time period i Perform zero mean and get acc i '; Calculate the acc i 'Get the average root mean square and obtain rms i , the rms i To characterize the starting time t i-strat To the end time t i-end Average vibration of the vehicle during the process; The acc is processed by a preset window function. i 'Process and obtain the array rms containing ceil((n-w+1) / s) numbers i_list , wherein w is the size of the window, the size of the window w≤50, s is the step size, the step size s<5, and ceil is a rounding function; Take the rms i_list The maximum value rms i_w_max , and calculate the vehicle vibration ratio R i =rms i_w_max / rms i , the rms i_w_max To characterize the starting time t i-strat To the end time t i-end The maximum short-term vibration of the vehicle within a time period.

2. The bridge joint height difference identification method based on image and vibration data analysis according to claim 1 is characterized in that: The number of sensors corresponding to the three-axis sensing data is a pair, which are used to generate R i_left , R i_right , when the R i_left With the R i_right Any of the above threshold values ​​R threshold When , it is determined that there is an abnormal height difference at the expansion joint that needs to be repaired, and the R i_left With the R i_right They are used to characterize the vibration and bumps experienced by the left and right wheels of a car.

3. The bridge joint height difference identification method based on image and vibration data analysis according to claim 1 is characterized in that: The relevant data specifically includes: the start time t corresponding to the expansion joint i-strat , the end time t i-end , as well as the longitude and latitude information of the vehicle corresponding to the time, the image information collected corresponding to the time, the vehicle's driving speed and heading angle.

4. The bridge joint height difference identification method based on image and vibration data analysis according to claim 1 is characterized in that: The sampling frequency of the three-axis sensor data is ≥250Hz, and the accuracy is ≥0.1g. The three-axis sensor data also includes x-axis acceleration data and y-axis acceleration data. The direction of the x-axis acceleration data is consistent with the vehicle's forward direction, and the z-axis acceleration data is acc i Perpendicular to the plane of the vehicle.

5. A bridge joint height difference recognition system based on image and vibration data analysis, used to implement any of the methods described in claims 1-4, characterized in that: include: An image trigger module, used to collect and generate image information at preset time intervals, and to identify and determine the image information through a preset neural network algorithm, and to generate a determination result, wherein the identification and determination is used to determine whether a vehicle passes through an expansion joint of a bridge, and a determination result is generated when the vehicle passes through the expansion joint; A data preparation module is used to respond to the determination result and record the start time t of the vehicle passing through the expansion joint. i-start and the end time t i-end , and obtain the three-axis sensing data of the corresponding time period, the three-axis sensing data includes the z-axis acceleration data acc i ; An operation and determination module is used to determine the i-start And the i-end The three-axis sensor data is preprocessed to calculate the vehicle vibration ratio R i , and the vehicle vibration ratio R is determined according to a preset determination threshold. i If the value exceeds the limit, it is determined that the expansion joint has an abnormal height difference and needs to be repaired. The synchronous recording module is used to store the relevant data of the abnormal height difference and synchronize it through the cloud server.

6. The bridge joint height difference identification system based on image and vibration data analysis according to claim 5 is characterized in that: The operation determination module comprises: A preprocessing unit is used to i-strat To the end time t i-end All the z-axis acceleration data acc within the time period i Perform zero mean and get acc i ', calculate the acc i 'Get the average root mean square and obtain rms i , the rms i To characterize the starting time t i-strat To the end time t i-end The average vibration of the vehicle during the process is calculated by a preset window function. i 'Process and obtain the array rms containing ceil((n-w+1) / s) numbers i_list , wherein w is the size of the window, the size of the window w≤50, s is the step size, the step size s<5, and ceil is a rounding function; Calculation unit for taking the rms i_list The maximum value rms i_w_max , and calculate the vehicle vibration ratio R i =rms i_w_max / rms i , the rms i_w_max To characterize the starting time t i-strat To the end time t i-end The maximum short-term vibration of the vehicle within the time period; The judgment unit is used to judge the threshold The vehicle vibration ratio R i Make a judgment, if it exceeds the threshold R threshold , it is determined that there is an abnormal height difference at the expansion joint that needs to be repaired.

Images