Underwater bridge pier structure inspection system based on unmanned underwater robot

Abstract

The application relates to the technical field of structure inspection, in particular to a bridge pier underwater structure inspection system based on an unmanned underwater robot, which comprises an inspection cycle adjustment module, an abnormality identification module, an automatic retake module, a damage evaluation module and a bridge pier abnormality evaluation module.The application can realize fine management of the inspection work by executing bridge pier underwater structure inspection through the unmanned underwater robot and dynamically adjusting the inspection cycle according to traffic flow, can adjust the inspection frequency according to actual use conditions and external influencing factors, can effectively prevent and find potential structure problems, can quickly and accurately mark out a damage area by using image acquisition and automatic abnormality identification of the unmanned robot and can execute retake when necessary, can improve image quality, can ensure the reliability of data, and can accurately evaluate the influence of damage on structure safety by comparing the position and coincidence degree of the damage area and a key load-bearing area of the bridge pier, so that scientific basis can be provided for maintenance decision.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of structure inspection, and in particular to a bridge pier underwater structure inspection system based on an unmanned underwater robot. BACKGROUND

[0002] The technical field of structure inspection involves periodic detection and evaluation of the integrity, safety, and durability of various structures such as building structures, transportation infrastructure, water conservancy projects, and large industrial facilities. This technical field includes crack detection, corrosion detection, settlement monitoring, fatigue damage detection, and other technical means, covering ultrasonic detection, magnetic powder detection, impact echo method, eddy current detection, three-dimensional laser scanning, acoustic emission detection, underwater sonar imaging, and visual imaging detection methods. The technical field of structure inspection focuses on defect identification, damage assessment, and life prediction of structures, aiming to achieve quantitative evaluation of structure state and safety risk determination through combination of various detection methods and analysis models, providing technical support for subsequent reinforcement, repair, or replacement.

[0003] Among them, the bridge pier underwater structure inspection system is a system for periodic inspection of the underwater structure of the bridge pier, specifically for detecting cracks, corrosion, erosion, shedding, and structural deformation of the underwater structure of the bridge pier. The system can complete the visualization modeling, defect automatic identification, and damage degree quantitative evaluation of the underwater area of the bridge pier by combining multiple underwater detection technologies and image processing algorithms. The purpose is to provide accurate technical basis for maintenance decision, risk control, and service life prediction of the underwater part of the bridge, ensuring long-term safe and stable operation of the overall structure of the bridge.

[0004] Traditional inspection systems rely on manual or semi-automatic detection methods, which are limited by the technical level and experience of the operators, making it difficult to achieve high efficiency and high precision detection. For example, the traditional inspection system requires divers or limited automatic equipment to perform inspection at fixed intervals, which is not only time-consuming and labor-intensive, but also difficult to perform in adverse weather or complex hydrological environments. In addition, fixed-period inspection cannot be adjusted according to actual use and environmental changes, resulting in missed detection during high-risk periods or waste of resources during low-risk periods. This rigid inspection mode limits the rational allocation and optimization of resources, and has defects in early damage identification and timely maintenance of critical parts, increasing the risk of structural safety, thus shortening the service life of the bridge. SUMMARY

[0005] The purpose of the present application is to solve the problems existing in the prior art, and to provide a bridge pier underwater structure inspection system based on an unmanned underwater robot.

[0006] To achieve the above purpose, the present application adopts the following technical scheme: a bridge pier underwater structure inspection system based on an unmanned underwater robot, comprising:

[0007] The inspection cycle adjustment module obtains bridge traffic flow data, calculates a difference value between a peak period cumulative traffic volume and a valley period cumulative traffic volume, calls a water flow shear force parameter, calculates a shear force increment value and a difference value between the shear force increment value and a safety critical shear force increment threshold value, adjusts a pier inspection cycle, and generates a dynamic inspection cycle adjustment result;

[0008] The abnormality identification module uses an unmanned underwater robot to perform underwater pier structure inspection based on the dynamic inspection cycle adjustment result, obtains underwater inspection images, identifies an abnormal area by comparing known pier abnormal images, marks the abnormal area, and generates an abnormal area marking result;

[0009] The automatic retake module calls the abnormal area marking result, extracts a pixel noise index and a definition index, evaluates the shooting quality of the abnormal area, adjusts light source brightness and exposure time if the shooting quality is lower than a set quality threshold value, performs retakes in situ, and generates local abnormal retake images;

[0010] The damage evaluation module calls the local abnormal retake images, extracts a damage center position, compares the damage center position with a pier key area, calculates a damage area overlap degree, calculates a damage severity in combination with a damage type, and generates damage evaluation information.

[0011] The present application improves that the dynamic inspection cycle adjustment result includes a cycle adjustment coefficient, a flow difference threshold trigger state, and a water flow shear force increment state, the abnormal area marking result specifically includes an abnormal boundary position parameter, an abnormal shape determination identifier, and an abnormal area center point coordinate, the local abnormal retake images include a retake light source brightness level, a retake exposure time setting value, and a retake area imaging data set, and the damage evaluation information specifically includes a damage type weight value, a sensitive area overlap coefficient, and a comprehensive influence path coefficient.

[0012] The present application improves that the inspection cycle adjustment module includes:

[0013] The traffic flow calculation sub-module obtains bridge traffic flow data, divides a time interval according to a peak period and a valley period, calculates a difference value between a peak period cumulative traffic volume and a valley period cumulative traffic volume, and obtains a peak-valley traffic difference value;

[0014] The shear force increment calculation sub-module calls the peak-valley traffic difference value, obtains a water flow shear force parameter, calculates a difference between a current shear force increment value and a safety critical shear force increment threshold value according to the safety critical shear force increment threshold value, and combines a preset pier inspection cycle and the peak-valley traffic difference value to use the formula:

[0015]

[0016] The operation obtains a cycle adjustment coefficient;

[0017] wherein, T adj represents a cycle adjustment coefficient, Q d represents a normalized value of the cumulative traffic volume in the peak period, Q p represents a normalized value of the cumulative traffic volume in the valley period, wherein S i represents a normalized value of the water flow velocity change amount in the i th time period, C s represents a safety critical shear force increment threshold value, F d represents a normalized value of the current shear force increment value, F s represents a normalized value of the shear force reference base value, n represents the total number of time periods in which the water flow velocity change amount participates in the calculation;

[0018] The cycle adjustment output submodule judges the cycle adjustment direction according to the cycle adjustment coefficient, combines a preset pier inspection cycle, calculates a post-adjustment cycle value, adjusts the pier inspection cycle, and generates an inspection cycle dynamic adjustment result.

[0019] The application improves that the abnormality identification module comprises:

[0020] The underwater image acquisition submodule schedules the unmanned underwater robot to perform pier underwater structure inspection in a specified time period based on the inspection cycle dynamic adjustment result, collects a continuous underwater image frame sequence of a pier bottom structure area, establishes an image sequence library, and acquires a pier underwater image sequence.

[0021] The feature matching degree calculation submodule calls the pier underwater image sequence, combines a known pier abnormal image, acquires a detection area gray mean value, a texture direction mean value and a boundary curvature sequence, compares the detection area gray mean value, the texture direction mean value and the boundary curvature sequence with a reference area gray mean value, a texture direction mean value and a boundary curvature sequence, and adopts the formula:

[0022]

[0023] The operation acquires an abnormal area matching degree value;

[0024] wherein, M d represents an abnormal area matching degree, G d represents a normalized value of the detection area gray mean value, G r represents a normalized value of the reference area gray mean value, T d represents a normalized value of the detection area texture direction mean value, T r represents a normalized value of the reference area texture direction mean value, C d,j represents the j th boundary curvature value of the detection area, C r,j represents the j th boundary curvature value of the reference area, and N represents the number of boundary curvature sampling points;

[0025] The region marking output submodule compares the abnormal region matching degree value with a preset matching degree threshold, identifies a region with a matching degree value greater than the threshold, combines the detection region center coordinates, marks the abnormal region position and boundary range, and generates an abnormal region calibration result.

[0026] The application improves that the automatic retake module comprises:

[0027] The image block extraction submodule calls the abnormal region calibration result, divides a pixel grid for the abnormal region, generates a fixed-size image block set according to the pixel grid coordinates, extracts an image block grayscale matrix, calculates the image block grayscale variance value and the mean gradient value, and obtains an image basic parameter group;

[0028] The quality evaluation submodule compares the grayscale variance value with an image definition reference value according to the image basic parameter group, judges the difference between the mean gradient value and a pixel noise reference value, combines the two judgment results to grade the image block, labels the quality grading label for each image block, and generates an image quality grading label group;

[0029] The image retake submodule calls the image quality grading label group, filters the image block labeled with a low quality grade, extracts the spatial position information of the filtered image block, adjusts the exposure time and light source brightness parameters of the unmanned underwater robot camera according to the position coordinates, performs in-situ retake, and generates a local abnormal retake image.

[0030] The application improves that the damage evaluation module comprises:

[0031] The structure parameter extraction submodule is based on the local abnormal retake image, divides each damage region in the image, extracts the damage region boundary line, calculates the boundary curvature value, counts the pixel proportion of the damage region in the entire image, and identifies the center position coordinates of the damage region, to obtain a damage structure parameter set;

[0032] The coincidence degree calculation submodule is based on the damage structure parameter set, calls the spatial boundary data of the shear concentration region, the axial pressure change region and the bending resistance bearing region of the pier, calculates the spatial overlapping range of the damage region and the key region, calculates the ratio of the overlapping area to the total area of the corresponding key region, and obtains the damage region coincidence degree value;

[0033] The severity evaluation submodule calls the damage region coincidence degree value and the damage structure parameter set, extracts the damage region boundary curvature, area proportion, spatial coincidence degree and corresponding damage type, and uses the formula:

[0034]

[0035] The operation obtains the damage severity value of the damage region, and generates damage evaluation information;

[0036] wherein S zr represents the damage severity value, C zr represents the normalized value of the boundary curvature of the damage area, A zr represents the area proportion of the damage area, R zr represents the coincidence value of the damage area and the key area, K zr represents the coincidence sensitivity reference value corresponding to the damage area type, D zr represents the damage type influence coefficient.

[0037] The system further comprises:

[0038] The pier abnormality evaluation module calls the damage evaluation information, screens the areas with damage severity exceeding a set threshold, counts the damage types and distribution ranges, calculates the pier abnormality influence degree in combination with the number and concentration trend of the damage areas, and generates a pier abnormality evaluation result.

[0039] The pier abnormality evaluation result specifically refers to the damage distribution density, the total amount of damage quantity, and the damage concentration trend coefficient.

[0040] The pier abnormality evaluation module comprises:

[0041] The high-risk area screening submodule extracts the severity value of each damage area based on the damage evaluation information, compares each area with a set severity threshold, screens the areas exceeding the threshold, extracts the damage type label and spatial range of the corresponding area, and generates a high-severity area set.

[0042] The distribution parameter extraction submodule calls the high-severity area set, counts the number of areas of each damage type, extracts the centroid coordinate set of each area, and calculates the density change trend value of the centroid distribution to obtain damage distribution characteristics.

[0043] The influence degree calculation submodule calls the damage distribution parameter set, extracts the number, distribution density, and concentration trend coefficient of each type of damage, combines the screened area set, and uses the formula:

[0044]

[0045] to obtain the pier abnormality influence degree and obtain the pier abnormality evaluation result.

[0046] wherein I b represents the pier abnormality influence degree, N as,k represents the number of the kth damage area, D as,k represents the distribution density of the kth damage area, T as,k represents the concentration trend coefficient of the kth damage area, K represents the total number of damage types, and S as,kThe average value of the damage severity of the kth damage area.

[0047] Compared with the prior art, the application has the advantages and positive effects that:

[0048] In the application, the underwater structure of the pier is inspected by the unmanned underwater robot, and the inspection cycle is dynamically adjusted according to the traffic flow, so that the fine management of the inspection work is realized, the inspection frequency can be adjusted according to the actual use condition and external influencing factors, potential structural problems can be effectively prevented and found, the image collection and automatic anomaly identification by the unmanned robot can quickly and accurately mark the damage area, and the image quality is improved when necessary, the reliability of the data is ensured, the influence of the damage on the structural safety can be accurately evaluated by comparing the position and coincidence degree of the damage area and the key bearing area of the pier, a scientific basis for maintenance decision is provided, the efficiency and accuracy of the inspection are improved by the automation and intelligent means, the uncertainty and risk of manual operation are reduced, and the timeliness and safety guarantee capability of the bridge maintenance are significantly improved. BRIEF DESCRIPTION OF DRAWINGS

[0049] Figure 1 The system flowchart of the application is shown in the figure;

[0050] Figure 2 The flowchart of the inspection cycle adjustment module of the application is shown in the figure;

[0051] Figure 3 The flowchart of the anomaly identification module of the application is shown in the figure;

[0052] Figure 4 The flowchart of the automatic retake module of the application is shown in the figure;

[0053] Figure 5 The flowchart of the damage evaluation module of the application is shown in the figure;

[0054] Figure 6 The flowchart of the pier anomaly evaluation module of the application is shown in the figure. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical scheme and advantages of the application more clear, the application is further described in detail below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the application, and are not used to limit the application.

[0056] In the description of the present application, it should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application. In addition, in the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise specifically limited.

[0057] Please refer to Figure 1 The present application provides a technical solution: a bridge pier underwater structure inspection system based on an unmanned underwater robot, the system comprising:

[0058] The inspection cycle adjustment module obtains bridge traffic flow data, segments and accumulates according to peak and valley sections, calculates the difference between the peak section cumulative traffic volume and the valley section cumulative traffic volume, calls the flow shear force parameter, calculates the difference between the shear force increment value and the safety critical shear force increment threshold value, combines the preset bridge pier inspection cycle, calculates the bridge pier inspection cycle adjustment value, adjusts the bridge pier inspection cycle, and generates the inspection cycle dynamic adjustment result;

[0059] The abnormality recognition module uses the unmanned underwater robot to perform bridge pier underwater structure inspection based on the inspection cycle dynamic adjustment result, obtains underwater inspection images, compares known bridge pier abnormal images, calculates the matching degree of the abnormal area according to the characteristics of the area, identifies the abnormal area, and marks the abnormal area to generate the abnormal area calibration result;

[0060] The automatic retake module calls the abnormal area calibration result, extracts local area image blocks, extracts pixel noise index and definition index, evaluates the shooting quality of the abnormal area, and if the shooting quality is lower than the set quality threshold, adjusts the light source brightness and exposure time, and retakes in situ to generate local abnormal retake images;

[0061] The damage evaluation module calls the local abnormal retake image, extracts the boundary curvature and area ratio of the damage area, and extracts the damage center position, compares it with the bridge pier key area, which includes the shear force concentration area, the axial pressure change area and the bending bearing area, calculates the damage area overlap degree, combines the damage type, calculates the damage severity, and generates the damage evaluation information;

[0062] The bridge pier abnormality evaluation module calls the damage evaluation information, filters the areas where the damage severity exceeds the set threshold, counts the damage type and distribution range, combines the number and concentration trend of the damage area, calculates the bridge pier abnormality influence degree, and generates the bridge pier abnormality evaluation result.

[0063] The patrol cycle dynamic adjustment result includes a cycle adjustment coefficient, a flow difference threshold trigger state and a water flow shear force increment state, the abnormal area calibration result is specifically an abnormal boundary position parameter, an abnormal shape judgment identifier and an abnormal area center point coordinate, the local abnormality supplementary image includes a supplementary light source brightness level, a supplementary exposure time setting value and a supplementary area imaging data set, the damage evaluation information is specifically a damage type weight value, a sensitive area coincidence coefficient and a comprehensive influence path coefficient, and the pier abnormality evaluation result is specifically a damage distribution density, a damage quantity total amount and a damage concentration trend coefficient.

[0064] Please refer to Figure 2 , the patrol cycle adjustment module includes:

[0065] The traffic flow calculation sub-module obtains bridge traffic flow data, divides the time interval according to the peak section and the trough section, calculates the cumulative traffic volume difference between the peak section and the trough section, and obtains the peak-trough traffic difference;

[0066] The traffic flow calculation sub-module obtains bridge traffic flow data, needs to call the vehicle counting device deployed at the entrance and exit of the bridge in the bridge traffic monitoring system. The device generates a set of vehicle passing data every 5 minutes. The collection object is a motor vehicle passing through the bridge. The recording content includes time stamp, vehicle type, traffic direction and vehicle speed parameters. The daily traffic volume data is formed by the data uploading module. In the execution, the 24 hours of each day are first divided into several time intervals. The peak section time is set to 7:00-9:00 and 17:00-19:00, and the trough section time is set to 1:00-5:00. The cumulative traffic volume of the vehicle in the corresponding time period is calculated respectively. For each peak section or trough section, the cumulative traffic volume is obtained by the following operation: the number of vehicles passing through every 5 minutes in the time period is added up. Taking the time period 7:00-9:00 as an example, if the 5-minute average traffic volume is 42 vehicles, then the cumulative traffic volume of the peak section is (2 hours x 60 / 5) x 42 = 1008 vehicles. The trough section time is set to 1:00-5:00, i.e. 4 hours, a total of 48 5-minute units. If the average traffic volume is 12 vehicles, then the cumulative traffic volume of the trough section is 48 x 12 = 576 vehicles. Then the cumulative traffic volume of the above peak section and trough section is normalized. The normalization reference value is set to the maximum traffic capacity of the bridge. Specifically, the upper limit of the hourly traffic capacity of the bridge is set to 800 vehicles. The value is determined based on the number of one-way traffic lanes, lane width and speed limit standard carried in the traffic design drawing, combined with the actual traffic density simulation. The value is periodically updated according to the holiday traffic statistical value and the highest traffic peak value in the past years. The current setting of the daily maximum single-period traffic capacity is 800 x 2 = 1600 vehicles. According to this, the normalized value of the peak section is Q d = 1008 / 1600 = 0.63, and the normalized value of the trough section is Q p= 576 / 1600 = 0.36, and then the difference between the two normalized values is calculated as the peak-to-trough traffic difference, i.e., Q d - Q p = 0.63-0.36 = 0.27, which represents the fluctuation amplitude of the traffic flow at different times, and can be used to show the difference in traffic pressure of the bridge at different times, and finally serves as one of the inputs for the subsequent shear force increment calculation.

[0067] The shear force increment calculation submodule calls the peak-to-trough traffic difference, obtains the water flow shear force parameter, calculates the difference between the current shear force increment value and the safety critical shear force increment threshold value according to the safety critical shear force increment threshold value, and combines the preset pier inspection period and the peak-to-trough traffic difference to use the formula:

[0068]

[0069] The operation obtains the period adjustment coefficient;

[0070] wherein, T adj represents the period adjustment coefficient, Q d represents the normalized value of the cumulative traffic volume in the peak period, Q p represents the normalized value of the cumulative traffic volume in the trough period, wherein S i represents the normalized value of the water flow velocity change in the i-th time period, C s represents the safety critical shear force increment threshold value, F d represents the normalized value of the current shear force increment value, F s represents the normalized value of the shear force reference baseline value, and n represents the total number of time periods in which the water flow velocity change participates in the calculation.

[0071] The shear force increment calculation submodule calls the aforementioned peak-to-trough traffic difference 0.27, and obtains the water flow velocity data from the flow speed monitoring device arranged in the water area under the bridge. The device records the water flow velocity with a sampling period of 15 minutes. It is assumed that n = 4 in a certain time period, i.e., 1 hour of data is selected as 1.2 m / s, 1.5 m / s, 1.7 m / s, and 2.1 m / s. Based on these data, the water flow velocity changes in adjacent time periods are calculated as 0.3 m / s, 0.2 m / s, and 0.4 m / s. The normalized processing is performed on these differences, and the normalized baseline value is set as the historical maximum flow speed variation amplitude 0.5 m / s in the monitoring area. This value is derived from the upper limit value of the 95% confidence interval in the daily maximum variation statistical results of the cross section of the water area where the pier foundation of the bridge is located in the past 5 years, which serves as the limit representative value of the hydrological disturbance intensity change. This value periodically fluctuates in the wet season and the dry season. The current selected 0.5 m / s is used as a static parameter for normalization processing, and the corresponding normalized results are S1 = 0.3 / 0.5 = 0.6, S2 = 0.2 / 0.5 = 0.4, and S3 = 0.4 / 0.5 = 0.8. The summation term is calculated as Subsequently, the safety critical shear force increment threshold C is obtained s The threshold is the maximum allowable shear force growth limit set by the bridge structure design unit, which is derived from the critical shear stress limit provided in the structure fatigue calculation, and is obtained through material test data combined with finite element stress simulation analysis. The current selected value is 0.55, which is normalized and dimensionless. The setting basis is the equivalent safety stress increment limit of the pier concrete wrapped steel structure under the temperature of 35°C and the water flow speed of 2.5m / s. The adjustment period is annual evaluation and update once. The current shear force increment F d The current conversion value is 0.47, and the reference baseline value F s =0.30, which represents the average shear force increment on the pier surface under the action of water flow in normal state. It is measured under the basis working condition of stable flow speed of 1.2m / s, and is the standard average value in spring low water period in recent years. It is not adjusted dynamically with seasons, but is corrected once every three years according to the water flow evolution trend. Similarly, after normalization, the difference between the current shear force increment value and the baseline shear force is |F d -F s |=|0.47-0.30|=0.17, the above data is substituted into the formula:

[0072]

[0073] The period adjustment coefficient is obtained as 0.675. The value reflects the influence degree of the bridge traffic pressure and the water flow impact on the pier stability, and provides a quantitative basis for the subsequent period adjustment submodule. The benefit of the formula is that the product calculation of the sum of the peak and valley traffic difference and the water flow speed change is combined with the safety critical shear force increment and the difference value of the current shear force change to form the denominator, so that the adjustment coefficient can reflect the coupling degree of traffic load change and fluid shear influence.

[0074] The period adjustment output submodule judges the period adjustment direction according to the period adjustment coefficient, combines the preset pier inspection period, calculates the adjusted period value, adjusts the pier inspection period, and generates the inspection period dynamic adjustment result;

[0075] The period adjustment output submodule receives the period adjustment coefficient T adj= 0.675, and combined with the bridge preset inspection cycle, the original cycle is set to 30 days, by comparing the cycle adjustment coefficient with the benchmark interval, the benchmark interval is set to [0.3, 0.5], which is determined by the average standard deviation range of the historical five-year cycle adjustment data of the bridge, and the lower limit 0.3 represents the stable state of the structure, and the upper limit 0.5 represents the boundary of shear risk, and the current value 0.675 exceeds the upper limit 0.5, which means that the inspection demand is rising, if the cycle adjustment coefficient is greater than 0.5, it is determined that the cycle is shortened, according to the adjustment ratio, the adjusted cycle is calculated, the adjustment ratio function is set to (1-T adj ), and the adjustment coefficient is 1-0.675=0.325, so the adjusted cycle is 30*0.325=9.75 days, rounded to 10 days, and the result is output to the inspection scheduling system to generate the dynamic adjustment result of the inspection cycle. The adjustment ratio function is a direct expression of the cycle shortening rate, which is represented by 1 minus the shear comprehensive risk index, which reflects its correction strength to the original cycle, and the coefficient value range is (0, 1), and the current value 0.325 represents that the original cycle is shortened by 32.5%, which ensures that the cycle adjustment operation has sufficient response strength, and the result shows that the current state of the pier needs to increase the detection frequency, and the adjusted inspection cycle value 10 days is used to replace the original cycle 30 days to dynamically update the maintenance plan.

[0076] Please refer to Figure 3 , the abnormality recognition module comprises:

[0077] The underwater image acquisition submodule is based on the dynamic adjustment result of the inspection cycle, and schedules the unmanned underwater robot to perform underwater structure inspection of the pier in a specified time period, collects continuous underwater image frame sequences of the bottom structure area of the pier, establishes an image sequence library, and obtains the underwater image sequence of the pier.

[0078] The underwater image acquisition submodule is based on the output of the aforementioned cycle adjustment submodule. After receiving the adjusted inspection cycle value of 10 days, the corresponding task scheduling instruction is generated through the bridge maintenance scheduling system to control the unmanned underwater robot to enter the working state within the 10th day. The specific working time window is set to the period of relatively stable water flow from 9:00 to 11:00 in the morning. The scheduling control instruction is composed of the task serial number, the inspection area number, the collection time period, and the preset path point. After receiving the instruction, the robot moves to the specified pier bottom area according to the path points, uniformly cruises at a speed of 0.5 m / s within a 5-meter radius, and collects underwater image frames at a frequency of 25 frames per second for 20 minutes. The total number of image frames is 20x60x25=30000. The image frames are numbered and organized into an image sequence library in chronological order. Each image is attached with the current water depth, attitude angle, and shooting timestamp. The image format is uniformly converted to grayscale for subsequent processing. The image resolution is set to 1920x1080, and the spatial resolution is set to 1 pixel corresponding to an actual size of 2 millimeters. After noise removal, sharpening, and equalization, the image enters the subsequent feature calculation link, and the acquisition of the underwater image sequence of the pier is finally completed.

[0079] The feature matching degree calculation submodule calls the underwater image sequence of the pier and combines the known pier abnormal image to obtain the detection area gray mean value, texture direction mean value, and boundary curvature sequence, and compares them with the reference area gray mean value, texture direction mean value, and boundary curvature sequence. The formula is:

[0080]

[0081] The operation obtains the abnormal area matching degree value.

[0082] Wherein, M d represents the abnormal area matching degree, G d represents the normalized value of the detection area gray mean value, G r represents the normalized value of the reference area gray mean value, T d represents the normalized value of the detection area texture direction mean value, T r represents the normalized value of the reference area texture direction mean value, C d,j represents the jth boundary curvature value of the detection area, C r,j represents the jth boundary curvature value of the reference area, and N represents the number of boundary curvature sampling points.

[0083] The feature matching degree calculation submodule calls the underwater image data in the above image sequence library and combines the known pier abnormal area image as a reference image. The boundary range of the to-be-detected area and the reference area is selected to be the same, and the image frame shooting angle is similar. The image block size is uniformly cropped to 256x256 pixels. First, the gray mean value Gd , the gray value distribution range is 0 to 255, if the total amount of gray value summation in a certain image block is 3276800, then the gray mean value is G d = 3276800 / (256 x 256) = 50.0, corresponding to the reference area gray mean value G r = 45.0, the gray mean value normalization processing uses the maximum gray value 255 as the reference, that is, after normalization G d = 0.196, G r = 0.176, the texture direction mean value T d , T r respectively represent the statistical value of the gradient direction in the image, the calculation method is to calculate the direction angle by performing Sobel gradient calculation on each pixel neighborhood, and then to calculate the mean value in the region, assuming that T d = 62.5°, T r = 60.0°, after conversion to radians, they are 1.091 and 1.047 respectively, the normalization reference is set to π, and the normalized value is T d = 1.091 / π ≈ 0.347, T r = 1.047 / π ≈ 0.333, the boundary curvature sequence C d,j , C r,j respectively represent the curvature sampling points on the boundary of the detection area and the reference area, the curvature calculation is based on the tangent change rate of the boundary contour point, assuming that the sampling point number N = 5, the curvature values of the five detection points are [0.18, 0.20, 0.22, 0.24, 0.26], the reference area is [0.15, 0.19, 0.21, 0.23, 0.25], the absolute value of the difference corresponding to each term is [0.03, 0.01, 0.01, 0.01, 0.01], the sum is 0.07, and the average difference is 0.07 / 5 = 0.014, which is substituted into the formula

[0084]

[0085] , the matching degree value is obtained as 0.9745, which is the final operation output result, the value closer to 1 indicates that the detection area is more similar to the reference abnormal area, the current value indicates that the detection image is more similar to the reference abnormal image, the gray mean value normalization reference value in the participation term is set to 255, which is defined according to the maximum value of the 8-bit image gray code, and the value is a constant value and does not change with the scene; the texture direction normalization takes π radians as the upper limit, which is derived from the physical definition range of the direction angle; the boundary curvature is not normalized, and the sampling value is directly used for matching, N is the curvature sampling number, which is set to 5, and is determined according to the boundary smoothness and contour point density, if the image boundary complexity increases, N value should be increased to 10 to improve the contour difference sensitivity.

[0086] The region marking output submodule compares the abnormal region matching degree value with a preset matching degree threshold value, identifies the region with a matching degree value greater than the threshold value, combines the detection region center coordinates, marks the abnormal region position and boundary range, and generates an abnormal region calibration result.

[0087] The region marking output submodule calls the aforementioned abnormal region matching degree value 0.9745 and compares it with a preset matching degree threshold value, which is set to 0.90. The threshold value is derived from the fitting analysis process of the comparison results of artificial sample labeling and images. After distributing analysis of the matching degree values in 100 groups of abnormal image and normal image comparison results, the 95% quantile is set as the lower limit of the matching degree judgment threshold value. If the matching degree value is higher than this value, it means that the detection region and the reference abnormal region have structural feature similarity. The current matching degree value 0.9745>0.90, so it is determined that the region is an abnormal region. Then the center point coordinates of the detection region in the image frame are read. A region range with a radius of 32 pixels is set as the boundary in the image pixel unit, and the corresponding actual distance is 32x2mm=64mm. The coordinates and boundary box parameters are labeled in the image sequence, the position information and boundary range of the abnormal region in the frame image are marked, and the abnormal region calibration result is generated. This result will be used as the original input data in the subsequent image recognition statistics and maintenance suggestion generation.

[0088] Please refer to Figure 4 , the automatic retake module includes:

[0089] The image block extraction submodule calls the abnormal region calibration result, divides the abnormal region into a pixel grid according to the pixel grid coordinates, generates a set of fixed-size image blocks according to the pixel grid coordinates, extracts the image block grayscale matrix, calculates the image block grayscale variance value and mean gradient value, and obtains the image basic parameter group.

[0090] The image block extraction submodule calls the aforementioned abnormal region calibration result. First, the center point coordinates and boundary pixel coordinates of the abnormal region in the image frame labeled as abnormal are obtained. The abnormal region is divided into equally spaced pixel grids according to the pixel coordinates, and each grid has a length of 64 pixels. A plurality of grid index regions are generated. Then, the image pixel data in each grid region is extracted to obtain fixed-size image blocks, and the size is uniform at 64x64 pixels. The image block is read in grayscale form, and the pixel value range is 0-255. The corresponding grayscale matrix data of each image block is extracted, and the grayscale matrix is defined as a two-dimensional array G i,j , where i and j are the row and column indices of the image block, for example, the random 5-row data in the grayscale matrix of a certain image block is as follows:

[0091] 52, 53, 55, 58, 60], [51, 54, 57, 59, 61], [50, 52, 54, 56, 58], [48, 50, 52, 53, 55], [47, 48, 50, 51, 53], the gray scale variance value is calculated by first calculating the gray scale mean value G of the image block avg , assuming that the sum of all pixel values is 208896 and the number of pixels is 64x64=4096, then G avg =208896 / 4096=51.0, the gray scale variance value is the square of the difference between all pixel values and the mean value, that is Assuming that the obtained value is 36.0, the mean gradient value is calculated by the absolute value of the gray scale difference between each adjacent pixel in the image block, both the horizontal and vertical directions are involved in the summation, and then divided by the number of participating items, assuming that the final mean gradient value is 8.4, the image basic parameter group is composed of the gray scale variance value 36.0 and the mean gradient value 8.4.

[0092] The quality evaluation submodule compares the gray scale variance value with the image sharpness reference value according to the image basic parameter group, judges the difference between the mean gradient value and the pixel noise reference value, and grades the image block according to the two judgment results, labels the quality grading label for each image block, and generates the image quality grading label group;

[0093] The quality evaluation submodule receives the image basic parameter group and quantitatively grades the image block quality. First, compare the gray scale variance value with the image sharpness reference value, set the image sharpness reference value to 25.0, set the gray scale variance statistical median value corresponding to the image that the human eye can distinguish the outline in the image collected by the robot platform under the condition of water depth 5 meters and water flow rate 1.5 m / s, which is stable under the condition of fixed resolution (1920x1080) and fixed focal length (6mm) of the image acquisition device, if different devices are used, the value needs to be reset, the current gray scale variance value of the image block is 36.0, which is higher than the reference value, meeting the sharpness requirement, then judge the difference between the mean gradient value and the pixel noise reference value, set the pixel noise reference value to 5.0, which is the mean value of the random offset of the pixels measured from the standard static image obtained under the condition that the shooting device is turned off, reflecting the degree of local brightness fluctuation in the image, the value is stable at 5.0 under the parameters of exposure time 1 / 50 second and ISO sensitivity 800 of the current device, the current mean gradient value is 8.4, which is higher than the reference value, indicating that the image edge information density exceeds the noise influence level, so the image block quality level can be determined as "high", the quality grading label is set to "A", if any parameter does not meet the reference value, it is marked as "B" or "C" level, according to the rule: those with gray scale variance value lower than 25.0 and mean gradient value lower than 5.0 are "C", the rest are "B", finally generate the image quality grading label group, which is composed of each image block and its corresponding quality level label.

[0094] The image re-shooting module calls the image quality grading label group, screens the image blocks marked as low quality, extracts the spatial position information of the screened image blocks, adjusts the exposure time and light source brightness parameters of the unmanned underwater robot camera according to the position coordinates, re-shoots in situ, and generates a local anomaly re-shooting image;

[0095] The image re-shooting module calls the image quality grading label group, screens the image blocks marked as low quality, extracts the spatial position information of the screened image blocks, adjusts the exposure time and light source brightness parameters of the unmanned underwater robot camera according to the position coordinates, re-shoots in situ, and generates a local anomaly re-shooting image; Figure 1 The image re-shooting module calls the image quality grading label group, screens the image blocks marked as low quality, extracts the spatial position information of the screened image blocks, adjusts the exposure time and light source brightness parameters of the unmanned underwater robot camera according to the position coordinates, re-shoots in situ, and generates a local anomaly re-shooting image;

[0096] Please refer to Figure 5 , the damage assessment module comprises:

[0097] The structure parameter extraction submodule is based on the local anomaly re-shooting image, segments each damage area in the image, extracts the damage area boundary line, calculates the boundary curvature value, counts the pixel proportion of the damage area in the whole image, and identifies the center position coordinates of the damage area to obtain a set of damage structure parameters;

[0098] The structure parameter extraction submodule calls the local exception retake image, after obtaining the exception region demarcation boundary in each frame of image, performs image segmentation operation on each exception region, extracts the boundary contour line segment, the contour extraction is based on the connectivity determination of image gray gradient calculation and edge pixel, obtains the closed or semi-closed boundary contour point set, assuming that the number of boundary points extracted from a certain exception region is 120, the continuous point column is constructed in the order distribution of the contour points in the image coordinate system, the curvature of any three points in the point column is calculated, the curvature calculation method adopted is the reciprocal of the circumcircle radius of the circular arc composed of three points, assuming that the curvature values of five representative points are [0.15, 0.20, 0.23, 0.18, 0.21], the normalized processing is performed on it, the normalized reference value is set to 0.25, which is the maximum historical value of the boundary curvature of the conventional crack of the pier concrete surface, which is derived from the 95% upper limit of the manual annotation curvature statistics in the underwater image data in the past three years, the normalized curvature values are [0.6, 0.8, 0.92, 0.72, 0.84] respectively, the average value is C zr =0.776, then the area ratio of the exception region is calculated, the total number of pixels in the region is set to 3200 pixels, the resolution of the whole image is 1920*1080, the total number of pixels is 2073600, then the area ratio is A zr =3200 /

[0099] 2073600≈0.00154, the geometric center of the boundary point set in the region is further extracted as the center point coordinate, for example, the horizontal pixel range is (640, 704), and the vertical range is (1280, 1344), then the center coordinates are (672, 1312), finally the damage structure parameter set is formed, including the boundary curvature normalized value C zr =0.776, the area ratio A zr =0.00154, the center position coordinates (672, 1312).

[0100] The coincidence degree calculation submodule calls the spatial boundary data of the shear concentrated region, the axial pressure change region and the bending resistance bearing region of the pier according to the damage structure parameter set, calculates the spatial overlapping range of the damage region and the key region, calculates the overlapping area and the total area of the corresponding key region by ratio, and obtains the damage region coincidence degree value;

[0101] The overlap calculation submodule calls the center coordinates and boundary range information of the above-mentioned damaged structural parameters, and reads the boundary data of the key areas defined in the pier structure model, including the two-dimensional image mapping bounding boxes of the shear force concentration area, the axial pressure variation area, and the bending bearing area. Let the boundary of the shear force concentration area in the current image coordinate system be a rectangular area [(600,1200)–(800,1400)], with an area of ​​400×200=80000 pixels. The current boundary range of the damaged area is [(640,1280)–

[0102] [704, 1344], with an area of ​​64 × 64 = 4096 pixels, and the coordinate range of the overlapping area is [(640, 1280) –

[0103] [704,1344], with an area of ​​64×64=4096 pixels, calculate the ratio of the overlapping area to the key region area as R. zr =4096 / 80000=0.0512, that is, the overlap is 5.12%. If the current damaged area crosses multiple key areas at the same time, the overlap ratio with each area is calculated separately and the maximum value is extracted as the final overlap value. This value represents the degree of spatial influence of the current abnormal area on the key structural area.

[0104] The severity assessment submodule calls upon the overlap values ​​of the damaged regions and the set of damage structure parameters to extract the boundary curvature, area ratio, spatial overlap, and corresponding damage type of the damaged regions, using the following formula:

[0105]

[0106] The severity of damage to the affected area is calculated, and damage assessment information is generated.

[0107] Among them, S zr C represents the severity of the injury. zr A represents the normalized value of the boundary curvature of the damaged region. zr R represents the area percentage of the damaged region. zr K represents the overlap value between the damaged area and the critical area. zr D represents the overlap sensitivity benchmark value corresponding to the damage area type. zr Indicates the influence coefficient of damage type;

[0108] The severity assessment submodule calls the aforementioned set of damage structure parameters and the overlap value of the damage area to extract parameter C. zr =0.776, A zr =0.00154, R zr =0.0512, and call the damage type influence coefficient D zr, and the type influence coefficient is set as 1.8, which is based on the structure experiment that longitudinal cracks have more serious influence on the bearing capacity of the axial compression member than transverse cracks, and is obtained by fitting the damage level distribution and the structure instability critical value. 1.8 is the standard value of the influence intensity of longitudinal cracks, and the coincidence degree sensitive reference value K zr is set as 0.035, which is based on the fact that if the coincidence area of longitudinal cracks and shear concentration area exceeds 3.5% of the total area, it has a significant impact on the safety of the structure, and this value is derived from the critical coincidence ratio of the ultimate bearing model of the shear compression member and the local force simulation of the structure. Substituting the formula

[0109]

[0110] The damage severity value S zr ≈0.002115, which is the quantitative expression result of the damage severity of the current area, and the small value indicates that the current damage area has cracks and falls into the key structure area, but the influence area is small. The usefulness of the formula is that by introducing the product of the normalized boundary curvature and the area ratio to construct the damage scale evaluation basis, and by correcting the risk amplitude through the difference value of the coincidence degree, and by introducing the damage type coefficient as the strength weighted item, the final result can take into account the comprehensive influence of the damage form, position and type on the safety of the bridge pier structure, and improve the resolution and structure correlation of quantitative identification.

[0111] Please refer to Figure 6 , the bridge pier abnormality evaluation module comprises:

[0112] The high-risk area screening sub-module extracts the severity value of each damage area based on the damage evaluation information, compares each area with the set severity threshold value, screens the areas exceeding the threshold value, extracts the damage type label and spatial range of the corresponding area, and generates a high severity area set;

[0113] The high-risk area screening submodule calls the foregoing damage assessment information, extracts the severity value of each damage area item by item, supposes that there are 5 damage areas in the current image, and the severity values thereof are 0.0021, 0.0054, 0.0098, 0.0009 and 0.0072, supposes that the severity threshold value is 0.005, the threshold value is set according to the standard crack warning value of the pier concrete structure and the damage tolerance of the underwater pressure-bearing area, and is set by selecting the 75th percentile value of the corresponding safety boundary in the 600 groups of structure failure samples in the early stage to ensure the convergence and stability of the risk boundary, and the severity values of the areas are compared item by item, and the severity value greater than 0.005 is determined as a high-risk area, and the current 2nd, 3rd and 5th items are 0.0054, 0.0098 and 0.0072, all of which are greater than the threshold value, and are screened as high-risk areas, and the damage types corresponding to the three items are marked as “longitudinal crack”, “peeling” and “hollow”, respectively, and the spatial boundary coordinates thereof are [(640, 1280)–(704, 1344)], [(880, 960)–(944, 1024)] and [(1200, 1500)–(1264, 1564)], respectively, and are stored in the high-severity area set in turn.

[0114] The distribution parameter extraction submodule calls the high-severity area set, counts the number of areas of each type of damage, extracts the centroid coordinate set of each area, and calculates the density trend value of the centroid distribution to obtain the damage distribution feature quantity;

[0115] The distribution parameter extraction submodule counts the number of areas of each type of damage based on the high-severity area set, supposes that there are 3 types in the current high-risk area set, among which there are 2 “longitudinal cracks”, 1 “peeling” and 1 “hollow”, calculates the total number and labels, then extracts the centroid coordinates of each area, the centroid coordinates are obtained by taking the median value of the boundary coordinates, for example, the “longitudinal crack” region center points are (672, 1312) and (1340, 980), and the centroid coordinate set thereof is stored as an array [(672, 1312), (1340, 980)], the density trend value of each type of centroid coordinate set is calculated, and the density trend value is defined as the ratio of the standard deviation to the mean of the shortest distance between the centroids, that is, D var =std(d ij ) / mean(d ij ), taking the two points of “longitudinal crack” as an example, the Euclidean distance is calculated as Since it is a single distance, the standard deviation is 0, and the density trend is 0, indicating that the concentration of this type of damage is high, if there are more centroid points, the average and standard deviation of the distance between two points are calculated in turn, and finally the number, centroid coordinates and density trend value of each type are integrated into the damage distribution feature quantity.

[0116] The influence degree calculation submodule calls the damage distribution parameter set, extracts the number, distribution density and concentration trend coefficient of each type of damage, combines the screening area set, and uses the formula:

[0117]

[0118] The abnormal influence degree of the pier is obtained by operation, and the pier abnormality evaluation result is obtained;

[0119] Where, I b represents the abnormal influence degree of the pier, N as,k represents the number of the kth damage area, D as,k represents the distribution density of the kth damage area, T as,k represents the concentration trend coefficient of the kth damage area, K represents the total number of damage types, and S as,k is the average value of the damage severity of the kth damage area.

[0120] The influence degree calculation submodule calls the aforementioned distribution parameter set, extracts the area number N as,k , the distribution density D as,k , the concentration trend coefficient T as,k and the average severity value S as,k of each type of damage, assuming that there are K=3 types of damage at present, and the specific values are as follows: longitudinal crack type: number N as,1 =2, density D as,1 =0.0025, trend coefficient T as,1 =0.1, and severity average S as,1 =0.00575; spalling type: number N as,2 =1, density D as,2 =0.0018, trend coefficient T as,2 =0.25, and severity average S as,2 =0.0098; hollow type: number N as,3 =1, density D as,3 =0.0016, trend coefficient T as,3 =0.32, and severity average S as,3 =0.0072, which are brought into the formula:

[0121]

[0122] The calculation is expanded as follows:

[0123]

[0124] Finally, the abnormal influence degree value I b≈0.00000815, which shows that the overall influence range of the current high-risk area of the pier is small, mainly caused by individual high severity damage and local aggregation characteristics. The usefulness of the formula is that the damage aggregation effect weight is adjusted by introducing the square root of the trend coefficient, and the density, quantity and average severity value are coupled to calculate, so that the evaluation results consider the damage spatial distribution, strength and category difference.

[0125] The above merely describes the preferred embodiments of the present application, and is not intended to limit the present application in other forms. Any skilled person in the art can modify or change the above disclosed technical content to equivalent embodiments applied to other fields, but any simple modification, equivalent change and modification made to the above embodiments without departing from the technical solution content of the present application, according to the technical essence of the present application, still belongs to the protection scope of the technical solution of the present application.

Claims

1. A bridge pier underwater structure inspection system based on an unmanned underwater robot, characterized in that, The system comprises: The inspection cycle adjustment module obtains bridge traffic flow data, calculates the difference between the cumulative traffic volume in the peak period and the cumulative traffic volume in the valley period, calls the water flow shear force parameter, calculates the difference between the shear force increment value and the safety critical shear force increment threshold value, adjusts the pier inspection cycle, and generates a dynamic inspection cycle adjustment result; The inspection cycle adjustment module comprises: The traffic flow calculation submodule obtains bridge traffic flow data, divides the time interval according to the peak period and the valley period, calculates the difference between the cumulative traffic volume in the peak period and the cumulative traffic volume in the valley period, and obtains the peak-valley traffic difference; The shear force increment calculation submodule calls the peak-valley traffic difference, obtains the water flow shear force parameter, calculates the difference between the current shear force increment value and the safety critical shear force increment threshold value according to the safety critical shear force increment threshold value, combines the preset pier inspection cycle and the peak-valley traffic difference, and uses the formula: ； to obtain the cycle adjustment coefficient; wherein, represents a cycle adjustment coefficient, represents a normalized value of a peak period cumulative traffic volume, represents a normalized value of a valley period cumulative traffic volume, wherein represents a normalized value of a water flow velocity change amount in a time period, represents a safety critical shear force increment threshold value, represents a normalized value of a current shear force increment value, represents a normalized value of a shear force reference reference value, represents a total number of time periods in which a water flow velocity change amount participates in calculation; The cycle adjustment output submodule combines the preset pier inspection cycle according to the cycle adjustment coefficient, judges the cycle adjustment direction, calculates the adjusted cycle value, adjusts the pier inspection cycle, and generates a dynamic inspection cycle adjustment result; The anomaly recognition module uses an unmanned underwater robot to perform pier underwater structure inspection based on the dynamic inspection cycle adjustment result, obtains underwater inspection images, identifies abnormal areas by comparing known pier abnormal images, marks the abnormal areas, and generates an abnormal area calibration result; The anomaly recognition module comprises: The underwater image acquisition submodule uses an unmanned underwater robot to perform pier underwater structure inspection within a specified time period based on the dynamic inspection cycle adjustment result, collects a continuous underwater image frame sequence of the pier bottom structure area, establishes an image sequence library, and obtains a pier underwater image sequence; The feature matching degree calculation submodule calls the pier underwater image sequence, combines the known pier abnormal image, obtains the detection area gray mean value, texture direction mean value, and boundary curvature sequence, compares them with the reference area gray mean value, texture direction mean value, and boundary curvature sequence, and uses the formula: ； to obtain the abnormal area matching degree value; wherein, a normalized value representing the abnormal region matching degree, a normalized value representing the detection region gray mean value, a normalized value representing the reference region gray mean value, a normalized value representing the detection region texture direction mean value, a normalized value representing the reference region texture direction mean value, a normalized value representing the detection region first boundary curvature value, a normalized value representing the reference region first boundary curvature value, a normalized value representing the boundary curvature sampling point number; The area marking output submodule compares the abnormal area matching degree value with the preset matching degree threshold value, identifies the area with a matching degree value greater than the threshold value, marks the abnormal area position and boundary range according to the detection area center coordinates, and generates an abnormal area calibration result; The automatic retake module calls the abnormal area calibration result, extracts the pixel noise index and the definition index, evaluates the shooting quality of the abnormal area, adjusts the light source brightness and the exposure time if the shooting quality is lower than the set quality threshold value, and performs retake in situ to generate a local abnormal retake image; The automatic retake module comprises: The image block extraction submodule calls the abnormal area calibration result, divides the abnormal area into a pixel grid, generates a fixed-size image block set according to the pixel grid coordinates, extracts the image block gray matrix, calculates the image block gray variance value and the mean gradient value, and obtains an image basic parameter group; The quality evaluation submodule compares the gray scale variance value with the image definition reference value and judges the difference between the mean gradient value and the pixel noise reference value according to the image basic parameter set, and grades the image block in combination with the two judgment results, labels the quality grading label for each image block, and generates an image quality grading label set; The image re-shooting submodule calls the image quality grading label set, filters the image block labeled with a low quality grade, extracts the spatial position information of the filtered image block, adjusts the exposure time and light source brightness parameters of the unmanned underwater vehicle shooting device according to the position coordinates, performs in-situ re-shooting, and generates a local anomaly re-shooting image; The damage evaluation module calls the local anomaly re-shooting image, extracts the damage center position, compares it with the key area of the pier, calculates the damage area overlap degree, combines the damage type, calculates the damage severity, and generates damage evaluation information; The damage evaluation module comprises: The structural parameter extraction submodule divides each damage area in the image based on the local anomaly re-shooting image, extracts the damage area boundary line, calculates the boundary curvature value, counts the pixel proportion of the damage area in the whole image, and identifies the center position coordinates of the damage area to obtain a damage structure parameter set; The overlap degree calculation submodule calls the spatial boundary data of the shear concentrated area, the axial pressure change area and the bending resistance bearing area of the pier according to the damage structure parameter set, calculates the spatial overlapping range of the damage area and the key area, calculates the ratio of the overlapping area to the total area of the corresponding key area, and obtains the damage area overlap degree value; The severity evaluation submodule calls the damage area overlap degree value and the damage structure parameter set, extracts the damage area boundary curvature, area proportion, spatial overlap degree and corresponding damage type, and uses the formula: to obtain the damage severity value of the damage area, and generates damage evaluation information; ； The patrol cycle dynamic adjustment result comprises a cycle adjustment coefficient, a flow difference threshold trigger state and a water flow shear force increment state, the abnormal area calibration result specifically comprises an abnormal boundary position parameter, an abnormal shape judgment identifier and an abnormal area center point coordinate, the local anomaly re-shooting image comprises a re-shooting light source brightness level, a re-shooting exposure time setting value and a re-shooting area imaging data set, and the damage evaluation information specifically comprises a damage type weight value, a sensitive area overlap coefficient and a comprehensive influence path coefficient. wherein, represents a normalized value of the boundary curvature of the damage region, represents a normalized value of the boundary curvature of the damage region, represents an area ratio of the damage region, represents a coincidence value of the damage region and the key region, represents a coincidence-sensitive reference value corresponding to the damage region type, represents a damage type influence coefficient.

2. The unmanned underwater vehicle-based pier underwater structure inspection system according to claim 1, wherein, The system further comprises:

3. The unmanned underwater vehicle-based pier underwater structure inspection system of claim 1, wherein, The pier anomaly evaluation module calls the damage evaluation information, filters the area whose damage severity exceeds the set threshold, counts the damage type and distribution range, calculates the pier anomaly influence degree in combination with the number and concentration trend of the damage area, and generates a pier anomaly evaluation result; The pier anomaly evaluation result specifically refers to a damage distribution density, a total amount of damage and a damage concentration trend coefficient. The pier anomaly evaluation module comprises:

4. The unmanned underwater vehicle-based pier underwater structure inspection system of claim 3, wherein, The high-risk area screening submodule extracts the severity value of each damage area based on the damage evaluation information, compares each area with the set severity threshold, filters the area exceeding the threshold, extracts the damage type label and spatial range of the corresponding area, and generates a high-severity area set; ​ The distribution parameter extraction submodule calls the high-severity region set, counts the number of regions of each type of damage, extracts the centroid coordinate set of each region, and calculates the density change trend value of the centroid distribution to obtain damage distribution characteristic quantities; The influence degree calculation submodule calls the damage distribution parameter set, extracts the number, distribution density and concentration trend coefficient of each type of damage, combines the screening region set, and uses the formula: ； The operation obtains the abnormal influence degree of the pier, and obtains the pier abnormality evaluation result. wherein, represents the abnormal influence degree of the bridge pier, represents the number of the first damage area, represents the distribution density of the first damage area, represents the concentration tendency coefficient of the first damage area, represents the total number of damage types, is the average value of the damage severity of the first damage area.

Images