Bridge underwater disease detection system and method based on array imaging

Abstract

The invention relates to the technical field of bridge disease detection, in particular to a bridge underwater disease detection system and method based on array imaging, and the system is characterized in that in a disease detection registration module, "sonar-laser-polarization" three-mode independent calculation and double-reference constraint are carried out; by combining the layered coaxial integrated design of a mark point, namely an acoustic reflection unit and an optical diffuse reflection unit, high-precision initial registration of sonar coordinate calculation errors, laser ranging correction errors and polarization three-dimensional conversion errors is realized, and the problem of multi-modal data scale drift is solved; the bridge pier path scanning module realizes complete coverage of bridge pier curved surface disease detection through scenarized adaptation of a spiral-radial swing track and a variable pitch spiral track aiming at the contradiction between the fixed curvature of a detection ring and the dynamic change of the curvature of a bridge pier, and effectively solves the problem of mismatching of a coverage gap and a posture of a traditional fixed path.

Description

Technical Field

[0001] The present invention relates to the technical field of bridge defect detection, and more particularly to an array imaging-based underwater bridge defect detection system and method. Background Art

[0002] Underwater bridge defect detection is crucial to ensuring the safety of bridge structures. Existing technologies often use multimodal sensors such as sonar, lidar, and polarization cameras for joint detection. However, traditional multimodal detection systems have significant flaws in the initial registration process: the coordinate system solutions of the three modalities of sonar, laser, and polarization lack effective benchmark constraints, multimodal data are prone to scale drift, and the landmark design does not achieve a collaborative benchmark for acoustic and optical modalities. This results in high errors in sonar coordinate solution and laser ranging correction, which cannot meet the needs of submillimeter-level bridge defect detection. In addition, the spatiotemporal registration accuracy of multimodal data in existing technologies is insufficient, making it difficult to establish a unified spatial benchmark, which seriously affects the accuracy of subsequent defect detection.

[0003] Traditional methods face significant technical bottlenecks in defect feature extraction and path planning: single-modal feature extraction fails to account for complex environmental interference such as underwater turbulence and sand content, resulting in low sensitivity for identifying subtle defects and a lack of monitoring capabilities for their dynamic development. Path planning generally employs fixed trajectories, which are unable to adapt to the dynamic changes in the curvature of the pier surface. The difference between the fixed curvature of the detection ring and the curvature of the pier surface leads to coverage gaps. Mismatches in the sonar / laser field of view cause repeated or missed scans, resulting in low coverage integrity and difficulty meeting the requirements for full-scene underwater bridge inspection. These issues result in existing technologies exhibiting deficiencies in underwater bridge defect detection, such as low accuracy, incomplete coverage, and poor dynamic adaptability, necessitating an urgent need for innovative solutions. Summary of the Invention

[0004] In view of the shortcomings of the existing technology, the purpose of the present invention is to provide a system and method for detecting underwater bridge defects based on array imaging.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] The underwater bridge damage detection system based on array imaging includes:

[0007] The defect detection and registration module configures detection rings and performs initial spatial registration of multimodal data before performing defect detection on underwater bridge piers.

[0008] Modal feature enhancement module, used to enhance the defect sensitivity of multimodal features, extract inter-frame change rate and multi-frame sliding window processing, and fuse 8-dimensional features;

[0009] The pier path scanning module is used to adapt the defect detection path planning to the pier surface.

[0010] Furthermore, the configured detection ring is divided into a lower layer, a middle layer, and an upper layer. The sensor type of the lower layer is sonar, the sensor type of the middle layer is lidar, and the sensor type of the upper layer is a polarization camera.

[0011] Furthermore, the initial spatial registration of multimodal data includes spatiotemporal synchronization and multimodal registration of the detection ring, independent solution of the three modalities, and dual benchmark constraints.

[0012] Furthermore, the detection ring is subjected to spatiotemporal synchronization and multimodal registration: n marking points are arranged on the detection ring, and the n marking points are evenly distributed in a circular direction. Each marking point is a cylinder, and the interior adopts "acoustic reflection unit + optical diffuse reflection unit" layered coaxial integration.

[0013] Furthermore, the three-modal independent solutions include independent solutions of the sonar coordinate system, the laser coordinate system, and the polarization coordinate system;

[0014] Sonar coordinate system:

[0015] Algorithm formula: ; c is the underwater sound speed, is the sonar echo phase difference, is the sonar frequency, v is the turbulent relative velocity measured by the inertial measurement unit, The compensation logic is to use the turbulent velocity v measured in real time by the inertial measurement unit to correct the sound speed fluctuation and relative motion interference;

[0016] Laser coordinate system:

[0017] Optical path correction formula: ; is the speed of light in vacuum, The speed of light underwater is calculated by real-time monitoring of water temperature and salinity. is the original distance directly measured by the lidar, is the corrected laser ranging value;

[0018] Polarization coordinate system:

[0019] Conversion process: Extract polarization images of the marker point at three wavelengths: 450nm, 550nm, and 650nm. Calculate the DoP and AoP of each pixel in the 450nm polarization image. Select the pixels of the marker point that satisfy a DoP greater than 0.8 and an AoP fluctuation less than ±5°. Annotate the intersection of the selected pixels as the location of the marker point in the image, and output its pixel coordinates (u, v). These are converted to 3D coordinates using the pre-calibrated camera intrinsic parameter matrix and the 3D mapping model.

[0020] Furthermore, the defect sensitivity of multimodal features is enhanced, including: feature conversion of single mode, where single mode includes sonar mode, laser mode, and polarization mode.

[0021] Furthermore, path planning for adaptive disease detection on pier curved surfaces is developed, including scenario-based compensation trajectory for curvature differences and multi-sensor collaborative step optimization with field of view constraints.

[0022] Furthermore, the curvature difference scenario compensation trajectory is defined, and R is defined as the curvature radius of the bridge pier. To detect the ring curvature radius, Trajectory type: spiral-radial oscillation trajectory, algorithm logic:

[0023] Axial step length L: Based on the effective coverage width of the sonar ;Pick ;

[0024] Circumferential rotation angle : , ensuring continuous circumferential coverage;

[0025] Radial swing: After completing one spiral circle, it swings slightly along the radial direction of the pier to compensate for the inner gap;

[0026] when , trajectory type: variable pitch spiral trajectory, algorithm logic:

[0027] Axial step length L: increases with the increase of R, the formula is ;

[0028] Circumferential rotation angle : decreases as R increases, ensuring stable circumferential coverage density;

[0029] Attitude adjustment: Detects micro-rotations around its own axis.

[0030] Furthermore, by taking into account the differences in field of view of sonar, laser, and polarization cameras (sonar ±45°, laser ±15°, polarization ±30°), a unified step-size calculation model was designed to ensure multi-modal coverage coordination:

[0031] Effective coverage width calculation: , , ;

[0032] Uniform step size ;

[0033] Dynamic adjustment logic: every 10 seconds, the real-time curvature R of the bridge pier is obtained through SLAM mapping, and L and rotation angle are recalculated. , adapting to the local curvature changes of the pier surface;

[0034] Using the bridge pier point cloud map generated by sonar-laser fusion SLAM, we can detect coverage blind spots in real time and dynamically insert supplementary scanning paths:

[0035] Identification of missed scanning areas: Using octree point cloud density analysis, the areas with insufficient marking density are .

[0036] Local replanning algorithm: Generate the shortest fill-sweep subpath nearby, with the following constraints: ,q: the pose node of the current detection ring, representing the current position in the path planning; : Missed scanning area The feature pose node represents the target position that needs to be scanned; : Detection ring fixed curvature Current curvature of the pier The absolute difference, coefficient 0.3: weight factor of curvature difference;

[0037] Path insertion and optimization: Insert the supplementary sweep sub-path into the main path, and achieve smooth transition through B-spline interpolation to ensure the smooth movement of the detection ring;

[0038] Define multimodal coverage quality functions, monitor detection integrity in real time, and trigger rescans or path replanning: ; is the coverage area of the sonar, is the coverage area of the laser, is the coverage area ratio of the polarization camera;

[0039] set up ,when : Start local replanning and give priority to scanning the areas with poor multimodal coordination; when : Continue executing the main path to reduce computational overhead.

[0040] Furthermore, the method for detecting underwater bridge defects based on array imaging has the following steps:

[0041] Step 1: Detection ring configuration and multi-modal initial registration;

[0042] Step 2: Multimodal data processing and feature enhancement;

[0043] Step 3: Scanning path planning for pier surface adaptation.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] The system's disease detection and registration module uses independent calculation of the "sonar-laser-polarization" three modes and dual benchmark constraints, combined with the layered coaxial integrated design of the landmark "acoustic reflection unit + optical diffuse reflection unit", to achieve high-precision initial registration of sonar coordinate solution errors, laser ranging correction errors, and polarization three-dimensional conversion errors, solving the problem of multimodal data scale drift. The modal feature enhancement module extracts single-modal defect features such as phase difference, depth residual, and DoP anomaly, and combines the inter-frame change rate with the dynamic feature mining of the 20-frame sliding window. The 8-dimensional feature space is integrated to improve the sensitivity of underwater bridge pier defect identification. The pier path scanning module addresses the contradiction between the fixed curvature of the detection loop and the dynamic change of the pier curvature. Through scenario-based adaptation of "spiral-radial swing trajectory" and "variable pitch spiral trajectory", combined with sonar-laser field of view collaborative step-size optimization, SLAM-driven missed area completion and dynamic feedback of the multimodal coverage quality function Q≥0.95, it achieves complete coverage of pier curved surface defect detection, effectively solving the coverage gap and posture mismatch problem of traditional fixed paths.

[0046] The method of the present invention realizes accurate monitoring of underwater bridge defects in an environment with high underwater sand content and high turbulent flow rate through a technical closed loop of "registration-planning-feature enhancement". BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is the principle block diagram of the underwater bridge damage detection system based on array imaging;

[0048] Figure 2 This is the principle block diagram of the independent solution of three modes;

[0049] Figure 3 A flowchart for modeling curvature differences. DETAILED DESCRIPTION

[0050] Example 1: Reference Figures 1 to 3 The underwater bridge defect detection system based on array imaging includes a defect detection registration module, a modal feature enhancement module, and a pier path scanning module.

[0051] Detection rings are configured for underwater bridge piers. The diameter of the detection rings is adjusted according to the diameter of the underwater piers. The detection rings are divided into a lower layer (acoustic zone), a middle layer (optical zone), and an upper layer (polarization zone). The sensor type in the lower layer (acoustic zone) is a miniature multi-beam sonar, with eight groups, parameters of 500kHz, and a resolution of 0.1mm when the water depth is ≤10m. The sensor type in the middle layer (optical zone) is a 16-channel lidar with parameters of 905nm, 200Hz scanning frequency, and 0.1° angular resolution. The sensor type in the upper layer (polarization zone) is an 8-channel multi-band polarization camera with parameters of 450 / 550 / 650nm three bands and an 8-channel beam splitter prism design.

[0052] Perform spatiotemporal synchronization and multimodal registration on the detection ring: n markers are placed on the detection ring, with n being at least 3. The n markers are evenly distributed around the ring, for example, 3 markers are evenly distributed around the ring at 120°. Each marker is a cylinder with a layered coaxial integration of "acoustic reflection unit + optical diffuse reflection unit" inside, achieving "one device, dual-modal benchmark";

[0053] Acoustic reflection unit (adaptive sonar)

[0054] Structure: A 3mm diameter titanium alloy reflective sphere is embedded in the bottom (density 4.5g / cm³, acoustic impedance ≈ 3 times that of seawater, achieving efficient sound reflection), and the surface is processed into a 30° cone angle scattering surface (rather than a traditional spherical surface);

[0055] Function: Enables sonar (500kHz, beam angle ±45°) to receive stable echoes within a ±45° viewing angle (echo intensity standard deviation <5%, traditional spherical echo attenuation > 30% at the edge of the viewing angle).

[0056] Optical diffuse reflection unit (compatible with laser / polarization camera)

[0057] Core structure: The top is integrated with a 1mm diameter alumina ceramic bull's eye (seawater corrosion resistant, hardness HRA>90), and the surface adopts a "dual-band functional film" design:

[0058] 550nm high-reflective film: Evaporated silicon dioxide-titanium oxide multilayer film, with a reflectivity of >95% for the 905nm band of lidar;

[0059] 450nm fluorescent coating: A phosphate material doped with europium ions absorbs 450nm polarized light and then emits 590nm fluorescence, enhancing the uniqueness of polarization camera identification (DoP>0.8 in the 450nm band, which is significantly different from the DoP<0.3 of background scattering).

[0060] The disease detection registration module completes initial registration through independent calculation of the "sonar-laser-polarization" tri-modality and dual-reference constraints:

[0061] Sonar coordinate system (phase difference-Doppler joint algorithm):

[0062] Algorithm formula: ; c is the underwater sound speed, is the sonar echo phase difference, =500kHz is the sonar frequency, v is the turbulence relative velocity measured by the IMU (IMU is configured on the detection ring), To collect the time difference; compensation logic: using the turbulence velocity v measured in real time by the inertial measurement unit (IMU), correct the sound speed fluctuation and relative motion interference to make the sonar coordinate solution error <0.2mm;

[0063] Laser coordinate system (turbulent optical path model correction):

[0064] Optical path correction formula: ; is the speed of light in vacuum, The speed of light underwater is calculated by real-time monitoring of water temperature and salinity. is the original distance directly measured by the lidar, is the corrected laser ranging value;

[0065] Polarization coordinate system (multi-band polarization feature conversion):

[0066] Conversion process:

[0067] Polarization images of the marker point at three wavelengths: 450nm (fluorescence), 550nm (high reflection), and 650nm (background) are extracted; 450nm (fluorescence): the europium ion fluorescent coating on the surface of the marker point absorbs 450nm polarized light and emits 590nm fluorescence (Stokes shift), making the polarization signal of the marker point in the 450nm band unique (DoP>0.8); 550nm (high reflection): the silica-titanium oxide multilayer film of the marker point is highly reflective to the 905nm (near-infrared) laser radar and also has high reflection in the 550nm (visible light) band (to assist positioning and enhance robustness); 650nm (background): background noise is collected for differential processing (to subtract interference from water scattering and biological attachment).

[0068] Hardware support: Customized polarization camera with an 8-channel beamsplitter prism for simultaneous acquisition of three-band polarization images (each band contains 0° / 45° / 90° / 135° linear polarization + RCP / LCP circular polarization, for a total of 8 channels).

[0069] Pixel coordinates are located using the 450nm fluorescence DoP > 0.8 and polarization angle (AoP) fluctuation < ±5° features. Positioning logic: Only pixels that simultaneously meet DoP > 0.8 and AoP fluctuation < ±5° are retained, identified as landmark areas, and their pixel coordinates (u, v) are output.

[0070] The pre-calibrated camera intrinsic parameter matrix + 3D mapping model is converted into 3D coordinates with an error of <0.3mm.

[0071] Improved ICP algorithm (turbulence robust version):

[0072] To address sensor jitter caused by turbulence, the traditional ICP algorithm has been improved with two key features: scale constraint and motion compensation.

[0073] Scale constraint (introduced by geometric prior):

[0074] Define the sonar-laser scale mapping relationship: sonar wavelength λ = 0.3mm (500kHz), laser wavelength λ = 905nm, and establish the scale conversion matrix: ;

[0075] This scale relationship is enforced during the registration process to avoid scale drift of multimodal data (traditional ICP is prone to 10%-20% scale deviation).

[0076] Motion compensation (turbulence vector pre-correction):

[0077] Extracting turbulent motion vectors from IMU composite timestamps , pre-compensate the ICP's "closest point search range": ; (The traditional search radius is fixed at 10mm, which is prone to missing the real corresponding points);

[0078] Modal feature enhancement module, which enhances the defect sensitivity of multimodal features;

[0079] Modality: Sonar, Original Features: 3D Point Cloud , defect sensitive feature conversion: phase difference is often : , >10°, the physical meaning is the phase distortion of sound propagation (cracks / corrosion cause changes in acoustic impedance); : The sonar echo phase value of the current detection point, : The “healthy benchmark phase” of the same area (such as the historical frame phase of the non-defective area of the pier, or the phase fitted by the non-defective points in the neighborhood) is used as the “non-defective reference value”;

[0080] Modality: LiDAR, Original feature: Depth map (z), Defect-sensitive feature conversion: Depth residual : , cracks >0.2mm, the physical meaning is the sudden change of surface depth (geometric characteristics of cracks / scour); : The depth value of the current point measured by the laser radar (unit: mm), reflecting the distance from the pier surface to the detection ring. : "Local healthy surface" fitted by RANSAC algorithm (using the neighborhood disease-free points to fit continuous planes / surfaces to simulate the disease-free geometry);

[0081] Modality: Polarization camera, Original features: DoP / AoP / R / B, Defect-sensitive feature conversion: DoP anomaly : , diseased area , the physical meaning is the polarization degree mutation (material anisotropy of biological attachment / corrosion); : Degree of polarization of the current point (0≤DoP≤1, reflecting the degree of polarization of light. The higher the DoP, the more significant the polarization characteristics). : background polarization value of the same area (such as the mean DoP value of the disease-free area, or the background value extracted by Gaussian filtering);

[0082] Combined with the dynamic changes of features caused by underwater water flow disturbance (flow velocity ≤ 0.5m / s), we can extract:

[0083] Inter-frame rate of change: such as the time derivative of the sonar phase difference ( ), the time derivative of the laser depth residual ( ), highlighting the dynamic development of diseases (such as crack expansion and intensified scour);

[0084] Multi-frame sliding window: 20 consecutive frames of data are spliced together in time and space to form a space-time cube (T=20, X=100, Y=100), preserving the dynamic evolution information of the disease.

[0085] Fusion of 8-dimensional features including geometry (x, y, z), acoustic phase (Δφ), optical depth (Δz), and polarization (ΔDoP, AoP, R / B);

[0086] Bridge pier path scanning module, bridge pier surface adaptation path planning;

[0087] When the curvature of the detection ring is fixed (set to , such as the design value of 3m), and the curvature of the pier When things change dynamically, we face the following challenges:

[0088] Coverage gap: There is a curvature difference between the detection ring and the pier surface ( ), resulting in the sensor field of view not being able to fully fit, forming a scanning blind area;

[0089] Attitude mismatch: Under fixed curvature, the projection area of the detection ring's sensor field of view (e.g., ±45° for sonar and ±15° for laser) on the pier surface changes with R. Traditional equidistant paths are prone to repeated scanning or missed scanning.

[0090] Lack of dynamic adjustment: Curvature differences cannot be compensated through hardware deformation, and must rely entirely on the intelligent adaptation of the path planning algorithm.

[0091] Curvature difference modeling and trajectory generation strategy:

[0092] against The resulting coverage gap (R is the curvature radius of the pier, To detect the ring curvature radius, design a "spiral offset" or "variable pitch spiral" trajectory and compensate for the curvature difference by adjusting the translation / rotation posture:

[0093] Scenario 1: The bridge pier is more curved ( ,like ):

[0094] Trajectory type: spiral-radial oscillating trajectory.

[0095] Algorithm logic:

[0096] Axial step length L: Based on the effective coverage width of the sonar ; (d is the distance between the detection ring and the bridge pier, is the sonar beam angle), take , ensuring a 30% overlap rate;

[0097] Circumferential rotation angle : , ensuring continuous circumferential coverage;

[0098] Radial swing: After completing one spiral, the pier swings slightly along the radial direction ( ), compensate for the inner gap (the curvature of the detection ring is gentler, and the inner side of the pier is easily missed).

[0099] Scenario 2: The bridge pier is flatter ( ,like ):

[0100] Trajectory type: variable pitch spiral trajectory.

[0101] Algorithm logic:

[0102] Axial step length L: increases with the increase of R ( ), the formula is ;

[0103] Circumferential rotation angle :Decreases as R increases( ), ensure that the circumferential coverage density is stable (such as maintaining 1 scanning point per degree in the circumference);

[0104] Attitude adjustment: Detects micro-rotation around its own axis ( ), so that the sector-shaped scanning surface of the lidar can better fit the curved surface of the bridge pier.

[0105] Multi-sensor collaborative step-size optimization with field-of-view constraints: Considering the field-of-view differences among sonar, laser, and polarization cameras (sonar ±45°, laser ±15°, and polarization ±30°), a unified step-size calculation model is designed to ensure multi-modal coverage coordination.

[0106] Calculation of effective coverage width (based on sonar, take the minimum value): , , ; uniform step size , ensuring that even the narrowest field of view sensor can provide continuous coverage.

[0107] Dynamic adjustment logic: every 10 seconds, the real-time curvature R of the bridge pier is obtained through SLAM mapping and recalculated. and rotation angle , adapting to local curvature changes on the pier surface (such as sudden curvature changes caused by scour pits).

[0108] Using the bridge pier point cloud map generated by sonar-laser fusion SLAM, real-time detection of coverage blind areas (point cloud density If it is determined to be a missed scan), a supplementary scan path is dynamically inserted.

[0109] Identification of missed scanning areas: Using octree point cloud density analysis, the areas with insufficient marking density are .

[0110] Local replanning algorithm (improved RRT): Generate the shortest fill-sweep subpath nearby, with the following constraints: , (Preferably select the scan path with small curvature difference and short path to avoid excessive detours). q: The pose node of the current detection ring (including three-dimensional coordinates x, y, z and attitude angles roll, pitch, yaw), representing the current position in the path planning; : Missed scanning area The characteristic pose node (usually the center point or boundary point of the missed scan area) represents the target position that needs to be scanned; : Detection ring fixed curvature Current curvature of the pier The absolute difference of ,coefficient 0.3: weighting factor of curvature difference, verified by experiments (when the weight is > 0.5, the path is overly detour, and when it is < 0.2, the curvature difference is undercompensated).

[0111] Path insertion and optimization: Insert the supplementary scanning sub-path into the main path, and achieve smooth transition through B-spline interpolation to ensure smooth movement of the detection ring (avoid sensor jitter caused by sudden stops and turns).

[0112] Define multimodal coverage quality functions, monitor detection integrity in real time, and trigger rescans or path replanning: ; is the coverage area of the sonar, is the coverage area of the laser, is the coverage area ratio of the polarization camera.

[0113] set up (can be set to 0.95), when : Start local replanning and give priority to rescanning areas with poor multimodal coordination (such as areas covered only by sonar but not by laser / polarization); when : Continue executing the main path to reduce computational overhead.

[0114] In the defect detection registration module of the above system, the "sonar-laser-polarization" three-modal independent solution and dual-reference constraint, combined with the layered coaxial integrated design of the landmark "acoustic reflection unit + optical diffuse reflection unit", achieves high-precision initial registration of sonar coordinate solution error, laser ranging correction error, and polarization three-dimensional conversion error, and solves the problem of multi-modal data scale drift. The modal feature enhancement module extracts single-modal defect features such as phase difference, depth residual, DoP anomaly, and combines the dynamic feature mining of inter-frame change rate and 20-frame sliding window. The 8-dimensional feature space is integrated to improve the sensitivity of bridge pier disease identification for underwater bridges. The pier path scanning module addresses the contradiction between the fixed curvature of the detection ring and the dynamic change of the pier curvature. Through the scenario adaptation of "spiral-radial swing trajectory" and "variable pitch spiral trajectory", combined with sonar-laser field of view collaborative step optimization, SLAM-driven missed area completion and dynamic feedback of multimodal coverage quality function Q≥0.95, it achieves complete coverage of bridge pier curved surface disease detection, effectively solving the coverage gap and posture mismatch problems of traditional fixed paths.

[0115] Example 2: A method for detecting underwater bridge defects based on array imaging, comprising the following steps:

[0116] Step 1: Detection ring configuration and multi-modal initial registration;

[0117] Step 2: Multimodal data processing and feature enhancement;

[0118] Step 3: Scanning path planning for pier surface adaptation.

[0119] The above method achieves accurate monitoring of underwater bridge defects in environments with high underwater sand content and high turbulent flow rate through the technical closed loop of "registration-planning-feature enhancement".

[0120] The above formulas are all dimensionless and numerically calculated, and the preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0121] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0122] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0123] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0124] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0125] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0126] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0127] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. The bridge underwater disease detection system based on array imaging is characterized by: include The defect detection and registration module configures detection rings and performs initial spatial registration of multimodal data before performing defect detection on underwater bridge piers. Modal feature enhancement module, used to enhance the defect sensitivity of multimodal features, extract inter-frame change rate and multi-frame sliding window processing, and fuse 8-dimensional features; The pier path scanning module is used to adapt the defect detection path planning to the pier surface.

2. The array imaging-based underwater bridge disease detection system according to claim 1 is characterized in that: The configured detection ring is divided into lower, middle and upper layers. The sensor type of the lower layer is sonar, the sensor type of the middle layer is lidar, and the sensor type of the upper layer is polarization camera.

3. The array imaging-based underwater bridge disease detection system according to claim 2 is characterized in that: Initial spatial registration of multimodal data, including spatiotemporal synchronization of detection rings and multimodal registration, independent solution of three modalities, and dual benchmark constraints.

4. The array imaging-based underwater bridge disease detection system according to claim 3 is characterized in that: Perform spatiotemporal synchronization and multimodal registration on the detection ring: Perform spatiotemporal synchronization and multimodal registration on the detection ring: Arrange n marking points on the detection ring, and the n marking points are evenly distributed in the ring. Each marking point is a cylinder, and the "acoustic reflection unit + optical diffuse reflection unit" is layered and coaxially integrated inside.

5. The array imaging-based underwater bridge disease detection system according to claim 3 is characterized in that: The three-modal independent solutions include independent solutions of the sonar coordinate system, independent solutions of the laser coordinate system, and independent solutions of the polarization coordinate system. Sonar coordinate system: Algorithm formula: ; c is the underwater sound speed, is the sonar echo phase difference, is the sonar frequency, v is the turbulent relative velocity measured by the inertial measurement unit, The compensation logic is to use the turbulent velocity v measured in real time by the inertial measurement unit to correct the sound speed fluctuation and relative motion interference; Laser coordinate system: Optical path correction formula: ; is the speed of light in vacuum, The speed of light underwater is calculated by real-time monitoring of water temperature and salinity. is the original distance directly measured by the lidar, is the corrected laser ranging value; Polarization coordinate system: Conversion process: Extract polarization images of the marker point at three wavelengths: 450nm, 550nm, and 650nm. Calculate the DoP and AoP of each pixel in the 450nm polarization image. Select the marker points whose DoP is greater than 0.8 and whose AoP fluctuation is less than ±5°. Annotate the intersection of these selected pixels as the marker point's position in the image, and output its pixel coordinates (u, v). These are converted to 3D coordinates using the pre-calibrated camera intrinsic parameter matrix and a 3D mapping model.

6. The array imaging-based underwater bridge disease detection system according to claim 1, characterized in that: Enhance the defect sensitivity of multi-modal features, including: single-mode feature conversion, single mode includes sonar mode, laser mode, and polarization mode.

7. The array imaging-based underwater bridge disease detection system according to claim 1 is characterized in that: Path planning for defect detection adapted to pier surfaces includes scenario-based compensation trajectory for curvature differences and multi-sensor collaborative step optimization with field of view constraints.

8. The array imaging-based underwater bridge disease detection system according to claim 1, characterized in that: Curvature difference scenario compensation trajectory, define R as the curvature radius of the bridge pier, To detect the ring curvature radius, Trajectory type: spiral-radial oscillation trajectory, algorithm logic: Axial step length L: Based on the effective coverage width of the sonar ;Pick ; Circumferential rotation angle : , ensuring continuous circumferential coverage; Radial swing: After completing one spiral circle, it swings slightly along the radial direction of the pier to compensate for the inner gap; when , trajectory type: variable pitch spiral trajectory, algorithm logic: Axial step length L: increases with the increase of R, the formula is ; Circumferential rotation angle : decreases as R increases, ensuring stable circumferential coverage density; Attitude adjustment: Detects micro-rotations around its own axis.

9. The array imaging-based underwater bridge disease detection system according to claim 1, characterized in that: Taking into account the differences in field of view of sonar, laser, and polarization cameras, a unified step-size calculation model was designed to ensure multi-modal coverage coordination: Effective coverage width calculation: , , ; Uniform step size ; Using the bridge pier point cloud map generated by sonar-laser fusion SLAM, we can detect coverage blind spots in real time and dynamically insert supplementary scanning paths: Identification of missed scanning areas: Using octree point cloud density analysis, the areas with insufficient marking density are ; Local replanning algorithm: Generate the shortest sweep subpath nearby, with the following constraints: ,q: the pose node of the current detection ring, representing the current position in the path planning; :Missed scanning area The feature pose node represents the target position that needs to be scanned; : Detection ring fixed curvature Current curvature of the pier The absolute difference, coefficient 0.3: weight factor of curvature difference; Path insertion and optimization: Insert the supplementary sweep sub-path into the main path, and achieve smooth transition through B-spline interpolation to ensure the smooth movement of the detection ring; Define multimodal coverage quality functions, monitor detection integrity in real time, and trigger rescans or path replanning: ; is the coverage area of the sonar, is the coverage area of the laser, is the coverage area ratio of the polarization camera; set up ,when : Start local replanning and give priority to scanning the areas with poor multimodal coordination; when : Continue executing the main path.

10. A method for detecting underwater bridge defects based on array imaging, applied to the system for detecting underwater bridge defects based on array imaging according to any one of claims 1 to 9, characterized in that: Here are the steps: Step 1: Detection ring configuration and multi-modal initial registration; Step 2: Multimodal data processing and feature enhancement; Step 3: Scanning path planning for pier surface adaptation.

Images