Displacement detection method and system for structural health monitoring

Abstract

The invention discloses a displacement detection method and system for structural health monitoring. The displacement detection method comprises the following steps: step 1, building a displacement detection system; step 2, carrying out image acquisition and preprocessing by using the displacement detection system; 3, performing light spot identification and center coordinate extraction on the preprocessed image; 4, performing error compensation on an output result in the step 3 to obtain compensated data; and 5, transmitting the compensated data to a cloud server, and carrying out structural health intelligent analysis. According to the invention, high-dynamic exposure control and image adaptive enhancement are supported; the system has the comprehensive anti-interference capability on strong light interference, shadow shielding, thermal drift, vibration disturbance and light spot distortion; a double-level error modeling mechanism is introduced into the filtering algorithm, and high-frequency jitter and low-frequency drift are effectively filtered out.

Description

Technical Field

[0001] The present invention relates to the field of structural health monitoring and intelligent image processing of optical measurements, and in particular to a displacement detection method and system for structural health monitoring, which is suitable for online monitoring and safety early warning of important structures such as bridges, dams, wind turbine towers, and high-rise buildings. Background Art

[0002] Currently, structural health monitoring (SHM) has gradually become a standard feature in infrastructure such as transportation, water conservancy, and energy. Existing mainstream methods such as fiber Bragg grating sensing, GPS measurement, laser ranging, and visual measurement have the following problems in practical applications:

[0003] It is difficult to achieve sub-millimeter accuracy;

[0004] Sensitive to environmental factors such as strong light, vibration, and temperature drift, and poor robustness;

[0005] Especially in visual measurement, insufficient accuracy in spot recognition and poor environmental adaptability are key bottlenecks restricting its practical application. Summary of the Invention

[0006] In view of the technical problems pointed out in the above background technology, the purpose of the present invention is to provide a displacement detection method and system for structural health monitoring.

[0007] To achieve the purpose of the present invention, the technical solution provided by the present invention is as follows:

[0008] First aspect

[0009] This application provides a displacement detection method for structural health monitoring, comprising the following steps:

[0010] Step 1: Build a displacement detection system, which includes multiple displacement detection units. Each displacement detection unit includes a laser emission module and a light spot processing module. The light spot processing module includes a whiteboard, a high-resolution camera, an acquisition terminal, and a mounting base.

[0011] Step 2: Using the displacement detection system to perform image acquisition and preprocessing, a high dynamic response image acquisition method capable of real-time adaptive adjustment based on rapid changes in ambient brightness, an adaptive Gaussian filtering noise reduction algorithm, and a CLAHE local contrast enhancement algorithm for enhancing the discernibility of the light spot against a non-ideal background are employed.

[0012] Step 3: Perform spot recognition and center coordinate extraction on the pre-processed image;

[0013] Step 4: Perform error compensation on the output result of step 3 to obtain compensated data;

[0014] Step 5: Transmit the compensated data to the cloud server and perform intelligent structural health analysis.

[0015] Among them, in step 2, the high dynamic response image acquisition method is specifically as follows: the high-resolution camera uses a built-in ambient light illumination sensor to obtain the light intensity value of the current acquisition point through continuous sampling, and introduces a parameter feedback adjustment mechanism based on the PID controller to dynamically control the key parameters of the camera.

[0016] Wherein, in step 2, the adaptive Gaussian filtering denoising algorithm evaluates the local complexity of the image through a structural tensor analysis method.

[0017] In step 2, the process of the CLAHE local contrast enhancement algorithm is as follows:

[0018] Step 2.1: Divide the image into small blocks;

[0019] Step 2.2: Perform histogram equalization on each block independently;

[0020] Step 2.3: Set the contrast limit factor to prevent small blocks from over-amplifying noise;

[0021] Step 2.4: Finally, perform bilinear interpolation fusion on the equalized image to form a global smooth result.

[0022] Wherein, the step 3 includes the following:

[0023] Step 3.1: Use adaptive threshold segmentation to divide the input grayscale image into blocks, and calculate the dynamic threshold of each local area based on its average brightness;

[0024] Step 3.2: After the initial extraction of the spot area contour, multiple rounds of contour feature filtering and geometric constraint enhancement are used;

[0025] Step 3.3: After preliminary identification of qualified light spots, perform sub-pixel light spot center positioning;

[0026] Step 3.4: Use the lightweight convolutional neural network coordinate fine-tuning module to perform sub-pixel coordinate fine-tuning on the preliminary recognition results.

[0027] In step 4, a two-level error compensation mechanism is adopted, and two complementary strategies are designed for short-term high-frequency noise and long-term slow-varying error drift respectively.

[0028] Wherein, the step 5 specifically includes the following:

[0029] Step 5.1: After error compensation is completed, the acquisition terminal packages the timestamp and displacement coordinate data and transmits them to the cloud server;

[0030] Step 5.2: The cloud server deploys a high-concurrency API interface service to receive the coordinate data stream uploaded by each node in real time. The received data is parsed and cached and then stored in a time series database for long-term storage.

[0031] At the same time, the front-end visualization platform uses the WebSocket protocol to achieve second-level refresh and dynamically display the following content:

[0032] Real-time two-dimensional coordinate displacement map: displays the motion trajectory of the light spot on the monitoring target surface;

[0033] Trend chart and fluctuation chart: plot the changing trend of xt, yt over a period of time to observe whether there are signs of periodic vibration or gradual displacement;

[0034] Structural heat map: When multiple monitoring nodes are deployed simultaneously, the relative displacement of each area is displayed through color gradients to achieve a visual structural deformation map;

[0035] Step 5.3: When one of the following alarm conditions is met, an alarm is issued:

[0036] (1) The absolute value of displacement exceeds the safety limit;

[0037] (2) the displacement slope increases abnormally within a certain time window;

[0038] When the above conditions are triggered, the following actions are automatically performed:

[0039] (1) Push alarm notification: real-time alarm via SMS, email or WeChat enterprise account;

[0040] (2) Start the video acquisition module: transmit the image and original data at that time for remote expert judgment;

[0041] (3) Interlocking control module: sends interlocking signals to the field control system.

[0042] Second aspect

[0043] The present application provides a displacement detection system for structural health monitoring, the displacement detection system comprising a plurality of displacement detection units, the displacement detection units comprising a laser emission module and a light spot processing module, the light spot processing module comprising a whiteboard, a high-resolution camera, an acquisition terminal, and a mounting base;

[0044] The displacement detection system performs the following displacement detection method, which specifically includes the following steps:

[0045] Step S1: performing image acquisition and preprocessing using the displacement detection system, wherein a high dynamic response image acquisition method capable of real-time adaptive adjustment according to rapid changes in ambient brightness, an adaptive Gaussian filtering noise reduction algorithm, and a CLAHE local contrast enhancement algorithm for enhancing the discernibility of light spots in non-ideal backgrounds are employed;

[0046] Step S2: performing spot recognition and center coordinate extraction on the pre-processed image;

[0047] Step S3: performing error compensation on the output result of step S2 to obtain compensated data;

[0048] Step S4: Transmit the compensated data to the cloud server and perform intelligent structural health analysis.

[0049] Among them, in step S1, the high dynamic response image acquisition method is specifically as follows: the high-resolution camera uses a built-in ambient light illumination sensor to obtain the light intensity value of the current acquisition point through continuous sampling, and introduces a parameter feedback adjustment mechanism based on the PID controller to dynamically control the key parameters of the camera.

[0050] Wherein, in step S1, the adaptive Gaussian filtering denoising algorithm evaluates the local complexity of the image through a structural tensor analysis method.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] 1. High-precision spot positioning capability

[0053] A composite center extraction algorithm (centroid method + Gaussian fitting + CNN fine-tuning) is used to achieve sub-pixel spot center positioning; the CNN model introduces an offset correction learning mechanism to deeply compensate for errors caused by boundary distortion and uneven illumination; the measured coordinate positioning error is as low as 0.2 to 0.4 pixels, and the accuracy is more than 60% better than traditional methods, supporting sub-millimeter displacement detection needs.

[0054] 2. Strong ability to adapt to the environment

[0055] It supports high-dynamic exposure control and adaptive image enhancement (structure tensor + CLAHE); the system has comprehensive anti-interference capabilities against strong light interference, shadow occlusion, thermal drift, vibration disturbance, and spot distortion; the filtering algorithm introduces a two-level error modeling mechanism (multi-window filtering + adaptive Kalman) to effectively filter out high-frequency jitter and low-frequency drift.

[0056] 3. Lightweight, modular, and highly integrated design

[0057] The computing module can be deployed on low-power edge devices such as Raspberry Pi / Jetson, with real-time inference latency less than 20ms; each functional module is independent (laser, acquisition, filtering, algorithm, transmission), which is conducive to maintenance and function expansion; and supports flexible networking and data transmission through USB, WiFi, 5G, and optical fiber.

[0058] 4. Cloud-based intelligent analysis and automated early warning mechanism

[0059] Combining cloud platform + local terminal to achieve dual redundant monitoring; data supports trend modeling, structural thermal map generation, and vibration spectrum analysis; safety warning supports rule setting, autonomous triggering, remote notification and linkage control. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 A schematic diagram of the structure of the displacement detection unit provided in this application;

[0061] In the figure, 1 is a laser emission module, 2 is a light spot processing module, 3 is a whiteboard, 4 is a high-resolution camera, and a collection terminal. DETAILED DESCRIPTION

[0062] The following is a clear and complete description of the technical solutions in the embodiments of the present invention, in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts are within the scope of protection of the present invention.

[0063] Existing laser spot center recognition algorithms often use simple centroid methods or image intensity weighting algorithms to extract the center. These methods are effective under ideal conditions, but in the presence of direct sunlight, noise interference, or complex backgrounds, spot center recognition is prone to deviation or failure, resulting in significantly increased displacement monitoring errors and making it difficult to meet the requirements of both high precision and high robustness.

[0064] To address the above issues, an embodiment of the present application provides a displacement detection method for structural health monitoring, comprising the following steps:

[0065] Step 1: Build a displacement detection system, which includes multiple displacement detection units. Each displacement detection unit includes a laser emission module 1 and a light spot processing module 2. The light spot processing module 2 includes a whiteboard 3, a high-resolution camera 4, an acquisition terminal 5, and a mounting base.

[0066] It should be noted that the detection unit mainly consists of two parts: the laser projection end and the receiving end:

[0067] Laser emission module: installed on a stable bracket, emitting a stable and collimated laser beam;

[0068] The light spot processing module includes: white board (diffuse reflection target surface), high-resolution camera, filter

[0069] chip module, anti-vibration and vibration reduction module, USB data transmission module, acquisition terminal, etc.

[0070] Both ends of the two modules are equipped with spirit levels to ensure that the projection axis is aligned with the imaging axis and improve measurement stability.

[0071] When the system is activated, the laser transmitter module generates a stable, collimated monochromatic laser beam and precisely projects it along the optical axis onto the target surface of the measurement area—a highly reflective whiteboard. The whiteboard, as a diffuse reflective medium, is treated with a diffuse reflective coating. Under laser illumination, a light spot pattern with clear boundaries and high intensity contrast is formed, effectively improving the signal-to-noise ratio (SNR) for subsequent image recognition. Once the light spot is formed on the whiteboard, it is captured by a high-resolution CMOS image acquisition device mounted within a specific angle and focal length range. This camera uses a bandpass filter to restrict the laser's wavelength (e.g., 650±10nm) to the sensor, thereby suppressing interference from ambient background light and enhancing the laser's signature information. Throughout the acquisition process, the image acquisition system scans the light spot on the whiteboard in real time, outputting a high-contrast image sequence. This provides a clear and stable input data source for subsequent light spot positioning, displacement calculation, and error analysis. The system also dynamically adjusts exposure time and gain to accommodate changes in external lighting, ensuring stable, high-quality image acquisition in both natural and bright outdoor conditions.

[0072] Step 2: Using the displacement detection system to perform image acquisition and preprocessing, a high dynamic response image acquisition method capable of real-time adaptive adjustment based on rapid changes in ambient brightness, an adaptive Gaussian filtering noise reduction algorithm, and a CLAHE local contrast enhancement algorithm for enhancing the discernibility of the light spot against a non-ideal background are employed.

[0073] In this system, the image acquisition and preprocessing stage is responsible for converting the optical information of the light spot into high-quality digital image data. It is a key front-end module in the entire light spot recognition and displacement extraction chain. Compared with traditional methods, we have made systematic improvements to the image acquisition hardware configuration, dynamic parameter control mechanism, and image enhancement process to adapt to changing environmental conditions and improve system robustness and accuracy.

[0074] The high dynamic response image acquisition method includes the following:

[0075] Traditional image acquisition systems operate under fixed parameter settings and are unable to make real-time adaptive adjustments based on rapid changes in ambient brightness, resulting in severe overexposure or underexposure, which in turn affects the accuracy of light spot recognition. Using a built-in ambient light sensor (with a wavelength range of 400-700nm), the system continuously samples the light intensity value L(t) at the current acquisition point and introduces a parameter feedback adjustment mechanism based on a PID (proportional-integral-differential) controller to dynamically control the key parameters of the camera:

[0076] Exposure time Te: automatically reduced in strong light to avoid image saturation;

[0077] Gain coefficient Ga: increases in low-light environments to improve image brightness;

[0078] White balance parameters: Real-time fine-tuning of the red and blue channel ratio to enhance the prominence of the red laser spot.

[0079] The PID regulator formula is as follows:

[0080]

[0081] Wherein, e(t)=Ltarget-L(t), and the target brightness is estimated by image entropy or mean grayscale.

[0082] To solve the problem of blurred edges in complex texture areas caused by traditional fixed kernel Gaussian filtering, we introduced the structural tensor analysis method to evaluate the local complexity of the image:

[0083] Calculate the local gradient tensor for the image I(x,y):

[0084]

[0085] Use the tensor eigenvalue to determine whether it is an edge / texture area and dynamically adjust the standard deviation σ of the Gaussian kernel;

[0086]

[0087] Among them, λmax(J) represents the main direction gradient strength of the local image.

[0088] This method can achieve an adaptive noise reduction effect with less blur in the edge area and enhanced smoothness in the texture area, which not only suppresses high-frequency noise but also retains the clear outline of the light spot edge.

[0089] To further enhance the discernibility of the light spot in a non-ideal background, the CLAHE (Contrast Limited Adaptive Histogram Equalization) technology is introduced to perform local contrast enhancement on the image. The algorithm process is as follows:

[0090] Split the image into small blocks (e.g. 8×8 or 16×16);

[0091] Perform histogram equalization on each block independently;

[0092] Set the contrast limit factor Climit to prevent excessive amplification of noise in small blocks;

[0093] Finally, bilinear interpolation fusion is performed on the equalized image to form a global smooth result.

[0094] The CLAHE algorithm can significantly improve the brightness difference between the light spot area and the background, enhancing the "detectability" of the laser point under background conditions such as strong light, dust, and dark walls.

[0095] Step 3: Perform spot recognition and center coordinate extraction on the pre-processed image;

[0096] In laser spot detection systems, accurately extracting the geometric center of the spot is a key step in displacement calculation. Traditional methods usually use the method of "image grayscale thresholding + contour extraction + geometric centroid calculation" to identify and locate the spot. Although this solution is simple to implement and has high computational efficiency, it has significant defects in complex field environments: the fixed threshold method is sensitive to changes in lighting, which can easily lead to the loss of spots in overexposed / underexposed areas of the image; when the background noise in the image (such as reflections and dust) is strong, non-spot areas may be misidentified; when the spot boundary is non-circular or distorted, the geometric centroid calculation deviates from the true center, increasing the measurement error.

[0097] To solve the above problems, this system has built a hierarchical, modular, and fusion-type light spot recognition solution, which integrates classic image algorithms and lightweight deep learning models to ensure sub-pixel center positioning in complex environments.

[0098] 1. Adaptive Gaussian Thresholding

[0099] The adaptiveThreshold method of OpenCV is used to divide the input grayscale image into blocks. Dynamic threshold calculation is performed on each local area based on its average brightness, effectively eliminating the segmentation failure problem caused by overall uneven illumination. The basic principle is as follows:

[0100] T(x,y)=μ N ( x,y) -C

[0101] Among them, μ N(x，y) represents the mean of the pixel neighborhood N, and C is the adjustment constant. This method ensures that the light spot can be effectively extracted under both bright and shadow conditions.

[0102] 2. Multi-round contour feature filtering and geometric constraint enhancement

[0103] To further eliminate false spots and highlight noise, this system designed a "three-dimensional comprehensive filtering algorithm" after initially extracting the outline of the spot area, and performed identification based on the following indicators:

[0104] Area A: The light spot should have a medium area in the image. If it is too small, it will be noise, and if it is too large, it will be overflow light.

[0105] Circularity:

[0106]

[0107] Where L is the perimeter of the contour. A circularity close to 1 indicates a nearly perfect circle. If Circularity < 0.7, it is excluded.

[0108] Edge gradient strength: The Sobel operator is used to evaluate edge sharpness. If the edge is too blurred or the gradient is discontinuous, it is considered an invalid contour.

[0109] 3. Sub-pixel light spot center positioning

[0110] After preliminary identification of qualified light spots, two methods are used to jointly extract the center coordinates:

[0111] Image geometric moment centroid method

[0112] Calculate image geometric moments based on regional image pixel intensities:

[0113]

[0114] Among them, M pq is the (p+q)-order moment of the image. This method is fast and effective and is applicable to light spots with regular boundaries.

[0115] Gaussian distribution fitting

[0116] When the light spot has a good shape and the light intensity distribution is approximately symmetrical, a two-dimensional Gaussian function model is used to fit the light spot to improve positioning accuracy:

[0117]

[0118] The fitting parameters (x0, y0) are the center of the spot with maximum intensity, achieving sub-pixel resolution. This algorithm is less robust to background interference and is suitable for scenes with pure images. It can be used in conjunction with the centroid method to compare and determine the most reliable value.

[0119] 4. Lightweight Convolutional Neural Network (CNN) Coordinate Fine-tuning Module

[0120] To further improve the accuracy of spot center coordinate recognition, especially in complex field conditions such as strong sunlight, dust, and reflections, this application designs and deploys a lightweight convolutional neural network (CNN) model based on the traditional centroid method and Gaussian fitting. This model is used to fine-tune the coordinates of the initial recognition results at the sub-pixel level. This module can be embedded in low-power devices (such as the Raspberry Pi 4B and Jetson Nano) and run in real time, balancing accuracy and speed.

[0121] The overall implementation of this module is divided into the following five logical steps:

[0122] (1) Model goal setting and function definition

[0123] The core task of this CNN model is to predict the offset (Δx, Δy) relative to the traditional recognition result based on the light spot area Patch intercepted in the image, and then accurately correct the results of the centroid method or Gaussian fitting. The final center coordinates are expressed as:

[0124] (x final ,y final )=(x raw +△x,y raw +△y)

[0125] This design effectively overcomes the problem of large recognition deviation in traditional methods under conditions of irregular spot edges and non-uniform lighting.

[0126] (2) Input data construction and preprocessing strategy

[0127] The network input is a 64×64 grayscale image patch, which is cropped from the original image with the initial center coordinate as the center. To improve the model's adaptability to various interference conditions, the following image enhancement operations are added to the training set:

[0128] Add Gaussian noise (σ = 5 to 20);

[0129] Perform brightness perturbation (±15%);

[0130] Implement affine transformations (rotation, scaling, translation);

[0131] All images are normalized to the range [0,1].

[0132] These enhancements help the model learn the "robust compensation capability" for factors such as light spot deformation, offset, and background noise.

[0133] (3) Model structure design and parameter configuration

[0134] The network adopts a concise and efficient seven-layer structure, which mainly includes two convolution-pooling combination layers, a flattening layer, and two fully connected layers. The structure of each layer is as follows:

[0135]

[0136]

[0137] The total number of parameters of the entire model is approximately 29,762, and the model file after training is less than 100KB, which is convenient for deployment on embedded devices.

[0138] (4) Training methods and optimization strategies

[0139] To improve model accuracy and generalization ability, the training process uses the following configuration:

[0140] Data source: constructed from real-life data and simulated images, all with manually labeled real-world center coordinates;

[0141] Loss function: Use mean square error (MSE) as the regression objective function:

[0142]

[0143] Optimizer: Adam, with an adaptive learning rate initially set to 1e-4;

[0144] Training mechanism: Using the Early Stopping strategy to prevent overfitting, 3,000 images were trained and 20 iterations were required to converge, ultimately achieving a residual error of less than 0.2px on the validation set.

[0145] (5) Embedded deployment performance and actual results

[0146] After model compression and acceleration optimization, the performance on edge devices is as follows:

[0147] Deployment platform: Raspberry Pi 5, Jetson Nano;

[0148] Inference time: Single-frame patch processing time is 13-18ms;

[0149] ·Precision improvement performance:

[0150] Traditional centroid residual: 0.6~1.1px

[0151] CNN fine-tuning residual: 0.2~0.4px

[0152] The average error is reduced by more than 60%.

[0153] Step 4: Perform error compensation on the output result of step 3 to obtain compensated data;

[0154] In actual operation, the positioning result of the laser spot center is inevitably affected by a variety of external and internal disturbance factors, including:

[0155] Slight vibration of the equipment (such as vibration of the bracket or ground);

[0156] Small mechanical deformations of the camera platform due to temperature fluctuations (thermal drift);

[0157] Image brightness fluctuations caused by drastic changes in external ambient lighting;

[0158] Sudden noise interference (such as fan airflow, dust obstruction).

[0159] To address these interference sources, this application proposes a two-level error compensation mechanism, designing two complementary strategies for short-term high-frequency noise and long-term slow-varying error drift:

[0160] 1. Multi-Window Adaptive Moving Average Filter: To improve the balance between real-time performance and smoothness, this system no longer uses a single-window averaging filter, but instead adopts a multi-window parallel evaluation mechanism. The specific approach is as follows:

[0161] At each moment t, the system uses different window lengths (n=3, 5, 7) to calculate the center coordinate sequence x of the current time series. t Perform a moving average:

[0162]

[0163] Calculate the local variance for the three window outputs:

[0164]

[0165] The window output with the smallest variance is selected as the current compensated coordinate value to avoid delayed response caused by an overly long window and enhance the adaptability to short-term interference.

[0166] 2. Improved Kalman Filtering (Adaptive Kalman Filtering with SensorFusion)

[0167] Traditional Kalman filtering assumes that process noise Q and observation noise R are constant, which makes it difficult to guarantee filtering accuracy in real nonlinear and non-stationary environments. To this end, this application integrates Kalman filtering with multi-source sensor input and introduces a dynamic noise estimation module to achieve adaptive adjustment.

[0168] Filtering process:

[0169] Prediction stage:

[0170]

[0171]

[0172] Update phase:

[0173]

[0174]

[0175]

[0176] in:

[0177] · Estimated coordinates of the current frame;

[0178] ·Z k : Current frame measurement coordinates (CNN / center of mass output);

[0179] ·P k : covariance matrix;

[0180] Q k 、R k : Process / measurement noise covariance, dynamic adjustment.

[0181] Noise dynamic estimation method:

[0182] The rate of change of light intensity (from the ambient light sensor) is R k Adjustment factor, increase R when high fluctuation k , suppressing observation instability caused by brightness fluctuations;

[0183] The accelerometer output mean variance (used to detect whether the system is in a vibration state) is used to dynamically update Q k , more accurately modeling system state changes.

[0184] Step 5: Transfer the compensated data to the cloud server and perform intelligent structural health analysis

[0185] (1) Edge-to-Cloud Transmission

[0186] The acquisition terminal (such as Raspberry Pi, Jetson Nano) has local caching and network communication capabilities. After error compensation is completed, the timestamp and displacement coordinate data are packaged and transmitted to the cloud through one of the following links:

[0187] 5G communication module: suitable for high-speed dynamic monitoring scenarios (such as bridge vibration and mobile platforms);

[0188] Wi-Fi LAN: Applicable to areas with network coverage, such as tunnels, dams, and factory areas;

[0189] Fiber-optic Ethernet: Suitable for fixed monitoring points, with high bandwidth and low latency performance.

[0190] The data packaging format uses the standard JSON format, and the content example is as follows:

[0191]

[0192] To ensure data integrity and security, TLS encryption is used and a retransmission mechanism is configured to prevent packet loss.

[0193] (2) Cloud data reception and visualization platform

[0194] The cloud server deploys high-concurrency API interface services (such as those based on Flask / FastAPI+Redis cache) to receive the coordinate data stream uploaded by each node in real time. The received data is parsed and cached and then stored in the

[0195] Sequential databases (InfluxDB, TimescaleDB) are used for long-term storage.

[0196] At the same time, the front-end visualization platform uses the WebSocket protocol to achieve second-level refresh and dynamically display the following content:

[0197] ·2D coordinate displacement map (real time): displays the motion trajectory of the light spot on the monitoring target surface;

[0198] Trend chart and fluctuation chart: plot x over a period of time t ,y t Change trends, observe whether there are signs such as periodic vibration, gradual displacement, etc.

[0199] Structural heat map (multi-point system): When multiple monitoring nodes are deployed simultaneously, the relative displacement of each area is displayed through color gradients, creating a visual map of structural deformation.

[0200] (3) Safety threshold alarm and linkage mechanism

[0201] The system allows users to customize alarm logic, such as setting the following threshold conditions:

[0202] The absolute value of displacement exceeds the safety limit (e.g. >3mm);

[0203] The displacement slope increases abnormally within a certain time window;

[0204] Once the above conditions are triggered, the system automatically performs the following operations:

[0205] Push alarm notification: real-time alarm via SMS, email or WeChat enterprise account;

[0206] Start the video acquisition module: transmit the image and original data at that time for remote expert judgment;

[0207] Linkage control module (optional): sends linkage signals to the on-site control system (such as stopping operation, closing valves, issuing sound and light alarms, etc.).

[0208] This application is applicable to the following scenarios: Bridge deformation measurement: Analyze the stress state of the bridge and assist in maintenance decision-making. Tunnel structure monitoring: Measure geological deformation and prevent landslide accidents. Dam settlement monitoring: Detect tiny displacements of the dam surface and provide early warning of crack risks. Mine surface deformation monitoring: Monitor surface collapse caused by mining. Wind turbine vibration monitoring: Analyze the stress condition of wind turbine blades and increase equipment life. High-rise building health monitoring: Long-term monitoring of building structure deformation to improve safety.

[0209] In addition, the present application provides a displacement detection system for structural health monitoring, the displacement detection system comprising a plurality of displacement detection units, the displacement detection units comprising a laser emission module and a light spot processing module, the light spot processing module comprising a whiteboard, a high-resolution camera, an acquisition terminal, and a mounting base;

[0210] The displacement detection system performs the following displacement detection method, which specifically includes the following steps:

[0211] Step S1: performing image acquisition and preprocessing using the displacement detection system, wherein a high dynamic response image acquisition method capable of real-time adaptive adjustment according to rapid changes in ambient brightness, an adaptive Gaussian filtering noise reduction algorithm, and a CLAHE local contrast enhancement algorithm for enhancing the discernibility of light spots in non-ideal backgrounds are employed;

[0212] Step S2: performing spot recognition and center coordinate extraction on the pre-processed image;

[0213] Step S3: performing error compensation on the output result of step S2 to obtain compensated data;

[0214] Step S4: Transmit the compensated data to the cloud server and perform intelligent structural health analysis.

[0215] Finally, it should be noted that the above embodiments are merely examples and illustrations of the present invention and are not intended to limit the present invention to the described embodiments. Furthermore, those skilled in the art will appreciate that the present invention is not limited to the above embodiments and that various variations and modifications may be made based on the teachings of the present invention, all of which fall within the scope of the present invention.

Claims

1. A displacement detection method for structural health monitoring, characterized in that: The steps include: Step 1: Build a displacement detection system, which includes multiple displacement detection units. Each displacement detection unit includes a laser emission module and a light spot processing module. The light spot processing module includes a whiteboard, a high-resolution camera, an acquisition terminal, and a mounting base. Step 2: Using the displacement detection system to perform image acquisition and preprocessing, a high dynamic response image acquisition method capable of real-time adaptive adjustment based on rapid changes in ambient brightness, an adaptive Gaussian filtering noise reduction algorithm, and a CLAHE local contrast enhancement algorithm for enhancing the discernibility of the light spot against a non-ideal background are employed. Step 3: Perform spot recognition and center coordinate extraction on the pre-processed image; Step 4: Perform error compensation on the output result of step 3 to obtain compensated data; Step 5: Transmit the compensated data to the cloud server and perform intelligent structural health analysis.

2. A displacement detection method for structural health monitoring according to claim 1, characterized in that: In step 2, the high dynamic response image acquisition method is specifically as follows: the high-resolution camera uses a built-in ambient light sensor to obtain the light intensity value of the current acquisition point through continuous sampling, and introduces a parameter feedback adjustment mechanism based on the PID controller to dynamically control the key parameters of the camera.

3. A displacement detection method for structural health monitoring according to claim 2, characterized in that: In step 2, the adaptive Gaussian filtering denoising algorithm evaluates the local complexity of the image through a structural tensor analysis method.

4. The displacement detection method for structural health monitoring according to claim 2, characterized in that: In step 2, the process of the CLAHE local contrast enhancement algorithm is as follows: Step 2.1: Divide the image into small blocks; Step 2.2: Perform histogram equalization on each block independently; Step 2.3: Set the contrast limit factor to prevent small blocks from over-amplifying noise; Step 2.4: Finally, perform bilinear interpolation fusion on the equalized image to form a global smooth result.

5. The displacement detection method for structural health monitoring according to claim 2, characterized in that: The step 3 includes the following: Step 3.1: Use adaptive threshold segmentation to divide the input grayscale image into blocks, and calculate the dynamic threshold of each local area based on its average brightness; Step 3.2: After the initial extraction of the spot area contour, multiple rounds of contour feature filtering and geometric constraint enhancement are used; Step 3.3: After preliminary identification of qualified light spots, perform sub-pixel light spot center positioning; Step 3.4: Use the lightweight convolutional neural network coordinate fine-tuning module to perform sub-pixel coordinate fine-tuning on the preliminary recognition results.

6. A displacement detection method for structural health monitoring according to claim 2, characterized in that: In step 4, a dual-level error compensation mechanism is adopted, and two complementary strategies are designed for short-term high-frequency noise and long-term slow-varying error drift respectively.

7. The displacement detection method for structural health monitoring according to claim 6, characterized in that: The step 5 specifically includes the following: Step 5.1: After error compensation is completed, the acquisition terminal packages the timestamp and displacement coordinate data and transmits them to the cloud server; Step 5.2: The cloud server deploys a high-concurrency API interface service to receive the coordinate data stream uploaded by each node in real time. The received data is parsed and cached and then stored in a time series database for long-term storage. At the same time, the front-end visualization platform uses the WebSocket protocol to achieve second-level refresh and dynamically display the following content: Real-time two-dimensional coordinate displacement map: displays the motion trajectory of the light spot on the monitoring target surface; Trend chart and fluctuation chart: plot the changing trend of xt and yt over a period of time to observe whether there are signs of periodic vibration or gradual displacement; Structural heat map: When multiple monitoring nodes are deployed simultaneously, the relative displacement of each area is displayed through color gradients to achieve a visual structural deformation map; Step 5.3: When one of the following alarm conditions is met, an alarm is issued: (1) The absolute value of displacement exceeds the safety limit; (2) the displacement slope increases abnormally within a certain time window; When the above conditions are triggered, the following actions are automatically performed: (1) Push alarm notification: real-time alarm via SMS, email or WeChat enterprise account; (2) Start the video acquisition module: transmit the image and original data at that time for remote expert judgment; (3) Interlocking control module: sends interlocking signals to the field control system.

8. A displacement detection system for structural health monitoring, characterized in that: The displacement detection system includes multiple displacement detection units, each of which includes a laser emission module and a light spot processing module. The light spot processing module includes a whiteboard, a high-resolution camera, an acquisition terminal, and a mounting base. The displacement detection system performs the following displacement detection method, which specifically includes the following steps: Step S1: performing image acquisition and preprocessing using the displacement detection system, wherein a high dynamic response image acquisition method capable of real-time adaptive adjustment according to rapid changes in ambient brightness, an adaptive Gaussian filtering noise reduction algorithm, and a CLAHE local contrast enhancement algorithm for enhancing the discernibility of light spots in non-ideal backgrounds are employed; Step S2: performing spot recognition and center coordinate extraction on the pre-processed image; Step S3: performing error compensation on the output result of step S2 to obtain compensated data; Step S4: Transmit the compensated data to the cloud server and perform intelligent structural health analysis.

9. The displacement detection system for structural health monitoring according to claim 8, characterized in that: In step S1, the high dynamic response image acquisition method is specifically as follows: the high-resolution camera uses a built-in ambient light sensor to obtain the light intensity value of the current acquisition point through continuous sampling, and introduces a parameter feedback adjustment mechanism based on the PID controller to dynamically control the key parameters of the camera.

10. The displacement detection method for structural health monitoring according to claim 8, characterized in that: In step S1, the adaptive Gaussian filtering denoising algorithm evaluates the local complexity of the image by using a structural tensor analysis method.

Images