Sub-pixel-level displacement monitoring method and monitoring system for steel rail telescopic regulator

Abstract

The invention discloses a sub-pixel-level displacement monitoring method and monitoring system for a steel rail telescopic regulator. The method comprises the following steps: S1, carrying out multi-modal data acquisition on the steel rail telescopic regulator; s2, carrying out sub-pixel-level edge positioning based on the gradient direction consistency constraint of the target edge; S3, carrying out multispectral cooperative imaging by adopting a self-adaptive weight fusion algorithm; s4, collecting temperature data in real time through a temperature sensor, and eliminating systematic deviation by applying a thermal expansion displacement compensation model; and S5, performing multi-modal data fusion on the data in the S2, the S3 and the S4. By innovatively fusing multispectral imaging, sub-pixel analysis and a physical compensation mechanism, a high-precision steel rail displacement monitoring system is constructed to form anti-interference and high-reliability monitoring output; the safety guarantee capability of the seamless track of the high-speed railway is remarkably improved, the monitoring precision is effectively improved, and the false alarm rate is reduced.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, specifically to a subpixel-level displacement monitoring method and system for rail expansion joints. The system is particularly suitable for real-time diagnosis of the safety status of seamless tracks in high-speed railway bridge sections and can effectively cope with the coupling effect of train impact loads and temperature stress. Background Technology

[0002] The relative displacement between the switch rail and the main rail of the rail expansion joint is a core parameter for assessing the safety status of seamless railway lines, and its monitoring accuracy is directly related to the operational safety of high-speed railways. Industry standards clearly require that an emergency warning must be triggered when the displacement exceeds the safety threshold, but existing technologies suffer from a high false alarm rate due to the coupling of environmental interference and physical effects. Traditional monitoring methods have three major defects: (1) Mechanical displacement sensors are affected by train vibration and cannot meet the sub-millimeter level accuracy requirements; (2) Optical equipment suffers from a deteriorated signal-to-noise ratio and a significant decrease in feature recognition rate in harsh environments such as rain and fog; (3) Existing methods ignore the physical mechanism of rail thermal expansion, resulting in systematic bias in the data. Current machine vision monitoring solutions still face fundamental limitations: the pixel-level edge detection algorithm is limited by the theoretical limit of spatial resolution, and the displacement measurement error is ≥0.3mm; at the same time, the feature recognition rate of visible light imaging is less than 60% under low illumination, strong backlight, or rain and snow interference. Engineering practice shows that the above defects lead to an unacceptable increase in the underreporting rate of extreme weather. Summary of the Invention

[0003] This invention addresses the problems in existing technologies by disclosing a subpixel-level displacement monitoring method and system for rail expansion joints. First, a composite target group containing geometric coding and a near-infrared light source is deployed. Images are simultaneously acquired using a visible-near-infrared dual-spectrum camera, and a signal-to-noise ratio adaptive weighted fusion algorithm overcomes interference from rain, fog, and strong light. Then, based on gradient direction consistency constraints and bicubic spline interpolation, 0.1-pixel-level edge localization is achieved, breaking through the accuracy limits of traditional visual measurements. Simultaneously, data is collected in real-time using a temperature sensor, and the thermal expansion equation is applied to eliminate systematic biases. Finally, an adaptive Kalman filter is used to fuse multi-source displacement data, forming an anti-interference, highly reliable monitoring output, effectively improving monitoring accuracy and reducing the false alarm rate.

[0004] This invention is achieved through the following technical solution: This invention first provides a subpixel-level displacement monitoring method for rail expansion joints, comprising the following steps: S1. Perform multimodal data acquisition on the rail expansion joint; S2. Sub-pixel-level edge localization based on gradient direction consistency constraints of the target edge; S3. Adaptive weight fusion algorithm is used for multispectral collaborative imaging; S4. Collect temperature data in real time through temperature sensors and apply a thermal expansion displacement compensation model to eliminate systematic deviations; S5. Perform multimodal data fusion on the data from S2, S3 and S4.

[0005] As a further solution, the specific method of S1 is as follows: Composite target groups are deployed on the sides of the switch rail and the base rail, and high-contrast geometric coding and near-infrared emitting units are set on the surface of the switch rail and the base rail. Images of the switch rail and the base rail are simultaneously acquired by a visible light-near-infrared dual-band camera, and temperature sensors collect the rail temperature data of the switch rail and the base rail in real time.

[0006] As a further option, the target in S2 is a checkerboard target, which is a PVC surface coated with a nano anti-fouling coating. This coating can prevent glare and diffuse reflection, thereby reducing interference with gradient calculation. The checkerboard grid is 4 rows × 9 columns, with a white area reflectivity of 92% and a black area reflectivity of 8%.

[0007] As a further solution, the specific method of S2 is as follows: S21. Calculate the gradient field of a visible light image. : First, converting the visible light image into a grayscale image effectively simplifies calculations and preserves the image's brightness information; Secondly, the Sobel operator is used to calculate the intensity change rate of each pixel in the image as a gradient vector, including the horizontal gradient formed by vertical edge convolution. and the vertical gradient formed by horizontal edge convolution. , representing the gradient component of each pixel, is used to calculate the gradient magnitude by substituting the grayscale coordinates as follows: the higher the value, the stronger the edge. ; Wherein, the pixel coordinates of the grayscale image are ; S22. Calculate the direction angle of the visible light image. : The direction angle represents the direction of gradient change and is calculated as follows: ; The atan2 function is an extension of the arctangent function, and its output angle is between -π and π radians.

[0008] S23, in the edge neighborhood Internally solve for the minimum point of the standard deviation of the gradient direction to accurately locate the target edge: ; in, This represents the coordinates of the target edge point, i.e., the coordinates of the point where the standard deviation of the gradient direction is minimized. Indicates at point The gradient direction at a given point is usually measured in radians or angles. This indicates the average gradient direction within the edge neighborhood Ω; S24. Accuracy is verified by bicubic spline interpolation, which improves the pixel positioning accuracy after visible light image processing, achieving an accuracy of 0.1 pixels. ; in, This represents the position increment after improving accuracy. This represents the original location estimate. The grayscale value function representing a visible light image. These represent the horizontal and vertical pixel coordinates of a visible light image, respectively.

[0009] As a further solution, the specific method of S3 is as follows: For the visible light and near-infrared images acquired by the visible light-near-infrared dual-band camera described in S1, an adaptive weight fusion algorithm is used: ; in, This represents the intensity value of the fused image. Weight and weight Dynamically adjusted based on real-time signal-to-noise ratio: , This represents the intensity value of visible light. This represents the intensity value of a near-infrared image.

[0010] but ;in, This represents the near-infrared compensation amount after weighted fusion.

[0011] As a further solution, the specific method of S4 is as follows: The rail temperature obtained from the temperature sensor in S1 and the visible light displacement obtained in S2 are compared... Establish a temperature-displacement coupled correction model to eliminate the influence of ambient temperature: ; In the formula, It represents the amount of temperature displacement change. This represents the actual measured displacement. This is the coefficient of linear expansion of the rail. The reference length for the expansion joint was obtained through on-site calibration using a total station. Data is collected in real time by a temperature sensor.

[0012] As a further solution, the specific method of S5 is as follows: The visible light shift value obtained in S2, the near-infrared compensation value obtained in S3, and the temperature correction value obtained in S4 are input into an adaptive Kalman filter for multimodal data fusion. ; in, This represents a visible light shift value with considerable accuracy. This represents the near-infrared compensation amount after weighted fusion. Indicates the temperature compensation amount. The Kalman gain determines the weights of displacement and compensation values ​​in state updates. The Kalman gain considers a sliding window of a certain size during the fusion process. The error variance matrix was calculated based on the most recent 30 sets of historical residual data from each sensor. The update cycle is 10 seconds, and the formula is as follows: ; Where H represents the state transition matrix of the Kalman filter (its state components remain constant). ; These correspond to the fusion weights of visible light shift value, near-infrared compensation amount, and temperature correction amount, respectively.

[0013] This invention also provides a subpixel-level displacement monitoring system for rail expansion joints, and the method for subpixel-level displacement monitoring of rail expansion joints includes: Composite target assembly: A ring code is etched on the surface of a titanium alloy substrate and an 850nm near-infrared LED is embedded therein, which is then mounted on the sides of the switch rail and the base rail. Multispectral imaging unit: Visible light + near-infrared dual sensor camera, used to acquire images of the switch rail and base rail in real time and transmit the data to the edge computing terminal; Temperature compensation module: used to collect the temperature of the switch rail and the base rail in real time and transmit the data to the edge computing terminal; Edge computing terminal: Processes the collected data using a built-in sub-pixel resolution algorithm and outputs the calculation results; Safety early warning platform: Transmits the calculation results of the edge computing terminal to the BIM model through a 5G private network to realize three-level alarm.

[0014] As a further solution, a three-level alarm includes: Level 1 alarm: An audible and visual alarm is triggered when the displacement exceeds the Level 1 warning threshold; Level 2 alarm: When the displacement exceeds the level 2 warning threshold, a text message is sent to maintenance personnel; Level 3 alarm: When the displacement exceeds the level 3 warning threshold, a line speed reduction command will be automatically triggered.

[0015] As a further solution, the edge computing terminal integrates the NVIDIA Jetson module, which has a built-in CUDA-accelerated subpixel resolution optimization library.

[0016] The features and beneficial effects of this invention are as follows: (1) This invention constructs a high-precision rail displacement monitoring system by innovatively integrating multispectral imaging, sub-pixel analysis and physical compensation mechanisms: First, a composite target group containing geometric coding and near-infrared light sources is deployed, and images are simultaneously acquired using a visible-near-infrared dual-spectral camera. The interference of rain, fog and strong light is overcome by a weighted fusion algorithm with adaptive signal-to-noise ratio. Then, 0.1-pixel-level edge positioning is achieved based on gradient direction consistency constraints and bicubic spline interpolation, breaking through the traditional visual measurement accuracy limit. At the same time, data is collected in real time through a temperature sensor, and the thermal expansion equation is applied to eliminate systematic deviations. Finally, an adaptive Kalman filter is used to fuse multi-source displacement data to form an anti-interference and highly reliable monitoring output. This scheme creatively integrates optical perception, physical modeling and edge computing, significantly improving the safety assurance capability of seamless high-speed railway lines. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the system deployment according to an embodiment of the present invention; Figure 2 This refers to the image with dynamically adjusted signal-to-noise ratio weights as described in this embodiment of the invention. Figure 3 This is a flowchart of the subpixel localization algorithm described in an embodiment of the present invention; Figure 4 This is a verification diagram of the temperature compensation effect described in an embodiment of the present invention; Figure 5 This is a schematic diagram of a chessboard target according to an embodiment of the present invention. Detailed Implementation

[0019] To facilitate understanding of the present invention, a more comprehensive description of the present invention will be given below, and embodiments of the present invention will be provided, but this does not limit the scope of the present invention.

[0020] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0021] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0022] Subpixel-level displacement monitoring method for rail expansion joints, such as Figures 1 to 5 As shown, it includes the following steps: S1. Perform multimodal data acquisition on the rail expansion joint. Composite target groups are deployed on the sides of the switch rail and the base rail, and high-contrast geometric coding and near-infrared emitting units are set on the surfaces of the switch rail and the base rail. Images of the switch rail and the base rail are simultaneously acquired by a visible light-near-infrared dual-band camera, and temperature sensors collect rail temperature data (of the switch rail and the base rail) in real time, effectively ensuring multimodal data coverage.

[0023] The composite target assembly has a circular code and an infrared LED etched on the substrate surface; Preferably, the infrared LED is embedded in the surface of the substrate, and the substrate is made of titanium alloy. The near-infrared emitting unit is an infrared LED. The wavelength of the infrared LED is not less than 850nm. The frame rate of the visible light-near infrared dual-band camera is ≥120fps; the protection level is IP68. The temperature sensor is a Pt100 platinum resistance temperature sensor with an accuracy of ±0.1℃. The geometric coding is a concentric ring structure with a line width of 0.5mm and a ring spacing of 1mm.

[0024] This method collects orbital temperature and image data from multiple aspects and merges these data to increase data coverage, providing more comprehensive data for subsequent processing and ensuring the accuracy and comprehensiveness of later data processing.

[0025] S2. Sub-pixel-level edge localization based on gradient direction consistency constraints of the target edge: Based on the gradient direction consistency constraint of the target edge, sub-pixel coordinates are calculated for the visible light image acquired by the visible light-near infrared dual-band camera described in S1: The target is a checkerboard target, made of PVC with a nano-anti-fouling coating. This coating prevents glare and diffuse reflection, thus reducing interference with gradient calculation. The checkerboard grid is 4 rows × 9 columns, with a white area reflectance of 92% and a black area reflectance of 8%. This contrast primarily affects the signal-to-noise ratio. Since the accuracy of the target directly affects the precision of sub-pixel coordinate calculation, errors that are easily caused during target generation and processing are calculated to ensure the accuracy verification experiment is conducted correctly. First, the experimental chessboard target was repeatedly measured 20 times using a laser tracker to obtain the theoretical and actual positions of the target's feature points, as shown in Table 1. Table 1

[0026] S21. Calculate the gradient field of a visible light image. : First, converting the visible light image into a grayscale image effectively simplifies calculations and preserves the image's brightness information; Secondly, the Sobel operator is used to calculate the intensity change rate of each pixel in the image as a gradient vector, including the horizontal gradient formed by vertical edge convolution. and the vertical gradient formed by horizontal edge convolution. , representing the gradient component of each pixel, is calculated by substituting the grayscale coordinates into the following formula. The higher the value, the stronger the edge.

[0027] ; Wherein, the pixel coordinates of the grayscale image are .

[0028] S22. Calculate the direction angle of the visible light image. : The direction angle represents the direction of gradient change and is calculated as follows.

[0029] ; The atan2 function is an extension of the arctangent function, and its output angle is between -π and π radians.

[0030] S23. Utilizing gradient field and visible light image orientation angle in the edge neighborhood Internally solve for the minimum point of the standard deviation of the gradient direction to accurately locate the target edge: ; in, This represents the coordinates of the target edge point, i.e., the coordinates of the point where the standard deviation of the gradient direction is minimized. Indicates at point The gradient direction at a given point is usually measured in radians or angles. This represents the average gradient direction within the edge neighborhood Ω.

[0031] S24. Accuracy is verified by cyclical calculation of bicubic spline interpolation gradient, which improves the pixel positioning accuracy after processing the visible light image, reaching the level of 0.1 pixels. ; in, This represents the position increment after improving accuracy. This represents the original location estimate. The grayscale value function representing a visible light image. These represent the horizontal and vertical pixel coordinates of a visible light image, respectively.

[0032] After improving the accuracy of the new position, a convergence test is performed on the position increment. If the accuracy meets the requirements, the process ends; otherwise, gradient direction constraints and bicubic spline interpolation are recalculated to determine the new position. This step breaks through the pixel-level resolution limit and effectively improves the theoretical recognition accuracy.

[0033] To compare the performance of this displacement monitoring method, the pure gradient direction method and the pure bicubic spline method were selected for verification experiments with this combined method. The specific experimental results are shown in Table 2 below.

[0034] Table 2

[0035] The experimental results in Table 2 show that the error accuracy of this combined method is controlled below 0.1 pixels, and the average error is significantly better than the other two methods, thereby improving the monitoring accuracy of the track and reducing the false alarm rate.

[0036] S3. Multispectral collaborative imaging is performed using an adaptive weight fusion algorithm: For the visible light and near-infrared images acquired by the visible light-near-infrared dual-band camera described in S1, an adaptive weight fusion algorithm is used: ; in, This represents the intensity value of the fused image. Weight and weight Dynamically adjusted based on real-time signal-to-noise ratio: , in, and The signal-to-noise ratios of visible light and near-infrared images are respectively. This represents the intensity value of visible light. This represents the intensity value of a near-infrared image.

[0037] but ,in, This indicates the near-infrared compensation amount after weighted fusion; this step effectively fuses visible light detail features by penetrating rain and fog through near-infrared active illumination.

[0038] S4. Collect temperature data in real time using a temperature sensor, and apply a thermal expansion displacement compensation model to eliminate systematic deviations: The rail temperature obtained from temperature sensor S1 and the visible light displacement obtained from S2. Establish a temperature-displacement coupled correction model to eliminate the influence of ambient temperature: ; In the formula, It represents the amount of temperature displacement change. This represents the actual measured displacement. This is the coefficient of linear expansion of the rail. The reference length for the expansion joint was obtained through on-site calibration using a total station. Data is collected in real time by a platinum resistance temperature sensor.

[0039] S5. Perform multimodal data fusion on the data from S2, S3 and S4; The visible light shift value obtained in S2, the near-infrared compensation value obtained in S3, and the temperature correction value obtained in S4 are input into an adaptive Kalman filter for multimodal data fusion. ; in, This represents a visible light shift value with considerable accuracy. This represents the near-infrared compensation amount after weighted fusion. Indicates the temperature compensation amount. The Kalman gain determines the weights of displacement and compensation values ​​in state updates. The Kalman gain considers a sliding window of a certain size during the fusion process. The error variance matrix was calculated based on the most recent 30 sets of historical residual data from each sensor. The update cycle is 10 seconds, and the formula is as follows: ; Where H represents the state transition matrix of the Kalman filter (its state components remain constant). ; These correspond to the fusion weights of visible light shift value, near-infrared compensation amount, and temperature correction amount, respectively.

[0040] Comparative Example 1: This comparative scheme only deploys ordinary black and white checkerboard targets without functional coatings at track monitoring points, uses a single visible light band camera for image acquisition, and completely lacks temperature sensing devices. The theoretical accuracy limit achieved by the Canny edge detection and Harris corner matching algorithm is less than 0.5 pixels. Due to the lack of a multispectral anti-interference mechanism, the probability of feature point loss of lock-on increases by 300% under rainy and foggy weather conditions. At the same time, the lack of a temperature compensation mechanism leads to a systematic error of 2.0-2 due to the thermal expansion effect of the rail under a 15°C temperature difference condition. The error is 0.3mm, exceeding the safety threshold for high-speed rail tracks. Field verification data shows that the average error of displacement monitoring under standard lighting conditions is 0.45 pixels, but it deteriorates to 0.68 pixels under low-light conditions at night, with a feature recognition rate of less than 55%. The false alarm rate exceeds the technical red line of 60% in rainy and foggy weather. More seriously, when the track temperature changes by 15°C, the displacement is not compensated and the false alarm is directly triggered, causing the false alarm rate to increase to 45-48%. Under the normal working condition of ±10°C temperature change, the monitoring confidence is less than 55%, which does not meet the engineering applicability of high-precision monitoring of high-speed rail tracks in all weather conditions.

[0041] Example 1: A specially designed composite target group is deployed on the sides of the switch rail and main rail, in conjunction with a high frame rate visible-near-infrared dual-band camera and a high-precision Pt100 temperature sensor to simultaneously acquire multimodal data. In step S2, image processing is performed using checkerboard targets coated with a nano-anti-fouling coating. Combined with Sobel gradient field calculation and gradient direction consistency constraints, sub-pixel-level edge features are accurately located, and bicubic spline interpolation technology is used for iterative verification, ultimately achieving a stable positioning accuracy of up to 0.1 pixels. In step S3, the system dynamically calculates the fusion weights of visible and near-infrared images based on the real-time signal-to-noise ratio, adaptively generating high-quality fused images, significantly improving feature recognition capabilities under adverse conditions such as rain, fog, and backlight. In step S4, a rail thermal expansion compensation model is established, using real-time rail temperature data to accurately calculate displacement deviations caused by temperature. For example, when the day-night temperature difference reaches 15°C, a systematic error of approximately 2.7 mm can be automatically deducted. In step S5, the visual displacement value, near-infrared compensation, and temperature correction are input into an adaptive Kalman filter. This filter dynamically updates its gain coefficient based on the most recent 30 sets of historical residual data, with a period of 10 seconds, and finally outputs a fused high-reliability displacement sequence. Actual test data shows that during 30 days of continuous line monitoring, the average error of feature point displacement was stably controlled within 0.07 pixels (approximately 0.1 mm), the maximum error did not exceed 0.097 pixels, and the overall false alarm rate was less than 5%.

[0042] This invention creatively integrates multimodal imaging and physical compensation mechanisms, achieving high-precision and robust monitoring of rail expansion joint displacement without the need for quantitative indicators. It effectively overcomes environmental interference through multispectral collaborative imaging, breaks through traditional accuracy limitations with sub-pixel positioning technology, and eliminates systematic deviations through embedded thermal expansion compensation. Ultimately, it constructs a real-time monitoring system that meets the stringent requirements of high-speed railways, significantly improving the safety assurance capability of seamless tracks.

[0043] A subpixel-level displacement monitoring method for rail expansion joints includes: Composite target assembly: A ring code is etched on the surface of a titanium alloy substrate and an 850nm near-infrared LED is embedded therein, which is then mounted on the sides of the switch rail and the base rail. Multispectral imaging unit: Visible light + near-infrared dual sensor camera, used to acquire images of the switch rail and base rail in real time and transmit the data to the edge computing terminal; Temperature compensation module: used to collect the temperature of the switch rail and the base rail in real time and transmit the data to the edge computing terminal; Edge computing terminal: Processes the collected data using a built-in sub-pixel resolution algorithm and outputs the calculation results; Safety early warning platform: Transmits the calculation results of the edge computing terminal to the BIM model through a 5G private network to realize three-level alarm.

[0044] This invention creatively integrates optical imaging technology, physical law modeling, and edge intelligent computing, solving the problems of large environmental interference, insufficient accuracy, and lack of physical effect compensation in traditional monitoring, and providing a reliable guarantee for the safe operation of seamless high-speed railway lines.

[0045] Preferably, the edge computing terminal integrates an NVIDIA Jetson module and has a built-in CUDA-accelerated subpixel resolution optimization library.

[0046] Preferably, the three-level alarm mechanism of the security early warning platform includes: Level 1 alarm: An audible and visual alarm is triggered when the displacement exceeds the Level 1 warning threshold; Level 2 alarm: When the displacement exceeds the level 2 warning threshold, a text message is sent to maintenance personnel; Level 3 alarm: When the displacement exceeds the level 3 warning threshold, a line speed reduction command will be automatically triggered.

[0047] The present invention also provides a non-transitory computer-readable storage medium (industrial-grade SD card) for storing the following computer instructions: Adaptive spectral fusion algorithm code; Subpixel edge localization optimization program; The Kalman filter dynamic weight update logic implements the displacement monitoring method described above when the processor executes instructions.

[0048] The present invention also provides a computer device, including a memory and a processor, wherein the processor and the memory communicate with each other, the memory stores program instructions executable by the processor, and the processor invokes the program instructions to execute the monitoring method described above, including the following steps: Visible and near-infrared images of the switch rail and the base rail are acquired simultaneously using a dual-spectrum camera. An adaptive weight fusion algorithm is used to enhance target features; Sub-pixel level edge localization based on gradient direction consistency constraints; The effects of thermal expansion are eliminated through a temperature-displacement coupling model; An adaptive Kalman filter is used to fuse multi-source displacement data; An early warning mechanism is triggered when the fusion displacement value exceeds the safety threshold.

[0049] The present invention also provides an electronic device, comprising: a processor, a memory, and a computer program; wherein the processor is connected to the memory, and the computer program is stored in the memory; when the electronic device is running, the processor executes the computer program stored in the memory to cause the electronic device to execute instructions for implementing the monitoring method described above, the instructions including: The multispectral imaging unit is controlled to simultaneously acquire visible light and near-infrared images at a frame rate of 120fps. Execute the sub-pixel localization algorithm to calculate the target edge coordinates; Real-time reading of temperature sensor data and calculation of thermal expansion compensation; Run an adaptive Kalman filter program to fuse multi-source displacement data; The displacement monitoring results are transmitted to the BIM safety platform via a 5G communication module. The audible and visual alarm device is activated when the displacement exceeds the limit. The electronic equipment is integrated into an IP68-rated edge computing terminal and deployed at railway sites to achieve unattended monitoring.

[0050] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0051] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0052] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0053] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment, whereby a series of operational steps are performed to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0054] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that, based on the technical solutions disclosed in the present invention, various modifications or variations that can be made by those skilled in the art without creative effort should be included within the scope of protection of the present invention.

Claims

1. A subpixel-level displacement monitoring method for rail expansion joints, characterized in that: Includes the following steps: S1. Perform multimodal data acquisition on the rail expansion joint; S2. Sub-pixel-level edge localization based on gradient direction consistency constraints of the target edge; S3. Adaptive weight fusion algorithm is used for multispectral collaborative imaging; S4. Collect temperature data in real time through temperature sensors and apply a thermal expansion displacement compensation model to eliminate systematic deviations; S5. Perform multimodal data fusion on the data from S2, S3 and S4.

2. The sub-pixel level displacement monitoring method for rail expansion joints according to claim 1, characterized in that: The specific method for S1 is as follows: Composite target groups are deployed on the sides of the switch rail and the base rail, and high-contrast geometric coding and near-infrared emitting units are set on the surface of the switch rail and the base rail. Images of the switch rail and the base rail are simultaneously acquired by a visible light-near-infrared dual-band camera, and temperature sensors collect the rail temperature data of the switch rail and the base rail in real time.

3. The sub-pixel level displacement monitoring method for rail expansion joints according to claim 1, characterized in that: The target in S2 is a checkerboard target, which is a PVC surface coated with a nano anti-fouling coating. The checkerboard grid has 4 rows and 9 columns, with a white area reflectivity of 92% and a black area reflectivity of 8%.

4. The sub-pixel level displacement monitoring method for rail expansion joints according to claim 1, characterized in that: The specific method for S2 is as follows: S21. Calculate the gradient field of a visible light image. : First, convert the visible light image into a grayscale image; Secondly, the Sobel operator is used to calculate the intensity change rate of each pixel in the image as a gradient vector, including the horizontal gradient formed by vertical edge convolution. and the vertical gradient formed by horizontal edge convolution. , representing the gradient component of each pixel, is used to calculate the gradient magnitude by substituting the grayscale coordinates as follows; ； Wherein, the pixel coordinates of the grayscale image are ; S22. Calculate the direction angle of the visible light image. : The direction angle represents the direction of gradient change and is calculated as follows: ； The atan2 function is an extension of the arctangent function, and its output angle is between -π and π radians. S23, in the edge neighborhood Minimum point of the standard deviation of the gradient direction in the inner solution: ； in, This represents the coordinates of the target edge point, i.e., the coordinates of the point where the standard deviation of the gradient direction is minimized. Indicates at point The gradient direction at a given point is measured in radians or angles. This indicates the average gradient direction within the edge neighborhood Ω; S24. Accuracy verification is performed through bicubic spline interpolation. ； in, This represents the position increment after improving accuracy. This represents the original location estimate. The grayscale value function representing a visible light image. , These represent the horizontal and vertical pixel coordinates of a visible light image, respectively.

5. The sub-pixel level displacement monitoring method for rail expansion joints according to claim 2, characterized in that: The specific method for S3 is as follows: For the visible light and near-infrared images acquired by the visible light-near-infrared dual-band camera described in S1, an adaptive weight fusion algorithm is used: ； in, The weights represent the intensity values ​​of the fused image. and weight Dynamically adjusted based on real-time signal-to-noise ratio: ， This represents the intensity value of visible light. Indicates the intensity value of a near-infrared image; but ;in, This represents the near-infrared compensation amount after weighted fusion.

6. The sub-pixel level displacement monitoring method for rail expansion joints according to claim 1, characterized in that: The specific method for S4 is as follows: The rail temperature obtained from the temperature sensor in S1 and the visible light displacement obtained in S2 are compared... Establish a temperature-displacement coupled correction model: ； In the formula, This represents the amount of temperature displacement change. This represents the actual measured displacement. This is the coefficient of linear expansion of the rail. The reference length for the expansion joint was obtained through on-site calibration using a total station. Data is collected in real time by a temperature sensor.

7. The sub-pixel level displacement monitoring method for rail expansion joints according to claim 1, characterized in that: The specific method for S5 is as follows: The visible light shift value obtained in S2, the near-infrared compensation value obtained in S3, and the temperature correction value obtained in S4 are input into an adaptive Kalman filter for multimodal data fusion. ； in, This represents a visible light shift value with considerable accuracy. This represents the near-infrared compensation amount after weighted fusion. Indicates the temperature compensation amount. The Kalman gain determines the weights of displacement and compensation values ​​in state updates; the Kalman gain considers a sliding window during the fusion process. The error variance matrix was calculated based on the most recent 30 sets of historical residual data from each sensor. The update cycle is 10 seconds, and the formula is as follows: ； Where H represents the state transition matrix of the Kalman filter. ; These correspond to the fusion weights of visible light shift value, near-infrared compensation amount, and temperature correction amount, respectively.

8. A subpixel-level displacement monitoring system for rail expansion joints, employing the subpixel-level displacement monitoring method for rail expansion joints as described in any one of claims 1 to 7, characterized in that: include: Composite target assembly: A ring code is etched on the surface of a titanium alloy substrate and an 850nm near-infrared LED is embedded therein, which is mounted on the sides of the switch rail and the base rail; Multispectral imaging unit: Visible light + near-infrared dual sensor camera, used to acquire images of the switch rail and base rail in real time and transmit the data to the edge computing terminal; Temperature compensation module: used to collect the temperature of the switch rail and the base rail in real time and transmit the data to the edge computing terminal; Edge computing terminal: Processes the collected data using a built-in sub-pixel resolution algorithm and outputs the calculation results; Safety early warning platform: Transmits the calculation results of the edge computing terminal to the BIM model through a 5G private network to realize three-level alarm.

9. The sub-pixel displacement monitoring system for rail expansion joints according to claim 8, characterized in that: Level 3 alarms include: Level 1 alarm: An audible and visual alarm is triggered when the displacement exceeds the Level 1 warning threshold; Level 2 alarm: When the displacement exceeds the level 2 warning threshold, a text message is sent to maintenance personnel; Level 3 alarm: When the displacement exceeds the level 3 warning threshold, a line speed reduction command will be automatically triggered.

10. The subpixel-level displacement monitoring system for rail expansion joints according to claim 8, characterized in that: The edge computing terminal integrates the NVIDIA Jetson module and has a built-in CUDA-accelerated subpixel resolution optimization library.

Images