Railway tunnel crack monitoring method and system

Abstract

The invention provides a railway tunnel crack monitoring method and system, and relates to the data processing technology, and the method comprises the steps: determining a starting point, an ending point and an extension route of a tunnel crack; controlling the high-precision acquisition unit to move along the extension route by taking the starting point as a starting point until the starting point is positioned on a window frame line of the acquisition window; determining a sub-route of the collection window corresponding to the extension route, and determining a first transverse point and a second transverse point which are located on the two sides of the sub-route and have distribution identification; controlling a transverse illumination unit carried by the high-precision acquisition unit to move to coincide with the transverse point, and controlling the high-precision acquisition unit to perform acquisition to obtain an acquisition image corresponding to the sub-route; and determining the route end point far away from the starting point as a new starting point based on the sub-route, and repeating the steps until the end point is located in the acquired image corresponding to any acquisition frequency, thereby obtaining detection data of the corresponding tunnel crack. The monitoring accuracy is at least improved.

Description

Technical Field

[0001] This invention relates to data processing technology, and more particularly to a method and system for monitoring cracks in railway tunnels. Background Technology

[0002] As a key infrastructure for rail transit, railway tunnels are subjected to multiple effects such as surrounding rock pressure, train vibration, temperature changes and groundwater erosion over a long period of time. This makes them highly susceptible to cracks on the surface of the lining structure. The development and expansion of cracks not only weaken the load-bearing capacity of the tunnel structure, but may also cause secondary diseases such as water leakage and lining spalling. In severe cases, it may even lead to structural instability and directly threaten the safety of train operation.

[0003] With the continuous increase in the operating mileage of my country's railway network, the aging problem of tunnels in service is becoming increasingly prominent, and crack monitoring has become a core aspect of tunnel operation and maintenance. How to accurately capture key parameters such as crack length, width, and extension direction to provide reliable data support for defect assessment and repair is an important issue for ensuring the long-term safe operation of railway tunnels. Currently, railway tunnel crack monitoring mainly adopts a manual inspection mode. Maintenance personnel use tools such as flashlights, tape measures, and crack width gauges to inspect the tunnel lining surface one by one, observe and record crack characteristics and measure dimensions with the naked eye. This mode relies on personnel experience, is easily affected by lighting conditions and visual fatigue, has limited ability to identify fine cracks (e.g., width < 0.2 mm), and the measurement data is highly subjective and has low accuracy. Therefore, there is an urgent need to provide a method and system for monitoring cracks in railway tunnels that can improve the accuracy of monitoring. Summary of the Invention

[0004] In view of the above problems, the present invention is proposed to provide a method and system for monitoring cracks in railway tunnels that overcomes or at least partially solves the above problems.

[0005] According to one aspect of the present invention, a method for monitoring cracks in railway tunnels is provided, comprising the following steps: Determine the starting and ending points of the corresponding tunnel cracks, as well as the extension route determined by the starting and ending points. The high-precision acquisition unit is controlled to move along the extended route from the starting point until the starting point is located at the window frame of the acquisition window of the high-precision acquisition unit. Determine the sub-route corresponding to the acquisition window of the extended route, and determine the first lateral point and the second lateral point with distribution characteristics on both sides of the sub-route based on the tunnel crack; The high-precision acquisition unit is controlled to move the lateral illumination unit to coincide with the lateral point, and the lateral illumination unit responds to the illumination, thereby controlling the high-precision acquisition unit to acquire images of the corresponding sub-route. Based on the sub-route, the route endpoint far from the starting point is determined as the new starting point. The above steps are repeated until the termination point is located in the image acquired for any number of acquisitions, thus obtaining the detection data of the corresponding tunnel crack.

[0006] Optionally, in the method according to the invention, determining the starting point, ending point, and extension route determined by the starting point and ending point of the corresponding tunnel crack includes: In response to receiving a crack monitoring signal carrying location information sent by any employee terminal, the system controls the drone to fly to the location information at a preset flight altitude to collect the information and sends the obtained flight image to the employee terminal. The system responds to the location determination signal sent by the receiving employee based on the flight image, determines the crack region located in the flight image that indicates the tunnel crack based on image recognition, and determines the two image contour points with the largest corresponding distance that make up the crack region as the start point and end point. The extension route of the corresponding tunnel crack is determined based on the established start and end lines connecting the starting and ending points.

[0007] Optionally, in the method according to the invention, determining the extension route of the corresponding tunnel crack based on the established start-end line connecting the starting point and the end point includes: Generate each first array line for each first array point obtained by arraying the start and end lines along a direction perpendicular to the start and end lines; Establish an image coordinate system, where the start and end lines coincide with the Y-axis of the image coordinate system; The average value of the corresponding horizontal coordinate values ​​of all image coordinate points that coincide with the region contour for each first array line is calculated, and the image coordinate points located in the crack region are determined as route points based on the obtained average horizontal coordinate values. Connect the route points that are adjacent to each other along the extension direction of the start and end lines to obtain the extended route.

[0008] Optionally, in the method according to the present invention, the high-precision acquisition unit is controlled to move along an extension path from a starting point until the starting point is located at the window frame of the acquisition window of the high-precision acquisition unit, and then the method further includes: The window frame line of the acquisition window with the starting point located at the high-precision acquisition unit is determined as the adjustment frame line, and the two window frame lines perpendicular to the adjustment frame line are determined as the verification frame lines; The response determines that any check frame line coincides with a tunnel crack based on the acquisition window. The acquisition multiplier of the corresponding acquisition window is then reduced by the same baseline multiplier to obtain the acquisition multiplier corresponding to different reduction times. The response determines that each check frame does not coincide with the tunnel crack based on the acquisition multiple corresponding to any decreasing number of acquisitions. It controls the high-precision acquisition unit to move away from the starting point along the extension route until the starting point is located at the adjustment frame.

[0009] Optionally, in the method according to the present invention, determining the sub-route corresponding to the acquisition window of the extended route, and determining the first lateral point and the second lateral point with distribution characteristics located on both sides of the sub-route based on the tunnel crack, includes: The sub-path is divided into arrays, and each second array line is generated along a direction perpendicular to the tangent direction of the sub-path for each second array point obtained by the respective path; Determine the first intersection point and the second intersection point where each second array line intersects with the crack profile of the tunnel crack, and determine the intersection distance between the first intersection point and the second intersection point; The maximum intersection distance is defined as the maximum distance. Then, with the second array line corresponding to the maximum distance as the center, a calculation process is constructed to calculate the difference between the corresponding intersection distance and the maximum distance in turn, and different distance differences are obtained. If any distance difference is less than a preset difference, the second array line corresponding to that distance difference is determined as the distribution characteristic line; otherwise, the calculation process is stopped. The first and second intersection points of the distribution characteristic lines corresponding to the center positions along the sub-route are determined as the first and second lateral points with distribution characteristics.

[0010] Optionally, in the method according to the invention, controlling the lateral illumination unit carried by the high-precision acquisition unit to move to coincide with the lateral point includes: If the center point of the response acquisition window does not coincide with the feature intersection point where the distribution feature line of the corresponding horizontal point intersects with the sub-line segment, the high-precision acquisition unit is controlled to move along the direction from the center point of the window toward the feature intersection point until the center point of the window coincides with the feature intersection point. Update the acquisition multiplier of the corresponding acquisition window until the starting point is located at the adjustment frame line, and obtain the corresponding updated value. The unit length between each horizontal point and the sub-route is determined based on the distribution feature line, and the update factor is determined based on the number of decreases corresponding to the acquisition factor of the acquisition window and the update value. The unit length is updated by a corresponding update factor, and the lateral illumination unit carried by the high-precision acquisition unit is controlled to move towards different lateral points based on the actual length of each lateral point.

[0011] Optionally, in the method according to the present invention, illumination is provided in response to the lateral illumination unit, and the high-precision acquisition unit is controlled to acquire images corresponding to the sub-route, and then the method further includes: Establish a real coordinate system, and define the starting point and the route endpoints far from the starting point, determined based on the sub-route, as the first vertical point and the second vertical point, respectively, wherein any distribution feature line coincides with the X-axis of the real coordinate system; In response to the existence of a horizontal coordinate difference between any vertical point and the center point of the window, the high-precision acquisition unit is controlled to move horizontally according to the corresponding coordinate difference, and the vertical coordinate difference between the vertical point and the center point of the window is obtained. The high-precision acquisition unit controls the vertical illumination unit carried by the high-precision acquisition unit to move towards the vertical point based on the vertical coordinate difference, and responds to the vertical illumination unit to illuminate based on the vertical point, controls the high-precision acquisition unit to acquire data, and updates the acquired image based on the obtained vertical supplementary image to obtain the updated acquired image.

[0012] Optionally, in the method according to the present invention, updating the acquired image based on the obtained vertical supplementary image to obtain an updated acquired image includes: The image portion that is the same as the vertically supplemented image and the acquired image is determined as the region to be supplemented, and the first contrast and second contrast of different image pixels that make up the region to be supplemented are determined based on the vertically supplemented image and the acquired image, respectively. Image pixels in the vertical supplementary image whose first contrast is greater than the second contrast are identified as supplementary pixels, and supplementary pixels in adjacent positions are connected to obtain each supplementary sub-region. The corresponding region to be supplemented in the acquired image is replaced based on each supplementary sub-region to obtain the updated acquired image.

[0013] Optionally, in the method according to the present invention, obtaining the detection data corresponding to the tunnel crack includes: Determine the acquisition factor for images acquired at different acquisition times, and set the smallest acquisition factor as the stitching factor; The captured images are reduced by the corresponding stitching factor using the image center point as the center to obtain the updated captured images; Create vertically arranged image fill slots on the data display interface, and fill each captured image into different image fill slots from top to bottom according to the capture order. The sub-routes of the captured images corresponding to different image fill slots are presented as line segment connections. Create filling slots of various lengths that are horizontally arranged with each image filling slot, and fill the filling slots of different lengths sequentially from top to bottom according to the acquisition order of the same acquired image, so as to obtain the detection data of the corresponding tunnel cracks composed of the data display interface.

[0014] According to another aspect of the present invention, a railway tunnel crack monitoring system is provided, comprising: The route determination module is configured to determine the start point, end point, and extended route determined by the start point and end point of the corresponding tunnel crack. The acquisition and movement module is configured to control the high-precision acquisition unit to move along the extended route from the starting point until the starting point is located at the window frame of the acquisition window of the high-precision acquisition unit. The distribution determination module is configured to determine the sub-route corresponding to the acquisition window of the extended route, and to determine the first lateral point and the second lateral point with distribution characteristics located on both sides of the sub-route based on the tunnel crack; The image acquisition module is configured to control the lateral illumination unit carried by the high-precision acquisition unit to move to coincide with the lateral point, and respond to the lateral illumination unit to provide illumination, thereby controlling the high-precision acquisition unit to acquire images of the corresponding sub-route. The data integration module is configured to determine the route endpoints far from the starting point as new starting points based on the sub-route, repeat the above steps until the termination point is located in the image acquired for any number of acquisitions, and obtain the detection data of the corresponding tunnel crack.

[0015] According to the solution of this invention, the invention effectively overcomes the bottleneck of insufficient accuracy in tunnel crack monitoring under manual inspection mode, significantly improves the accuracy and reliability of crack parameter measurement, and provides high-quality data support for tunnel structural defect assessment. Specifically, this can be reflected in the following aspects: 1. In terms of precise crack detail capture, this invention completely solves the problems of uneven lighting and difficulty in identifying minute cracks during manual inspection through the collaborative design of lateral illumination adaptation and high-precision acquisition. Specifically, based on the first and second lateral points on both sides of the sub-route, the lateral illumination unit can be precisely moved to the target position and turned on to form a directional light source perpendicular to the crack direction. Thus, based on the directional illumination, a clear contrast of light and dark can be formed at the crack edge, clearly distinguishing minute cracks from interference features such as stains and scratches on the lining surface, avoiding the problem of blurred details caused by the scattered flashlight light during manual inspection. At the same time, the high-precision acquisition unit performs image acquisition under the support of directional illumination, which can completely preserve the microscopic features such as the edge contour and width change of the crack. The detail clarity of the acquired image is dozens of times higher than that of manual observation, laying the foundation for the subsequent accurate measurement of crack width and shape. 2. In terms of continuous monitoring of the entire crack, this invention achieves complete coverage of the crack extension trajectory through route tracking and segment-by-segment iterative acquisition mode. First, the starting point, ending point and extension route of the crack are identified. Then, the high-precision acquisition unit is controlled to move along the route segment by segment. Sub-routes are divided with acquisition windows as units. After each segment of the sub-route is acquired, the starting point is updated and the operation is repeated until the entire crack is covered. Continuous acquisition images of the crack from the beginning to the end can be obtained, which avoids the omission of the middle section of the crack and can accurately capture the changes in the direction of crack extension. This solves the technical problem that it is difficult to form a complete crack data chain by manual inspection. 3. Regarding the improvement of measurement accuracy, this invention significantly reduces measurement errors through clear positioning benchmarks and data traceability. Using the crack extension route as a benchmark, the acquisition window is precisely aligned with the sub-route, ensuring that the shooting angle of each acquired image is always focused on the core area of ​​the crack, avoiding the perspective deviation that occurs when measuring with handheld tools. At the same time, the precise overlap of the lateral illumination unit and the lateral point ensures the consistency of illumination conditions for images acquired from different sub-routes, eliminating image comparison errors caused by illumination differences. Furthermore, the detection data acquired segment by segment can be spliced ​​together through sub-route association to form a complete digital archive of the crack. This not only allows for the direct extraction of macroscopic parameters such as crack length and extension direction, but also enables the calculation of width changes in each segment based on continuous images, thus improving monitoring accuracy. Attached Figure Description

[0016] Figure 1 A flowchart of a railway tunnel crack monitoring method according to an embodiment of the present invention is shown; Figure 2 This embodiment shows a schematic diagram of the structure of the high-precision acquisition unit and the illumination unit. Figure 3 A system block diagram of a railway tunnel crack monitoring system according to another embodiment of the present invention is shown. Detailed Implementation

[0017] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0018] To address the problems existing in the prior art, the inventors proposed the solution of this invention. One embodiment of this invention provides a method for monitoring cracks in railway tunnels. This method can be executed in a computing device, which can be understood as a terminal with data processing capabilities, such as a mobile phone or computer.

[0019] Figure 1 A flowchart of a railway tunnel crack monitoring method according to an embodiment of the present invention is shown. like Figure 1 As shown, the method proposed in this embodiment begins with step S1, which includes the following: Determine the starting and ending points of the corresponding tunnel cracks, as well as the extension route determined by the starting and ending points.

[0020] For example, in this embodiment, by determining the starting point, ending point, and extension route of the tunnel crack, a clear location benchmark and path guidance can be provided for the movement and data acquisition of the subsequent high-precision acquisition unit, ensuring the targeting and completeness of tunnel crack monitoring. It can be explained that the starting point is the initial endpoint of the tunnel crack's extension in space, and the ending point is the final endpoint of the tunnel crack's extension. These are key location nodes defining the overall range of the crack. For example, in long-distance tunnel crack monitoring, determining the starting and ending points allows the monitoring process to form a clear "from A to B" task loop, ensuring that every crack segment is included in the monitoring range. Furthermore, the extension route is the trajectory line connecting the starting and ending points, conforming to the actual direction of the tunnel crack. It is not a simple straight line connection, but a path formed based on the actual extension shape of the crack (e.g., a curve or broken line adapting to the crack's bending and turning). Determining this extension route provides a precise "navigation path" for the movement of the subsequent high-precision acquisition unit, allowing the high-precision acquisition unit to gradually advance along the extension route, ensuring that it is always aligned with the crack area and avoiding the acquisition image deviating from the crack target due to path deviation.

[0021] Furthermore, in this embodiment, the aforementioned "determining the starting point, ending point, and extension route determined by the starting point and ending point of the corresponding tunnel crack" may further include the following steps: In response to receiving a crack monitoring signal carrying location information sent by any employee terminal, the system controls the drone to fly to the location information at a preset flight altitude to collect the information and sends the obtained flight image to the employee terminal. The system responds to the location determination signal sent by the receiving employee based on the flight image, determines the crack region located in the flight image that indicates the tunnel crack based on image recognition, and determines the two image contour points with the largest corresponding distance that make up the crack region as the start point and end point. The extension route of the corresponding tunnel crack is determined based on the established start and end lines connecting the starting and ending points.

[0022] For example, in this embodiment, the determination of the starting point, ending point, and extension route can be specifically implemented based on the following method steps: First, in response to receiving a crack monitoring signal carrying location information from any employee terminal, the server involved in this embodiment can control a drone to fly to the location information at a preset flight altitude to collect data and send the obtained flight images to the employee terminal. It can be explained that the employee terminal can be understood as the terminal device operated by the on-site staff, and the location information is the approximate spatial location reported by the on-site staff after discovering a suspected crack (e.g., the left arch at tunnel K1+200). The preset flight altitude is a safe height set based on the tunnel clearance height that can cover the monitoring area (e.g., 5 meters from the tunnel wall to ensure a clear shooting range and no risk of collision). Here, by flying to the location information location with a drone carrying image acquisition equipment to collect flight images, it is possible to quickly reach areas that are difficult for humans to reach, avoiding the limitations of manual surveying. At the same time, the images are transmitted back to the employee terminal in real time, providing a visual basis for subsequent location confirmation and greatly improving the efficiency of initial crack location. Secondly, in response to the location determination signal sent by the employee terminal based on the flight image, the server can determine the crack region located in the flight image that indicates the tunnel crack based on image recognition, and determine the two image contour points with the largest corresponding distance to the region contour that makes up the crack region as the start point and end point; here, the location determination signal is the trigger command after the employee confirms the location of the tunnel crack in the flight image through the employee terminal, which can further confirm the location of the tunnel crack, and the crack region indicating the tunnel crack can be accurately extracted from the flight image based on image recognition technology (e.g., edge detection algorithm), eliminating background interference (e.g., tunnel wall stains, pipeline shadows); it can be explained that since the region contour is the edge contour line of the crack region, the distance between all image contour points on the contour can be calculated, and the two points with the largest distance can be determined as the start point and end point; Finally, the server can determine the extension route of the corresponding tunnel crack based on the established start-end line connecting the starting and ending points. Here, the start-end line is a line segment formed by connecting the starting and ending points with a straight line, which can initially outline the extension direction of the crack. Based on this start-end line, and combined with the actual direction of the crack in the flight image, the extension route that fits the true extension shape of the crack can be determined. It can be said that the establishment of the start-end line provides a clear benchmark framework for the extension route, avoids the randomness of route determination, and ensures that the extension route always revolves around the core area of ​​the crack. This provides accurate path guidance for the subsequent high-precision acquisition unit to move and monitor along the route, solves the problem of easy path deviation without a benchmark, and ensures that intelligent monitoring can cover the entire crack in an orderly manner.

[0023] Furthermore, in this embodiment, the aforementioned "determining the extension route of the corresponding tunnel crack based on the established start and end lines connecting the starting and ending points" may further include the following steps: Generate each first array line for each first array point obtained by arraying the start and end lines along a direction perpendicular to the start and end lines; Establish an image coordinate system, where the start and end lines coincide with the Y-axis of the image coordinate system; The average value of the corresponding horizontal coordinate values ​​of all image coordinate points that coincide with the region contour for each first array line is calculated, and the image coordinate points located in the crack region are determined as route points based on the obtained average horizontal coordinate values. Connect the route points that are adjacent to each other along the extension direction of the start and end lines to obtain the extended route.

[0024] For example, in this embodiment, determining the extension route based on the established start and end lines can be specifically implemented based on the following method steps: First, the server can generate first array lines for each first array point obtained by arraying the start and end lines along a direction perpendicular to the start and end lines. Here, the start and end lines are straight lines connecting the start and end points. The arraying process is to evenly divide the start and end lines into multiple first array points with consistent intervals (for example, setting one point every 10 pixels). By generating a straight line passing through each first array point along a direction perpendicular to the start and end lines (i.e., the corresponding horizontal direction), the first array line is generated. It can be explained that each generated first array line can be like a "ruler scale" to divide the narrow crack area into multiple horizontal analysis units at longitudinal intervals, realizing segmented and refined exploration of the crack outline, avoiding misjudgment of the direction caused by overall coarse analysis, and providing a uniformly distributed analysis benchmark for subsequent extraction of route points. Secondly, the server can establish an image coordinate system based on the start and end lines coinciding with the Y-axis of the image coordinate system. Here, the image coordinate system is a Cartesian coordinate system established based on the flight image. By coinciding the start and end lines with the Y-axis (vertical axis), the first array line can be naturally parallel to the X-axis (horizontal axis), realizing the coordinate alignment of the crack extension direction (vertical) and the analysis direction (horizontal). This provides a unified standard for subsequent quantitative analysis of the lateral position of the crack profile, ensuring that the description of the profile point position on different first array lines is consistent, avoiding calculation deviations caused by coordinate confusion, and laying the coordinate system foundation for accurate extraction of route points. Next, the server can calculate the average of the corresponding horizontal coordinate values ​​of all image coordinate points that correspond to the same vertical coordinate value and overlap with the region contour for each first array line. Based on the obtained average horizontal coordinate value, the image coordinate points located in the crack region are determined as route points. Here, the same first array line corresponds to the same vertical coordinate value (Y value), and its intersection with the region contour of the crack region will form multiple image coordinate points. Calculating the average of the horizontal coordinate values ​​(X values) of these points can filter out noise points or irregular protrusions at the contour edge, and obtain the core horizontal center position of the crack at that vertical position. Determining the image coordinate points located in the crack region corresponding to the average value as route points can ensure that each route point accurately corresponds to the center direction of the crack at that vertical position, thereby effectively avoiding the point deviation caused by the irregularity of the crack contour, so that the route points can truly reflect the core extension trajectory of the crack. Finally, the server can connect adjacent route points along the extension direction of the start and end lines to obtain the extended route. Here, the extension direction of the start and end lines is the Y-axis direction of the image coordinate system. The adjacent route points arranged in vertical order are connected in sequence, and the resulting curve or broken line can completely fit the actual direction of the crack from the start point to the end point. For example, when the tunnel crack bends to the left in a certain section and shifts to the right in another section, the connected extended route will synchronously show the corresponding bending shape, rather than a simple straight line. The extended route, which is spliced ​​from refined route points, can provide "close-fitting" movement path guidance for the subsequent high-precision acquisition unit, ensuring that the acquisition unit is always aligned with the core area of ​​the crack, avoiding monitoring omissions caused by path deviation, and significantly improving the accuracy and completeness of intelligent monitoring of railway tunnel cracks.

[0025] Step S2 includes the following steps: The high-precision acquisition unit is controlled to move along the extended route from the starting point until the starting point is located within the window frame of the acquisition window of the high-precision acquisition unit.

[0026] For example, in this embodiment, by controlling the high-precision acquisition unit to move along the extension route with the starting point as a reference and setting the window frame termination condition, a standardized monitoring starting position can be established, providing a unified benchmark for subsequent segmented acquisition and image stitching, ensuring the orderliness and completeness of crack monitoring. It can be explained that the high-precision acquisition unit is a device used to acquire high-definition images of tunnel cracks (e.g., high-definition industrial cameras, 3D scanners, etc.), and its acquisition accuracy directly affects the capture effect of crack details. The starting point is the previously determined initial endpoint of the tunnel crack extension, and the extension route is the trajectory line that conforms to the actual direction of the crack. By using the starting point as the starting point of movement, it can be ensured that the monitoring action of the high-precision acquisition unit starts from the initial end of the crack, avoiding the omission of the initial segment caused by cutting into the middle area of ​​the crack during monitoring. Extending the route ensures that the acquisition unit always adjusts its position around the core area of ​​the crack, avoiding defocusing of the acquisition target due to deviation from the route, and laying the positional foundation for subsequent accurate acquisition of crack images. In addition, the acquisition window is the field of view of the high-precision acquisition unit, and the window frame is the edge boundary of this field of view (e.g., the left, right, top, and bottom edge lines of the acquisition window). When the high-precision acquisition unit moves to the starting point and falls exactly on the window frame, it means that the acquisition unit has reached the preset initial monitoring position. At this time, the field of view of the acquisition window not only includes the starting segment of the crack corresponding to the starting point, but also reserves subsequent field of view space for continued acquisition along the extended route. This allows different crack monitoring tasks or multiple monitoring of the same crack to start with a unified initial field of view, avoiding the confusion of monitoring benchmarks caused by the randomness of the initial position.

[0027] Here, the high-precision acquisition unit can be carried by a drone for corresponding movement. The drone mentioned in this embodiment can refer to the same device or different devices.

[0028] It can be further explained that, based on the above process, the high-precision acquisition unit obtains a standardized monitoring starting posture: starting from the beginning of the crack, it is adjusted along the crack direction to the position where the edge of the acquisition window is aligned with the starting point. Based on the setting of this starting position, it is ensured that the monitoring starts from the very front of the crack without missing the starting segment, and it also provides a clear boundary reference for defining the acquisition range of each subsequent moving acquisition. For example, during subsequent moving acquisition, the starting point on the current window frame can be used as a reference, and the other side of the window away from the starting point can be used as a new connection boundary to ensure that images from different acquisition times can be connected in an orderly manner along the extension route, avoiding repeated acquisition or omission of crack segments. This provides an orderly starting basis for the entire intelligent monitoring process of tunnel cracks and significantly improves the standardization and completeness of the monitoring.

[0029] Since the high-precision acquisition unit moves to the point where the starting point is located within the frame of the acquisition window, directly acquiring data based on the current acquisition magnification may result in an excessively narrow field of view within the acquisition window due to an excessively high acquisition magnification. For example, if the width of a tunnel crack is large or the acquisition magnification is too high, the edge of the tunnel crack may extend beyond the lateral boundary of the acquisition window, making it impossible to capture the entire lateral view of the crack within the acquisition window in a single acquisition. This ultimately reduces the completeness and accuracy of crack monitoring. Therefore, to solve this problem, in this embodiment, the aforementioned "controlling the high-precision acquisition unit to move along the extended route from the starting point until the starting point is located within the frame of the acquisition window of the high-precision acquisition unit" may further include the following steps: The window frame line of the acquisition window with the starting point located at the high-precision acquisition unit is determined as the adjustment frame line, and the two window frame lines perpendicular to the adjustment frame line are determined as the verification frame lines; The response determines that any check frame line coincides with a tunnel crack based on the acquisition window. The acquisition multiplier of the corresponding acquisition window is then reduced by the same baseline multiplier to obtain the acquisition multiplier corresponding to different reduction times. The response determines that each check frame does not coincide with the tunnel crack based on the acquisition multiple corresponding to any decreasing number of acquisitions. It controls the high-precision acquisition unit to move away from the starting point along the extension route until the starting point is located at the adjustment frame.

[0030] For example, in this embodiment, the position adjustment of the high-precision acquisition unit can be achieved based on the following method steps: First, the server can define the window frame line of the acquisition window where the starting point is located as the adjustment frame line, and define the two window frame lines perpendicular to the adjustment frame line as the verification frame lines. Here, the adjustment frame line is the edge of the acquisition window where the starting point is located (for example, if the starting point is located on the upper frame line of the window, then the lower frame line is the adjustment frame line), which serves as the reference boundary for subsequent position calibration. The verification frame line is the window edge perpendicular to the adjustment frame line (for example, when the adjustment frame line is the left frame line, the verification frame line consists of the upper and lower frame lines), used to determine whether the lateral field of view of the acquisition window can completely cover the tunnel crack. By clearly defining the functional division of the frame lines, a clear judgment object is provided for subsequent magnification adjustment and position calibration, avoiding operational confusion caused by confusion of frame line functions, and ensuring that the adjustment and calibration actions have a clear target orientation. Secondly, based on the acquisition window, if any verification frame line coincides with a tunnel crack, the server can decrease the acquisition multiplier of the corresponding acquisition window by the same baseline multiplier to obtain acquisition multipliers corresponding to different decrease numbers. It can be explained that when any verification frame line coincides with a tunnel crack, it means that the current acquisition multiplier is too high and the field of view is too narrow, and the lateral edge of the tunnel crack exceeds the acquisition window range. The same baseline multiplier is a preset fixed decrease range (such as decreasing by 0.2 times each time). By gradually reducing the acquisition multiplier according to this range, the field of view of the acquisition window can be gradually expanded. Based on the step-by-step decrease method, the degree of field of view expansion can be precisely controlled, avoiding excessive adjustment of the multiplier at one time, which would lead to field of view redundancy or image clarity reduction. This ensures that sufficient acquisition accuracy is maintained while expanding the field of view, providing suitable field of view conditions for fully capturing the lateral details of the crack. Finally, based on the acquisition multiplier corresponding to any decreasing number of acquisitions, if each verification frame does not coincide with the tunnel crack, the server can control the high-precision acquisition unit to move away from the starting point along the extension route until the starting point is located at the adjustment frame line. That is, when all verification frames do not coincide with the tunnel crack, it indicates that the field of view at the current acquisition multiplier can completely cover the lateral range of the tunnel crack. At this time, by moving the high-precision acquisition unit away from the starting point along the extension route, the acquisition window can be adjusted to a better monitoring starting position while maintaining the integrity of the field of view. The termination condition of "until the starting point is located at the adjustment frame line" ensures that the adjustment frame line is still used as the reference after the movement, maintaining the calibration standard consistent with the initial position setting. This not only solves the problem of incomplete crack coverage caused by the narrow field of view, but also ensures that the high-precision acquisition unit always uses the adjustment frame line as the reference through position calibration. This provides a stable position reference for the orderly connection of images when moving and acquiring along the extension route, avoiding the omission or repeated acquisition of tunnel crack segments due to position offset, and significantly improving the integrity and accuracy of intelligent monitoring of railway tunnel cracks.

[0031] Step S3 includes the following steps: The sub-route corresponding to the acquisition window of the extended route is determined, and the first and second lateral points with distribution characteristics located on both sides of the sub-route are determined based on the tunnel cracks.

[0032] For example, in this embodiment, by dividing the acquisition window into sub-routes and locating the identifying lateral points on both sides of the crack, a clear path segment and positional reference are provided for subsequent precise illumination and image acquisition, ensuring the targeted monitoring and image quality of the tunnel crack. It can be explained that the extended route is a complete trajectory line running through the entire tunnel crack, and the acquisition window is the current field of view of the high-precision acquisition unit. The sub-route is a segment of the extended route within the acquisition window; it is the core monitoring object of the current acquisition action. By determining the sub-route, the complete extended route can be broken down into several segments adapted to the acquisition window, ensuring that each acquisition focuses on a specific crack segment, avoiding over-acquisition of non-crack areas due to ambiguous acquisition range, improving the targeted nature of the acquisition, and ensuring that the acquisition action accurately corresponds to the local area of ​​the crack. Furthermore, since the tunnel crack is located on both sides of the sub-route... Since the cracks are distributed laterally, the lateral points with distinctive distribution characteristics are key locations that reflect the lateral boundary features of the cracks (e.g., the characteristic endpoints of the left and right edges of the tunnel crack, or the two endpoints of the widest lateral point of the tunnel crack). This allows for precise definition of the lateral range of the crack within the corresponding section of the sub-route, providing clear location markers for subsequent operations. In other words, it can be understood that the sub-route specifies "which crack section to collect," and the first and second lateral points specify "where the lateral range of the tunnel crack is." This ensures precise focusing of the collection range, avoids resource waste, and provides reliable positioning data for the lateral illumination units, guaranteeing the effectiveness of supplemental lighting. This, in turn, improves the clarity of the collected images and the detail integrity of the tunnel cracks, providing high-quality image data support for intelligent monitoring of railway tunnel cracks and significantly enhancing the accuracy and reliability of monitoring.

[0033] Furthermore, in this embodiment, the aforementioned "determining the sub-route corresponding to the acquisition window of the extended route, and determining the first lateral point and the second lateral point with distribution characteristics located on both sides of the sub-route based on the tunnel crack" may also include the following steps: The sub-path is divided into arrays, and each second array line is generated along a direction perpendicular to the tangent direction of the sub-path for each second array point obtained by the respective path; Determine the first intersection point and the second intersection point where each second array line intersects with the crack profile of the tunnel crack, and determine the intersection distance between the first intersection point and the second intersection point; The maximum intersection distance is defined as the maximum distance. Then, with the second array line corresponding to the maximum distance as the center, a calculation process is constructed to calculate the difference between the corresponding intersection distance and the maximum distance in turn, and different distance differences are obtained. If any distance difference is less than a preset difference, the second array line corresponding to that distance difference is determined as the distribution characteristic line; otherwise, the calculation process is stopped. The first and second intersection points of the distribution characteristic lines corresponding to the center positions along the sub-route are determined as the first and second lateral points with distribution characteristics.

[0034] For example, in this embodiment, the determination of the sub-route, the first lateral point, and the second lateral point can be specifically implemented based on the following method steps: First, the server can divide the sub-route into arrays and generate second array lines for each second array point obtained through the path along a direction perpendicular to the tangent direction of the sub-route. As mentioned above, a sub-route is a segment of the extended route located within the acquisition window. Array division involves uniformly dividing the sub-route into multiple second array points with consistent intervals (e.g., setting one point every 5 pixels) to ensure that the lateral distribution analysis of the entire sub-route is comprehensive. The tangent direction of the sub-route is its extension direction. A straight line passing through each second array point is generated along a direction perpendicular to the tangent (i.e., lateral), which is the second array line. It can be explained that the second array line can densely cover the crack segment corresponding to the sub-route, just like a "lateral detection line," to achieve point-by-point exploration of the lateral width of the tunnel crack, avoiding the omission of distribution features due to sparse analysis points, and laying the foundation for subsequent accurate capture of the lateral boundary of the tunnel crack. Secondly, the server can determine the first and second intersection points of each second array line intersecting with the crack outline of the tunnel crack, and determine the intersection distance between the first and second intersection points. Here, the crack outline is the edge outline of the tunnel crack in the acquisition window. Each second array line intersects with this outline to form two intersection points (i.e., the first intersection point and the second intersection point), which correspond to the two sides of the tunnel crack respectively. The intersection distance is the straight-line distance between these two intersection points, which directly reflects the actual width of the tunnel crack at this lateral position. That is, by calculating the intersection distance on all second array lines, the distribution variation law of the tunnel crack width along the sub-route can be quantified, providing objective data support for screening key lateral points and avoiding point deviation caused by relying solely on visual judgment. Next, the server can determine the maximum intersection distance as the maximum distance, and use the second array line corresponding to the maximum distance as the center to construct a calculation process that calculates the difference between the corresponding intersection distance and the maximum distance in turn, obtaining different distance differences. Here, the maximum distance is the maximum value among all intersection distances, corresponding to the second array line at the widest point of the crack. This position can most clearly reflect the lateral distribution range of the tunnel crack, so it is used as the central reference. The difference between the intersection distance and the maximum distance (i.e., the distance difference) is calculated sequentially from the center to the second array lines on both sides, which can intuitively show the shrinking trend of the tunnel crack width from the widest point to both sides, providing a gradient basis for defining the effective lateral distribution range of the tunnel crack. Then, in response to any distance difference less than a preset difference, the server can identify the second array line corresponding to that distance difference as a distribution feature line; otherwise, the calculation process stops. It can be explained that the preset difference is a threshold (e.g., 2 pixels) pre-set based on the monitoring accuracy of tunnel cracks. When the distance difference is less than this threshold, it means that the width of the tunnel crack at that location is very close to the widest point and is still within the core distribution area of ​​the tunnel crack. Identifying this type of array line as a distribution feature line can accurately delineate the effective range of the lateral distribution of tunnel cracks, avoiding the mistaken inclusion of excessively narrow edge areas into the core monitoring range. At the same time, stopping the calculation when the distance difference is greater than or equal to the preset difference can promptly define the boundary of the distribution feature line, ensuring the accuracy and efficiency of the screening results. Finally, the first and second intersection points of the distribution feature lines corresponding to the center position along the sub-route are determined as the first and second lateral points with distribution characteristics. Here, based on the above, it is known that the second array line determined as the distribution feature line is still in the core distribution area of ​​tunnel cracks. Therefore, the distribution feature line at the center position can characterize the lateral distribution of tunnel cracks. Thus, the two intersection points of the distribution feature lines corresponding to the center position can be determined as the first and second lateral points, which can truly reflect the lateral distribution characteristics of tunnel cracks and provide a reliable position reference for subsequent image acquisition and parameter analysis of tunnel cracks, significantly improving the accuracy and reliability of intelligent monitoring of railway tunnel cracks.

[0035] It can be explained that when the number of line segments corresponding to all distribution characteristic lines of the same sub-route is even, the first intersection point and the second intersection point of the two distribution characteristic lines with the largest distance difference at the corresponding center position can be determined as the first lateral point and the second lateral point with distribution characteristics.

[0036] Step S4 includes the following: The high-precision acquisition unit is controlled to move the lateral illumination unit it carries to coincide with the lateral point, and responds to the lateral illumination unit to provide illumination. The high-precision acquisition unit is then controlled to acquire images of the corresponding sub-route.

[0037] For example, in this embodiment, the transverse illumination unit can be precisely controlled to align with the transverse points and the acquisition can be triggered by illumination actions to ensure that the acquired sub-route images clearly show crack details, providing high-quality data support for subsequent monitoring and analysis, and improving the effectiveness of intelligent monitoring of tunnel cracks. It can be noted that the transverse illumination unit is a device equipped in the high-precision acquisition unit used to provide transverse directional illumination, and its illumination range directly affects the brightness and contrast of the crack area. The transverse points are the previously determined first and second transverse points located on both sides of the sub-route, which define the transverse core distribution range of the crack. By controlling the transverse illumination unit to move to coincide with the transverse points, the illumination can be precisely focused on the transverse boundary and core area of ​​the crack. To avoid over-illumination of non-crack areas or insufficient illumination of crack areas due to lighting deviation, for example, when the first and second lateral points correspond to the left and right edges of the crack respectively, the light emitted by the lateral illumination unit after moving to coincide with these two points can evenly cover the lateral range of the crack, enhance the contrast between the crack and the tunnel wall, and lay the lighting foundation for clearly capturing the crack outline. This ensures that the crack area is in a sufficient and suitable lighting environment when the high-precision acquisition unit acquires images, avoiding blind acquisition when the lighting is unstable or does not cover the target area. This ensures that each acquired sub-route image has high definition and high contrast, providing reliable image data for subsequent intelligent monitoring links such as crack parameter measurement and status analysis, and significantly improving the accuracy and effectiveness of intelligent monitoring of railway tunnel cracks.

[0038] It can be explained that the horizontal illumination unit can be specifically an LED lamp bead or other light-emitting element, which can be slidably connected to the horizontal moving slide rails pre-set on both sides of the high-precision acquisition unit. There can be two horizontal illumination units, which are respectively set on the horizontal moving slide rails on both sides, so that the horizontal illumination unit can slide based on the horizontal moving slide rails.

[0039] Furthermore, in this embodiment, the aforementioned "controlling the lateral illumination unit carried by the high-precision acquisition unit to move to coincide with the lateral point" may further include the following steps: If the center point of the response acquisition window does not coincide with the feature intersection point where the distribution feature line of the corresponding horizontal point intersects with the sub-line segment, the high-precision acquisition unit is controlled to move along the direction from the center point of the window toward the feature intersection point until the center point of the window coincides with the feature intersection point. Update the acquisition multiplier of the corresponding acquisition window until the starting point is located at the adjustment frame line, and obtain the corresponding updated value. The unit length between each horizontal point and the sub-route is determined based on the distribution feature line, and the update factor is determined based on the number of decreases corresponding to the acquisition factor of the acquisition window and the update value. The unit length is updated by a corresponding update factor, and the lateral illumination unit carried by the high-precision acquisition unit is controlled to move towards different lateral points based on the actual length of each lateral point.

[0040] For example, in this embodiment, the control and movement of the lateral illumination unit can be specifically based on the following method steps: First, if the center point of the acquisition window does not coincide with the intersection point of the distribution feature line and the sub-segment of the corresponding lateral point, the server can control the high-precision acquisition unit to move along the direction from the center point of the window towards the intersection point until the center point of the window coincides with the intersection point. It can be explained that the distribution feature line is a pre-determined straight line that reflects the lateral distribution characteristics of tunnel cracks, and the sub-segment is the sub-path within the acquisition window. The intersection point is the point where the distribution feature line and the sub-segment intersect, serving as the core reference point for the lateral distribution of tunnel cracks. When the center point of the window deviates from this point, the field of view reference of the acquisition window is biased, and the coordinate positioning of the lateral point lacks a reliable reference. In this case, the server can control the high-precision acquisition unit to move towards the intersection point and make them coincide, calibrating the center of the field of view of the acquisition window to the core position of the lateral distribution of tunnel cracks. This ensures that the coordinate description of the lateral point in the acquisition window is accurate and consistent, providing a reliable reference for subsequent position calculations and avoiding positioning errors caused by field of view reference offset. Secondly, the server can update the acquisition multiplier of the corresponding acquisition window until the starting point is located at the adjustment frame line, and obtain the corresponding updated value. Based on the above, it can be seen that the adjustment frame line is the frame line of the acquisition window where the starting point is located, which is a fixed reference for position calibration. After updating the acquisition multiplier (such as increasing or decreasing the multiplier according to the monitoring accuracy requirements), the high-precision acquisition unit needs to be moved to make the starting point return to the adjustment frame line, ensuring that the position reference of the acquisition window is always aligned with the adjustment frame line. The obtained updated value is the change parameter before and after the acquisition multiplier update (for example, if the acquisition multiplier is updated from 1.0 to 1.5, the updated value is 0.5). This value can quantify the impact of the multiplier change on the image scale, providing data basis for subsequent correction of the distance parameter of the horizontal point, and avoiding distance calculation distortion caused by the multiplier change. Next, the server can determine the unit length between each horizontal point and the sub-route based on the distribution feature line, and determine the update factor based on the number of decreases corresponding to the acquisition factor of the acquisition window and the update value. It can be explained that the unit length is the distance from the horizontal point along the distribution feature line to the sub-route (based on the image scale of the current acquisition factor); the number of decreases in the acquisition factor is the number of operations when adjusting the acquisition factor previously. Combined with the update value, an update factor reflecting the actual image scale can be calculated. Here, the factor is the calibrated actual distance conversion coefficient, which can correct the distance deviation caused by the change in acquisition factor. That is, through this calculation, the unit length based on the image scale can be converted into a distance parameter that fits the actual physical space, solving the problem of inaccurate distance positioning caused by image scale distortion, and providing a precise numerical reference for the movement distance of the illumination unit. Finally, the server can update the unit length according to the corresponding update factor and control the lateral illumination unit carried by the high-precision acquisition unit to move towards different lateral points based on the actual length of each lateral point. Here, multiplying the unit length by the update factor yields the actual length between the lateral point and the sub-route (i.e., the actual distance the illumination unit needs to move). Then, the illumination unit can be controlled to move this actual length in the direction towards the lateral point, ensuring that the illumination unit accurately coincides with the lateral point. For example, if the unit length is 10 pixels and the update factor is 0.8 (corresponding to 1 pixel = 0.8 mm), the actual length is 8 mm. Moving the illumination unit 8 mm will accurately align it with the lateral point. The movement control with benchmark calibration and parameter correction completely solves the problem of illumination unit positioning deviation, ensuring that the illumination can accurately cover the lateral core area of ​​the crack, providing clear imaging conditions for the high-precision acquisition unit, and significantly improving the accuracy and reliability of intelligent monitoring of cracks in railway tunnels.

[0041] Furthermore, in this embodiment, if only the horizontal illumination unit is used to supplement illumination and acquire images, the coordinate deviation between the center point of the acquisition window and the vertical key points of the sub-route (e.g., the starting point, the end point of the sub-route) may lead to insufficient illumination or image offset in the crack area near the vertical point. Simultaneously, the lack of a directional illumination and supplementary acquisition mechanism for vertical points will cause the crack details around the vertical point in the acquired image (e.g., the vertical extension shape and width variation of tunnel cracks) to be blurred or missing, failing to fully reflect the overall crack appearance corresponding to the sub-route, ultimately reducing the integrity of the image data and the reliability of subsequent analysis. Therefore, based on a similar approach to the horizontal illumination unit, the aforementioned "responding to the horizontal illumination unit to provide illumination, controlling the high-precision acquisition unit to acquire images, and obtaining the acquired images of the corresponding sub-route" may further include the following steps: Establish a real coordinate system, and define the starting point and the route endpoints far from the starting point, determined based on the sub-route, as the first vertical point and the second vertical point, respectively, wherein any distribution feature line coincides with the X-axis of the real coordinate system; In response to the existence of a horizontal coordinate difference between any vertical point and the center point of the window, the high-precision acquisition unit is controlled to move horizontally according to the corresponding coordinate difference, and the vertical coordinate difference between the vertical point and the center point of the window is obtained. The high-precision acquisition unit controls the vertical illumination unit carried by the high-precision acquisition unit to move towards the vertical point based on the vertical coordinate difference, and responds to the vertical illumination unit to illuminate based on the vertical point, controls the high-precision acquisition unit to acquire data, and updates the acquired image based on the obtained vertical supplementary image to obtain the updated acquired image.

[0042] For example, in this embodiment, the vertical key points of the sub-route are illuminated based on the set vertical illumination unit to ensure the integrity of the image data. This can be achieved specifically based on the following method steps: First, the server can establish a corresponding real-world coordinate system, defining the starting point and the endpoint of the sub-route far from the starting point as the first vertical point and the second vertical point, respectively. Any distribution feature line coincides with the X-axis of the real-world coordinate system. This real-world coordinate system, a two-dimensional coordinate system based on the actual space of the tunnel, can transform abstract coordinates in the image into real spatial locations. Since the first vertical point corresponds to the starting point and the second vertical point corresponds to the endpoint of the sub-route far from the starting point, these are the key boundary points of the sub-route in the vertical direction. Therefore, coinciding the distribution feature line (the straight line reflecting the transverse distribution of cracks) with the X-axis (transverse axis) can align the coordinates of the crack's transverse and vertical positions, providing a unified standard for subsequent coordinate difference calculations. Establishing this coordinate system provides a spatial reference for accurate description and deviation correction of point locations, avoiding positioning errors caused by a chaotic coordinate system and ensuring that the positional analysis of vertical points has practical significance. Secondly, in response to the existence of a horizontal coordinate difference between any vertical point and the center point of the window, the server can control the high-precision acquisition unit to move horizontally according to the corresponding horizontal coordinate difference, and obtain the vertical coordinate difference between the vertical point and the center point of the window. Based on the above, it can be seen that the center point of the window is the geometric center of the acquisition window. If there is a horizontal coordinate difference (distance in the X-axis direction) between the first or second vertical point and this point, it means that the horizontal position of the acquisition window deviates from the vertical point position, which will lead to incomplete imaging of the area around the vertical point position. Therefore, the high-precision acquisition unit can be controlled to move horizontally according to the corresponding difference, so that the horizontal position of the center point of the window and the vertical point can be aligned, correcting the horizontal offset. At the same time, the vertical coordinate difference (distance in the Y-axis direction) can be obtained, which can clarify the vertical position relationship between the vertical point position and the center point of the window in the vertical direction, providing accurate distance basis for the subsequent movement of the vertical illumination unit, and avoiding illumination deviation caused by unclear vertical position. Next, the vertical illumination unit carried by the high-precision acquisition unit is controlled to move towards the vertical point based on the vertical coordinate difference. In response to the vertical illumination unit illuminating the vertical point, the high-precision acquisition unit is controlled to acquire the image and obtain a vertical supplementary image. It can be noted that, similar to the horizontal illumination unit, the vertical illumination unit is also a device used to provide vertical directional illumination. Moving the vertical illumination unit along the direction towards the vertical point by the distance corresponding to the vertical coordinate difference allows the illumination to be precisely focused on the vertical point and the surrounding crack area, enhancing the brightness and contrast of the area and solving the problem of insufficient vertical illumination that may have existed with the previous horizontal illumination. By using the illumination action as the acquisition trigger condition, the vertical supplementary image is acquired under the most suitable illumination conditions. This image can clearly capture the crack details around the vertical point (e.g., the morphological changes of the tunnel crack at the end), making up for the lack of information in the vertical dimension of the original acquired image. Finally, the acquired image is updated based on the obtained vertical supplementary image, resulting in an updated acquired image. It can be seen that the vertical supplementary image focuses on clear details around the vertical points. Fusing and updating it with the original acquired image can supplement and improve blurry or missing vertical crack information in the original image. For example, if the crack edge near the second vertical point in the original image is blurry, replacing the corresponding area with clear details from the supplementary image can clearly present the complete shape of the crack corresponding to the sub-route from the first vertical point to the second vertical point. This integrates horizontal and vertical acquisition information, eliminates the limitations of single-light acquisition, and ensures that the final acquired image contains both horizontal crack details and complete coverage of the vertical boundary area. This provides comprehensive and clear image data support for intelligent monitoring of railway tunnel cracks, significantly improving the completeness and reliability of the monitoring results.

[0043] It can be noted that, similar to the horizontal illumination unit, the vertical illumination unit can also be specifically an LED lamp bead or other light-emitting element, which can be slidably connected to the vertical moving slide rails pre-set on both sides of the high-precision acquisition unit. The vertical moving slide rails are perpendicular to the horizontal moving slide rails, and the corresponding vertical illumination unit can also include two units, which are respectively set on the vertical moving slide rails on both sides, so that the vertical illumination unit can slide based on the vertical moving slide rails.

[0044] Figure 2 The diagram shows the structure of the high-precision acquisition unit and the illumination unit in this embodiment. Figure 2 As shown, the high-precision acquisition unit is located at the center, specifically... Figure 2 In the high-precision acquisition unit, the vertical and horizontal illumination units are located around the high-precision acquisition unit, corresponding to L1, L2, L3 and L4 respectively. Each illumination unit can change its position based on an individual moving slide rail. The length of each moving slide rail can be preset, but this embodiment does not limit the specific value.

[0045] Furthermore, in this embodiment, the aforementioned "updating the acquired image based on the obtained vertical supplementary image to obtain the updated acquired image" may further include the following steps: The image portion that is the same as the vertically supplemented image and the acquired image is determined as the region to be supplemented, and the first contrast and second contrast of different image pixels that make up the region to be supplemented are determined based on the vertically supplemented image and the acquired image, respectively. Image pixels in the vertical supplementary image whose first contrast is greater than the second contrast are identified as supplementary pixels, and supplementary pixels in adjacent positions are connected to obtain each supplementary sub-region. The corresponding region to be supplemented in the acquired image is replaced based on each supplementary sub-region to obtain the updated acquired image.

[0046] For example, in this embodiment, updating the acquired image based on the vertical supplementary image may specifically include the following method steps: First, the server can identify the identical portion of the vertically supplemented image and the acquired image as the area to be supplemented. Then, based on the vertically supplemented image and the acquired image, it determines the first contrast and second contrast of different image pixels that make up the area to be supplemented. It can be explained that the area to be supplemented is the overlapping portion of the vertically supplemented image and the acquired image. Accurately locating this area avoids interference with clear, non-overlapping areas, ensuring that updates only apply to the overlapping range that needs optimization. Contrast is a quantitative indicator of the difference in brightness between pixels in an image, directly reflecting the distinction between the crack outline and the background. Specifically, the first contrast is the contrast of each pixel in the area to be supplemented in the vertically supplemented image, and the second contrast is the contrast of the corresponding pixel in the acquired image. By calculating the contrast of both separately, the clarity of the same pixel location in the two images can be objectively determined, providing data for subsequent selection of high-quality pixels and avoiding selection bias caused by relying solely on visual judgment. Secondly, the server can identify image pixels in the vertical supplementary image whose first contrast is greater than the second contrast as supplementary pixels, and connect supplementary pixels in adjacent positions to obtain supplementary sub-regions. That is, when the first contrast of a pixel in the vertical supplementary image is higher than the second contrast of the acquired image, it means that the pixel is clearer in the supplementary image, which may be due to the illumination based on the vertical illumination unit. Therefore, it can be identified as a supplementary pixel to filter out high-quality details in the supplementary image. By connecting adjacent supplementary pixels to form supplementary sub-regions, scattered high-quality pixels can be integrated into continuous area blocks, avoiding image splicing traces caused by scattered replacement, ensuring a natural transition of the updated image, and clarifying the specific range that needs to be replaced, thus improving the accuracy of the update operation. Finally, the server can replace the corresponding area to be supplemented in the acquired image based on each supplementary sub-region to obtain an updated acquired image. It can be explained that replacing the corresponding area (low-contrast blurry details) in the acquired image with the supplementary sub-region (high-contrast clear details) in the vertical supplementary image can achieve the image fusion effect of "selecting the best and removing the worst". This allows the updated image to retain the original clear information of other areas of the acquired image while supplementing the crack details around the vertical points. This avoids the image quality degradation caused by overall superposition and maximizes the use of the advantageous information of the supplementary image. It ensures that the updated acquired image achieves detail optimization in the area to be supplemented, and presents the complete and clear picture of the cracks corresponding to the sub-route. This provides high-quality image support for subsequent intelligent monitoring links such as crack width measurement and morphological analysis, and significantly improves the accuracy and reliability of railway tunnel crack monitoring.

[0047] Step S5 may also include the following: Based on the sub-route, the route endpoint far from the starting point is determined as the new starting point. The above steps are repeated until the termination point is located in the image acquired for any number of acquisitions, thus obtaining the detection data of the corresponding tunnel crack.

[0048] For example, in this embodiment, by dynamically updating the starting point, cyclically executing the acquisition process, and setting termination conditions, the embodiment achieves full-coverage monitoring of tunnel cracks from the starting point to the termination point, ensuring that the acquired detection data is complete and coherent, providing a comprehensive basis for crack assessment, and ensuring that each acquisition sub-route can be connected in an orderly manner along the extension route of the crack, avoiding duplication or breakage of the acquisition range. For example, if the initial starting point is crack start point A, and the first acquisition sub-route is AB, after setting B as the new starting point, the next acquisition sub-route can extend from B to C, achieving seamless connection of crack segments, laying the positional foundation for full-process monitoring, and finally obtaining the detection data of the corresponding tunnel crack.

[0049] Furthermore, in this embodiment, the aforementioned "obtaining detection data of corresponding tunnel cracks" may further include the following steps: Determine the acquisition factor for images acquired at different acquisition times, and set the smallest acquisition factor as the stitching factor; The captured images are reduced by the corresponding stitching factor using the image center point as the center to obtain the updated captured images; Create vertically arranged image fill slots on the data display interface, and fill each captured image into different image fill slots from top to bottom according to the capture order. The sub-routes of the captured images corresponding to different image fill slots are presented as line segment connections. Create filling slots of various lengths that are horizontally arranged with each image filling slot, and fill the filling slots of different lengths sequentially from top to bottom according to the acquisition order of the same acquired image, so as to obtain the detection data of the corresponding tunnel cracks composed of the data display interface.

[0050] For example, in this embodiment, the acquisition of detection data can be specifically implemented based on the following method steps: First, the server can determine the acquisition magnification for images acquired at different acquisition times, and set the minimum acquisition magnification as the stitching magnification. It should be noted that the acquisition magnification of images acquired at different acquisition times may have been adjusted due to the aforementioned monitoring accuracy requirements (for example, some images are 1.5x and some are 1.0x), resulting in inconsistent image scaling ratios. The minimum acquisition magnification ensures that all images retain their original content after scaling, avoiding the cropping of some crack details due to excessive magnification. By setting it as the stitching magnification, a standard benchmark is provided for subsequent unified image size, ensuring that images acquired at different acquisition times have a consistent scale after scaling, laying the foundation for subsequent orderly stitching, and solving the problem of inconsistent image size display. Secondly, the server can scale down each acquired image by a corresponding stitching factor, using the image center point as the center, to obtain an updated acquired image. Using the image center point as the scaling center ensures that the core area of ​​the image (i.e., the crack segment corresponding to the sub-route) is always in the center position of the image, preventing the crack from deviating from the field of view after scaling. Scaling down the image by the stitching factor ensures that all updated acquired images have a uniform size specification. For example, images with a 1.5x scaling factor and images with a 1.0x scaling factor are both scaled down by 1.0x (stitching factor), resulting in consistent size and stable core crack area position, providing suitable image material for subsequent orderly filling and arrangement. Next, the server can create vertically arranged image fill slots on the data display interface, and fill each acquired image into different image fill slots sequentially from top to bottom based on the acquisition order. The sub-routes of the acquired images corresponding to different image fill slots are presented as line segments. It can be explained that the vertically arranged image fill slots provide an orderly display space for the acquired images. Filling based on the acquisition order (i.e., the order in which the high-precision acquisition unit moves along the extension route) can restore the extension process of the crack from the starting point to the ending point. The sub-routes are presented as line segments, which means that the end of the sub-routes of the previous image and the beginning of the sub-routes of the next image are connected in the display interface, intuitively presenting the continuous shape of the crack. Thus, scattered images can be transformed into "visualized crack chains", solving the problem that a single image cannot display the complete crack, enabling monitoring personnel to quickly grasp the overall direction and segment details of the crack. Finally, the server can further create horizontally arranged length fill slots corresponding to each image fill slot, and fill the different length fill slots sequentially from top to bottom based on the acquisition order of the same acquired image. This results in the detection data of the corresponding tunnel cracks, presented in a data display interface. In this data display, the horizontally arranged length fill slots correspond one-to-one with the image fill slots, ensuring that each crack image can be directly associated with its actual length data (for example, the actual length of the crack corresponding to a certain sub-route is 0.8 meters). The length data is filled according to the acquisition order, maintaining consistency with the image arrangement logic, achieving a clear association of "one image corresponding to one length". This allows the detection data to include both the visual morphology of the tunnel crack and precise dimensional information. Monitoring personnel can quickly obtain complete information about each crack segment without repeated comparisons, significantly improving the readability and utilization efficiency of the data. This provides intuitive and reliable comprehensive detection data support for the assessment, analysis, and decision-making of railway tunnel cracks.

[0051] In summary, this embodiment effectively overcomes the bottleneck of insufficient accuracy in tunnel crack monitoring under manual inspection mode, significantly improves the accuracy and reliability of crack parameter measurement, and provides high-quality data support for tunnel structural defect assessment. Specifically, this can be reflected in the following aspects: 1. In terms of precise crack detail capture, this embodiment completely solves the problems of uneven lighting and difficulty in identifying minute cracks during manual inspection through the collaborative design of lateral illumination adaptation and high-precision acquisition. Specifically, based on the first and second lateral points on both sides of the sub-route, the lateral illumination unit can be precisely moved to the target position and turned on to form a directional light source perpendicular to the crack direction. Thus, based on the directional illumination, a clear contrast of light and dark can be formed at the crack edge, clearly distinguishing minute cracks from interference features such as stains and scratches on the lining surface, avoiding the problem of blurred details caused by the scattered flashlight light during manual inspection. At the same time, the high-precision acquisition unit performs image acquisition under the support of directional illumination, which can completely preserve the microscopic features such as the edge contour and width change of the crack. The detail clarity of the acquired image is dozens of times higher than that of manual observation, laying the foundation for the subsequent accurate measurement of crack width and shape. 2. In terms of continuous monitoring of the entire crack, this embodiment achieves complete coverage of the crack extension trajectory through route tracking and segment-by-segment iterative acquisition mode. First, the starting point, ending point and extension route of the crack are identified. Then, the high-precision acquisition unit is controlled to move along the route segment by segment. Sub-routes are divided with acquisition windows as units. After each segment of sub-route acquisition is completed, the starting point is updated and the operation is repeated until the entire crack is covered. Continuous acquisition images of the crack from the beginning to the end can be obtained, which avoids the omission of the middle section of the crack and can accurately capture the changes in the direction of crack extension. This solves the technical problem that it is difficult to form a complete crack data chain by manual inspection. 3. Regarding the improvement of measurement accuracy, this embodiment significantly reduces measurement errors through clear positioning benchmarks and data traceability. Using the crack extension route as a benchmark, the acquisition window is precisely aligned with the sub-route, ensuring that the shooting angle of each acquired image is always focused on the core area of ​​the crack, avoiding the perspective deviation that occurs when measuring with handheld tools. At the same time, the precise overlap of the lateral illumination unit and the lateral point ensures the consistency of illumination conditions for images acquired from different sub-routes, eliminating image comparison errors caused by illumination differences. Furthermore, the detection data acquired segment by segment can be spliced ​​together through sub-route association to form a complete digital archive of the crack. This not only allows for the direct extraction of macroscopic parameters such as crack length and extension direction, but also enables the calculation of width changes in each segment based on continuous images, thus improving monitoring accuracy.

[0052] Another embodiment of the present invention provides a railway tunnel crack monitoring system. Figure 3 Its corresponding system block diagram, such as Figure 3 As shown, the system includes: The route determination module is configured to determine the start point, end point, and extended route determined by the start point and end point of the corresponding tunnel crack. The acquisition and movement module is configured to control the high-precision acquisition unit to move along the extended route from the starting point until the starting point is located at the window frame of the acquisition window of the high-precision acquisition unit. The distribution determination module is configured to determine the sub-route corresponding to the acquisition window of the extended route, and to determine the first lateral point and the second lateral point with distribution characteristics located on both sides of the sub-route based on the tunnel crack; The image acquisition module is configured to control the lateral illumination unit carried by the high-precision acquisition unit to move to coincide with the lateral point, and respond to the lateral illumination unit to provide illumination, thereby controlling the high-precision acquisition unit to acquire images of the corresponding sub-route. The data integration module is configured to determine the route endpoints far from the starting point as new starting points based on the sub-route, repeat the above steps until the termination point is located in the image acquired for any number of acquisitions, and obtain the detection data of the corresponding tunnel crack.

[0053] In the specification provided herein, the algorithms and displays are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used with the examples of this invention. The required structure for constructing such systems is apparent from the above description. Furthermore, this invention is not directed to any particular programming language. It should be understood that the contents of the invention described herein can be implemented using various programming languages, and the above description of specific languages ​​is for the purpose of disclosing preferred embodiments of the invention.

[0054] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0055] Similarly, it should be understood that, in order to streamline this disclosure and aid in understanding one or more of the various aspects of the invention, in the above description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof.

[0056] Those skilled in the art will understand that modules, units, or components of the devices disclosed in the examples herein can be arranged in the devices described in this embodiment, or alternatively, can be located in one or more devices different from the devices in this example. The modules in the foregoing examples can be combined into a single module or, in addition, can be divided into multiple sub-modules.

[0057] Those skilled in the art will understand that the modules in the device of the embodiment can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiment can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components.

[0058] Furthermore, those skilled in the art will understand that although some embodiments described herein include certain features included in other embodiments but not others, combinations of features from different embodiments are meant to be within the scope of the invention and form different embodiments.

[0059] Furthermore, some of the embodiments described herein are methods or combinations of method elements that can be implemented by a processor of a computer system or by other means of performing the functions. Therefore, a processor having the necessary instructions for implementing the methods or method elements forms means for implementing the methods or method elements. Furthermore, the elements described herein in the apparatus embodiments are examples of means for implementing the functions performed by elements for the purposes of carrying out the invention.

[0060] As used herein, unless otherwise specified, the use of ordinal numbers such as “first,” “second,” “third,” etc., to describe ordinary objects merely indicates different instances of similar objects and is not intended to imply that the objects being described must have a given order in time, space, ordering, or any other manner.

[0061] Although the invention has been described with respect to a limited number of embodiments, those skilled in the art will understand from the foregoing description that other embodiments are conceivable within the scope of the invention described herein. Furthermore, it should be noted that the language used in this specification has been chosen primarily for readability and edibility purposes, and not for the purpose of explaining or limiting the subject matter of the invention.

Claims

1. A method for monitoring cracks in railway tunnels, characterized in that, include: Determine the starting and ending points of the corresponding tunnel cracks, as well as the extension route determined by the starting and ending points. The high-precision acquisition unit is controlled to move along the extended route from the starting point until the starting point is located at the window frame of the acquisition window of the high-precision acquisition unit. Determine the sub-route corresponding to the acquisition window of the extended route, and determine the first lateral point and the second lateral point with distribution characteristics on both sides of the sub-route based on the tunnel crack; The high-precision acquisition unit is controlled to move the lateral illumination unit to coincide with the lateral point, and the lateral illumination unit responds to the illumination, thereby controlling the high-precision acquisition unit to acquire images of the corresponding sub-route. Based on the sub-route, the route endpoint far from the starting point is determined as the new starting point. The above steps are repeated until the termination point is located in the image acquired for any number of acquisitions, thus obtaining the detection data of the corresponding tunnel crack.

2. The method according to claim 1, characterized in that, include: Determine the starting and ending points of the corresponding tunnel cracks, as well as the extension route determined by the starting and ending points, including: In response to receiving a crack monitoring signal carrying location information sent by any employee terminal, the system controls the drone to fly to the location information at a preset flight altitude to collect the information and sends the obtained flight image to the employee terminal. The system responds to the location determination signal sent by the receiving employee based on the flight image, determines the crack region located in the flight image that indicates the tunnel crack based on image recognition, and determines the two image contour points with the largest corresponding distance that make up the crack region as the start point and end point. The extension route of the corresponding tunnel crack is determined based on the established start and end lines connecting the starting and ending points.

3. The railway tunnel crack monitoring method according to claim 2, characterized in that, Based on the established start and end lines connecting the starting and ending points, the extension route of the corresponding tunnel crack is determined, including: Generate each first array line for each first array point obtained by arraying the start and end lines along a direction perpendicular to the start and end lines; Establish an image coordinate system, where the start and end lines coincide with the Y-axis of the image coordinate system; The average value of the corresponding horizontal coordinate values ​​of all image coordinate points that coincide with the region contour for each first array line is calculated, and the image coordinate points located in the crack region are determined as route points based on the obtained average horizontal coordinate values. Connect the route points that are adjacent to each other along the extension direction of the start and end lines to obtain the extended route.

4. The method according to claim 1, characterized in that, The high-precision acquisition unit is controlled to move along the extended route from the starting point until the starting point is located within the window frame of the acquisition window of the high-precision acquisition unit. This process then includes: The window frame line of the acquisition window with the starting point located at the high-precision acquisition unit is determined as the adjustment frame line, and the two window frame lines perpendicular to the adjustment frame line are determined as the verification frame lines; The response determines that any check frame line coincides with a tunnel crack based on the acquisition window. The acquisition multiplier of the corresponding acquisition window is then reduced by the same baseline multiplier to obtain the acquisition multiplier corresponding to different reduction times. The response determines that each check frame does not coincide with the tunnel crack based on the acquisition multiple corresponding to any decreasing number of acquisitions. It controls the high-precision acquisition unit to move away from the starting point along the extension route until the starting point is located at the adjustment frame.

5. The method according to claim 4, characterized in that, The sub-route corresponding to the acquisition window of the extended route is determined, and based on the tunnel cracks, the first lateral point and the second lateral point with distribution characteristics located on both sides of the sub-route are determined, including: The sub-path is divided into arrays, and each second array line is generated along a direction perpendicular to the tangent direction of the sub-path for each second array point obtained by the respective path; Determine the first intersection point and the second intersection point where each second array line intersects with the crack profile of the tunnel crack, and determine the intersection distance between the first intersection point and the second intersection point; The maximum intersection distance is defined as the maximum distance. Then, with the second array line corresponding to the maximum distance as the center, a calculation process is constructed to calculate the difference between the corresponding intersection distance and the maximum distance in turn, and different distance differences are obtained. If any distance difference is less than a preset difference, the second array line corresponding to that distance difference is determined as the distribution characteristic line; otherwise, the calculation process is stopped. The first and second intersection points of the distribution characteristic lines corresponding to the center positions along the sub-route are determined as the first and second lateral points with distribution characteristics.

6. The method according to claim 5, characterized in that, Controlling the lateral illumination unit carried by the high-precision acquisition unit to move until it coincides with the lateral point includes: If the center point of the response acquisition window does not coincide with the feature intersection point where the distribution feature line of the corresponding horizontal point intersects with the sub-line segment, the high-precision acquisition unit is controlled to move along the direction from the center point of the window toward the feature intersection point until the center point of the window coincides with the feature intersection point. Update the acquisition multiplier of the corresponding acquisition window until the starting point is located at the adjustment frame line, and obtain the corresponding updated value. The unit length between each horizontal point and the sub-route is determined based on the distribution feature line, and the update factor is determined based on the number of decreases corresponding to the acquisition factor of the acquisition window and the update value. The unit length is updated by a corresponding update factor, and the lateral illumination unit carried by the high-precision acquisition unit is controlled to move towards different lateral points based on the actual length of each lateral point.

7. The method according to claim 6, characterized in that, The system responds to illumination from the lateral illumination unit, controls the high-precision acquisition unit to acquire images of the corresponding sub-route, and then includes: Establish a real coordinate system, and define the starting point and the route endpoints far from the starting point, determined based on the sub-route, as the first vertical point and the second vertical point, respectively, wherein any distribution feature line coincides with the X-axis of the real coordinate system; In response to the existence of a horizontal coordinate difference between any vertical point and the center point of the window, the high-precision acquisition unit is controlled to move horizontally according to the corresponding coordinate difference, and the vertical coordinate difference between the vertical point and the center point of the window is obtained. The high-precision acquisition unit controls the vertical illumination unit carried by the high-precision acquisition unit to move towards the vertical point based on the vertical coordinate difference, and responds to the vertical illumination unit to illuminate based on the vertical point, controls the high-precision acquisition unit to acquire data, and updates the acquired image based on the obtained vertical supplementary image to obtain the updated acquired image.

8. The method according to claim 7, characterized in that, The acquired image is updated based on the obtained vertical supplementary image to obtain the updated acquired image, including: The image portion that is the same as the vertically supplemented image and the acquired image is determined as the region to be supplemented, and the first contrast and second contrast of different image pixels that make up the region to be supplemented are determined based on the vertically supplemented image and the acquired image, respectively. Image pixels in the vertical supplementary image whose first contrast is greater than the second contrast are identified as supplementary pixels, and supplementary pixels in adjacent positions are connected to obtain each supplementary sub-region. The corresponding region to be supplemented in the acquired image is replaced based on each supplementary sub-region to obtain the updated acquired image.

9. The method according to claim 6, characterized in that, Obtain the detection data for the corresponding tunnel cracks, including: Determine the acquisition factor for images acquired at different acquisition times, and set the smallest acquisition factor as the stitching factor; The captured images are reduced by the corresponding stitching factor using the image center point as the center to obtain the updated captured images; Create vertically arranged image fill slots on the data display interface, and fill each captured image into different image fill slots from top to bottom according to the capture order. The sub-routes of the captured images corresponding to different image fill slots are presented as line segment connections. Create filling slots of various lengths that are horizontally arranged with each image filling slot, and fill the filling slots of different lengths sequentially from top to bottom according to the acquisition order of the same acquired image, so as to obtain the detection data of the corresponding tunnel cracks composed of the data display interface.

10. A railway tunnel crack monitoring system, characterized in that, include: The route determination module is configured to determine the start point, end point, and extended route determined by the start point and end point of the corresponding tunnel crack. The acquisition and movement module is configured to control the high-precision acquisition unit to move along the extended route from the starting point until the starting point is located at the window frame of the acquisition window of the high-precision acquisition unit. The distribution determination module is configured to determine the sub-route corresponding to the acquisition window of the extended route, and to determine the first lateral point and the second lateral point with distribution characteristics located on both sides of the sub-route based on the tunnel crack; The image acquisition module is configured to control the lateral illumination unit carried by the high-precision acquisition unit to move to coincide with the lateral point, and respond to the lateral illumination unit to provide illumination, thereby controlling the high-precision acquisition unit to acquire images of the corresponding sub-route. The data integration module is configured to determine the route endpoints far from the starting point as new starting points based on the sub-route, repeat the above steps until the termination point is located in the image acquired for any number of acquisitions, and obtain the detection data of the corresponding tunnel crack.

Images