A method and system for detecting apparent communication signal cable detachment in a two-stage tunnel.

Abstract

This invention relates to a method and system for detecting the detachment of communication signal cables in a two-stage tunnel. The detection method includes: S1, dynamically acquiring original images of tunnel surface ancillary facilities and equipment using an image acquisition module; S2, preprocessing the original images acquired by the image acquisition module; S3, classifying the tunnel surface ancillary facilities and equipment based on the preprocessed image data; and S4, detecting the slope of the communication signal cables to achieve detachment detection. The detection system includes: an image acquisition module for dynamically acquiring original images of tunnel surface ancillary facilities and equipment; and a processing module for preprocessing the original images acquired by the image acquisition module, classifying the tunnel surface ancillary facilities and equipment based on the preprocessed image data, and detecting the slope of the communication signal cables to achieve detachment detection.

Description

Technical Field

[0001] This invention relates to the field of urban rail transit communication technology, and in particular to a method and system for detecting the detachment of apparent communication signal cables in a two-stage tunnel. Background Technology

[0002] As the "central nervous system" of urban rail transit, the communication and signaling system plays an irreplaceable role in ensuring train safety, improving transportation efficiency, and optimizing train operation. With the continuous increase in the operating speed of urban rail transit (urban rail) and the increasing service life of infrastructure equipment, equipment failures and aging facilities are becoming increasingly apparent. This leads to frequent operational failures and rising maintenance costs, posing a serious challenge to the safe and stable operation of urban rail transit. Especially for communication and signaling cables that are exposed to complex environments for extended periods, their health directly affects the accurate transmission of train control information. Any damage or detachment can cause serious safety hazards, even leading to accidents involving encroachment on safety barriers, threatening the lives and property of passengers. Therefore, regularly inspecting, monitoring, and maintaining these critical facilities has become an important task in ensuring the stable operation of the urban rail system.

[0003] CN104410820A discloses a vehicle-mounted trackside equipment box and cable appearance inspection system. It includes an image acquisition device installed diagonally above and outside the track on the vehicle body, used to acquire image data of the trackside equipment box and cables in real time according to control signals from a controller; a controller used to send control signals to a plurality of image acquisition devices according to image acquisition parameters input from a storage and display server; and a storage and display server used to acquire environmental information from an environmental information acquisition device and obtain image acquisition parameters matching the environmental information.

[0004] CN112200483A discloses an automatic inspection system and method for subway trackside equipment. The inspection vehicle is equipped with an image acquisition module, a positioning module, and a fault detection module. The image acquisition module acquires image data of the subway trackside equipment, the positioning module obtains the real-time location information of the inspection vehicle, and the fault detection module identifies and analyzes the subway trackside equipment based on the image data and the location information corresponding to the image capture, obtains the location of the faulty equipment, and clarifies the fault type.

[0005] CN111208146A discloses a tunnel cable detection system and method. The system includes: an image acquisition device installed outside the inspection vehicle's cargo compartment; speed sensors installed on the wheels of the inspection vehicle; and an image processing device communicatively connected to the image acquisition device. The image acquisition device acquires image information of the cable to be inspected and sends the image information to the image processing device. The image processing device includes an image data acquisition control module and an image data processing module. The image data acquisition control module includes a synchronous acquisition control submodule, used to trigger the image acquisition device to acquire image information based on the speed information of the inspection vehicle's wheels acquired by the speed sensors. The image data processing module processes the image information. This method is a way to detect tunnel cables using the above system.

[0006] Existing methods for detecting damage and detachment of communication signal cables mainly rely on manual inspection or detection systems mounted on low-speed electric drive integrated testing platforms. The former, requiring personnel to personally enter the track area for visual inspection, has significant limitations: firstly, manual inspection is extremely inefficient, especially on long-distance lines, as completing a comprehensive inspection is time-consuming and labor-intensive; secondly, this method is constrained by the technical skill and personal experience of the inspectors, leading to significant discrepancies in results between different personnel and the potential for overlooking hidden dangers. Furthermore, manual inspection carries high safety risks, particularly at night or in inclement weather conditions, where inspectors face greater physical and psychological stress. While the latter utilizes mechanized methods, it still has many shortcomings. For example, the speed of testing using electric drive integrated testing platforms is typically limited to around 15 km / h, far below the actual train operating speed, forcing such inspections to be scheduled during non-operational hours, severely limiting the frequency and timeliness of testing. Simultaneously, the detection accuracy and reliability of such systems are also limited, especially in complex and variable tunnel environments, where factors such as lighting conditions and background interference can affect the accuracy of the final results. In conclusion, both traditional and currently advanced detection technologies exhibit varying degrees of lag and limitations in meeting the demands of rapidly developing urban rail transit, necessitating the exploration of more efficient, accurate, and safe next-generation detection solutions.

[0007] Furthermore, on the one hand, there are differences in understanding among those skilled in the art; on the other hand, the applicant studied a large number of documents and patents when making this invention, but due to space limitations, not all details and contents were listed in detail. However, this does not mean that the present invention does not possess the features of these prior art. On the contrary, the present invention already possesses all the features of the prior art, and the applicant reserves the right to add relevant prior art to the background art. Summary of the Invention

[0008] In view of the shortcomings of the prior art, the present invention provides a method and system for detecting the detachment of apparent communication signal cables in a two-stage tunnel, so as to solve at least some of the above-mentioned technical problems.

[0009] In this invention, "two-stage" refers to the two main steps included in the detection method. First, the images are classified and identified. Then, the slope of the communication signal cable is specifically detected to determine whether a detachment has occurred. Further, "two-stage" specifically refers to a systematic, step-by-step detection process. The first stage involves preprocessing and classifying the acquired images of equipment inside the tunnel to ensure accurate identification of various types of auxiliary facilities and equipment. The second stage focuses on the status analysis of the communication signal cable, particularly assessing the risk of detachment by calculating the cable's slope. These two stages together constitute a complete automated detection scheme aimed at improving the safety and reliability of urban rail transit communication systems.

[0010] In this invention, "tunnel appearance" refers to the visible surfaces inside the tunnel and the various devices and structures installed on them. More specifically, "tunnel appearance" refers to the parts of the tunnel interior that can be directly observed, including but not limited to tunnel walls, ceilings, and all ancillary facilities and equipment fixed thereto, such as junction boxes, loudspeakers, communication signal cables, cable trays, etc. These elements constitute part of the tunnel's visual environment, and their condition is crucial to ensuring the safe operation of trains. Therefore, when detecting detached communication signal cables, the entire tunnel appearance must be considered comprehensively to obtain the most accurate results.

[0011] In this invention, "communication signal cables" are critical lines used for transmitting data or commands, playing a vital role in urban rail transit systems. Further, "communication signal cables" refer to the physical medium connecting different communication nodes, responsible for transmitting information (such as control commands, monitoring data, etc.). When the line is in good health, these cables ensure efficient communication between the train control system and ground stations, maintaining normal train operation. However, damage or detachment can lead to serious safety hazards and even accidents. Therefore, regularly inspecting and maintaining the condition of communication signal cables to prevent them from failing due to aging or other reasons has become an indispensable part of ensuring the safe operation of urban rail transit.

[0012] In this invention, "detachment detection" is a mechanism designed to prevent potential hazards by detecting whether communication signal cables have detached from their original position. Furthermore, "detachment detection" is a specific technical means to identify whether communication signal cables have been displaced or fallen from a predetermined location. This work is crucial because any accidental movement of the cable can interfere with the normal communication link, thereby affecting the safe operation of the train. Traditional detection methods often rely on manual inspections, but this method is inefficient and prone to missing potential hazards. In contrast, the detection method and system provided by this invention can more quickly and accurately locate problem points and issue timely alarms, thereby greatly reducing the risk of accidents.

[0013] This invention discloses a method for detecting the detachment of apparent communication signal cables in a two-stage tunnel, comprising the following steps:

[0014] S1. Dynamically acquire original images of tunnel surface ancillary facilities and equipment using the image acquisition module;

[0015] S2. Preprocess the raw images acquired by the image acquisition module;

[0016] S3. Based on the preprocessed image data, classify the tunnel's apparent ancillary facilities and equipment;

[0017] S4. Detect the slope of the communication signal cable to detect the detachment of the communication signal cable.

[0018] This detection method first dynamically acquires raw images of the tunnel's external facilities and equipment. This not only captures real-time information on changes in the tunnel's internal environment but also avoids potential omissions due to long time intervals in traditional manual inspections. Next, preprocessing the acquired raw images significantly improves image quality and feature identifiability, laying a solid foundation for subsequent classification and detection. Finally, based on the classification of the preprocessed image data, the slope of communication signal cables is detected, enabling effective identification of detached communication signal cables. This process not only improves detection accuracy but also greatly reduces the possibility of false alarms, ensuring the safe operation of the urban rail transit system. Automating the entire detection process reduces labor costs and improves efficiency, making detection work more efficient and timely. Furthermore, this method can adapt to different types of tunnel environments and can complete detection during normal train operation, further enhancing the safety and reliability of the detection.

[0019] According to a preferred embodiment, the image acquisition module is mounted on a carrier vehicle, so that when the carrier vehicle moves on the track, the image acquisition module can respond to the received pulse signal to realize acquisition triggering based on preset second basic information including trigger mode and trigger parameters, wherein the trigger mode includes time triggering and distance triggering.

[0020] By mounting the image acquisition module on the carrier vehicle and triggering the acquisition based on preset trigger modes (such as time-triggered or distance-triggered) and responding to pulse signals, this design cleverly combines the characteristics of train operation with the inspection requirements. In practical applications, as the carrier vehicle moves along the track, the image acquisition module installed on it can automatically initiate the image acquisition process according to the set trigger parameters. For the distance-triggered mode, precise calculations of the encoder frequency and wheel circumference ensure that each acquisition occurs at the accurate location, which not only improves the spatial consistency of image data but also provides a reliable data foundation for subsequent analysis. Furthermore, by adjusting the trigger mode and parameter settings, the acquisition frequency can be flexibly controlled according to the specific conditions of different track sections, ensuring comprehensive inspection while avoiding unnecessary repetitive operations, thereby effectively improving inspection efficiency. This design not only meets the requirements of efficient operation in modern rail transit but also provides technical support for achieving intelligent and automated inspection, especially by maintaining high-precision image acquisition even under high-speed operating conditions, greatly enhancing the applicability and robustness of the inspection system.

[0021] According to a preferred embodiment, the preprocessing of the original image acquired by the image acquisition module includes image geometric correction using affine transformation, noise removal using Gaussian filtering, and image contrast enhancement using the Retinex algorithm, so as to improve image quality and feature recognizability.

[0022] In the preprocessing of the raw images acquired by the image acquisition module, affine transformation is used for geometric correction, Gaussian filtering for noise removal, and the Retinex algorithm for contrast enhancement. These steps address issues such as geometric distortion, random noise interference, and uneven illumination in the images. First, affine transformation eliminates distortions caused by shooting angle or positional deviations by adjusting parameters such as image angle and scaling, ensuring all images have a consistent standard view. Next, Gaussian filtering effectively smooths high-frequency noise in the image while preserving edge details, resulting in a clearer and cleaner image. Finally, the Retinex algorithm enhances the contrast of local areas by separating the reflection and illumination components in the image, revealing previously obscured details. After this series of preprocessing operations, the image quality is significantly improved, and features are more prominent. This not only improves the accuracy of subsequent classification and recognition but also provides high-quality data support for the final detection of detached communication signal cables. Furthermore, these preprocessing steps reduce the computational burden in subsequent processing, improving processing speed and response time.

[0023] According to a preferred embodiment, the YOLO v5m algorithm is used to classify the tunnel appearance ancillary facilities and equipment for the preprocessed image data, so as to identify different types of equipment and store the tunnel appearance ancillary facilities and equipment images in the corresponding database according to the types of equipment contained therein. YOLO v5m is used as the base model, and an SE module containing compression and excitation processes is introduced into the backbone layer of YOLO v5m.

[0024] Based on preprocessed image data, the YOLO v5m algorithm is used to classify the apparent ancillary facilities and equipment in tunnels, and the images of different types of equipment are stored in corresponding databases. This approach fully leverages the advantages of deep learning technology. YOLO v5m, as an advanced object detection algorithm, is fast and accurate, capable of processing large numbers of images and providing reliable classification results in a short time. Especially after introducing the SE module in its backbone layer, the model can adaptively recalibrate the channel feature response, further enhancing useful features and suppressing irrelevant information. This means that when facing complex and changing tunnel environments, the YOLO v5m+SE model can more accurately identify various types of communication signal cables and their related equipment, reducing the possibility of misclassification. Furthermore, storing the identification results by category not only facilitates subsequent data management and retrieval but also provides convenience for long-term monitoring and trend analysis. In this way, the system not only improves the efficiency and accuracy of detection but also provides maintenance personnel with detailed reference materials, helping to formulate scientific and reasonable maintenance plans and ensure the safe and reliable operation of the urban rail transit system. This classification method can also continuously improve its performance and accuracy through machine learning as new data accumulates.

[0025] According to a preferred embodiment, images of various communication signal cables in the communication signal cable library are used as input images. By performing edge detection on the input images and applying Hough transform, votes are accumulated in the parameter space to identify straight segments of the communication signal cables. Then, the specific position of the cable is reconstructed based on the local maximum value in the accumulator array, thereby determining the slope of the communication signal cable in the input image.

[0026] For images of various communication signal cables in a communication signal cable library, edge detection and Hough transform are applied to accumulate votes in the parameter space to identify straight cable segments. The specific location of the cable is reconstructed based on local maxima in the accumulator array, thus determining the slope of the communication signal cable in the input image. This method cleverly utilizes mathematical transformation principles to simplify complex image processing problems. First, edge detection extracts pixels potentially belonging to straight lines from the image, providing the foundational data for the subsequent Hough transform. Then, the Hough transform achieves efficient searching of straight line features by mapping each edge point to ρ (distance from the line to the origin) and θ (angle between the line and the positive x-axis) in the parameter space. The accumulator array plays a crucial role in this process: it records the number of support votes for each possible straight line, highlighting those line segments that truly exist in the image. Finally, by analyzing the local maxima in the accumulator array, the location and slope of the communication signal cable can be accurately determined. This method is not only unaffected by line rotation and scaling but also effectively handles noise and occlusion in the image, greatly improving the robustness and accuracy of detection. This precise slope detection can promptly identify abnormal conditions in communication signal cables, providing early warnings of potential safety hazards and offering strong technical support for ensuring the safety of urban rail transit.

[0027] According to a preferred embodiment, a cable slope threshold is set to detect the detachment of communication signal cables by combining the cable slope threshold with the characteristics of abnormal cable trays. If the slope of any communication signal cable exceeds the cable slope threshold, it is judged as a suspected detachment. Then, it is determined whether the communication signal cable has been detached by judging whether the cable tray associated with the suspected detached communication signal cable has experienced abnormalities including positional displacement, posture change and / or structural loss.

[0028] The above solution provides an intelligent judgment mechanism based on multi-feature fusion. First, by using a pre-set cable slope threshold, the system can quickly filter out suspected detached cables whose absolute slope value exceeds the threshold (e.g., 0.3). This threshold is an empirical value derived from extensive data analysis and has high reliability. Then, for these suspected detached cables, the system further checks for anomalies such as positional shifts, posture changes, or structural defects in the associated cable trays. This two-step approach considers not only the physical characteristics of the cable itself but also the state of its fixing devices, forming a more comprehensive evaluation system. Specifically, if the cable slope exceeds the threshold and the corresponding tray shows an anomaly, it can be determined with relatively high confidence that the cable has detached. Conversely, if only one condition is met, further investigation and confirmation are required. This method is not only simple and easy to implement, requiring no additional complex algorithms, but also has high accuracy. It can significantly improve the sensitivity and reliability of communication signal cable detachment detection without affecting the overall system performance. Through this dual verification mechanism, not only can the accuracy of detection be improved, but false alarms can also be reduced, providing maintenance personnel with more reliable decision-making basis and ensuring the safe operation of urban rail transit.

[0029] The present invention also discloses a two-stage tunnel surface communication signal cable detachment detection system, which includes: an image acquisition module for dynamically acquiring original images of tunnel surface auxiliary facilities and equipment; and a processing module for preprocessing the original images acquired by the image acquisition module, and then classifying the tunnel surface auxiliary facilities and equipment based on the preprocessed image data, and detecting the slope of the communication signal cable to realize the detachment detection of the communication signal cable.

[0030] This invention constructs a two-stage tunnel surface communication signal cable detachment detection system. This system integrates functional units such as an image acquisition module and a processing module. It aims to effectively monitor the detachment of communication signal cables by dynamically acquiring raw images of tunnel surface ancillary facilities and equipment, performing preprocessing, classification, and slope detection, among other operations. Compared to traditional manual inspection methods, this system not only significantly improves the automation level of the inspection work and reduces human interference, but also ensures the consistency and accuracy of data acquisition. More importantly, through integrated hardware design and software algorithm optimization, the system can complete the processing and analysis of a large number of images in a short time, greatly improving detection efficiency.

[0031] According to a preferred embodiment, it is configured with an integrated device external to a carrier vehicle and a mounting bracket for connecting the integrated device to the carrier vehicle. The image acquisition module is installed inside the integrated device. The mounting bracket includes two non-coplanar mounting surfaces, a first mounting surface for connecting to the carrier vehicle and a second mounting surface for connecting to the integrated device.

[0032] This testing system features an integrated device externally mounted on a carrier vehicle, secured using a mounting bracket with two non-coplanar mounting surfaces. This design simplifies installation and enhances system stability and applicability. The compact structure of the integrated device facilitates installation and disassembly while ensuring tight fit between components, minimizing the impact of external environmental factors on performance. The first mounting surface of the bracket connects to the carrier vehicle, while the second connects to the integrated device. This non-coplanar design cleverly solves spatial layout issues, ensuring the device's versatility and compatibility across various vehicle models. Furthermore, the robust connection of the mounting bracket guarantees stability during high-speed operation, preventing loosening or detachment due to vibration or impact. This design not only facilitates operation by on-site personnel but also extends the device's lifespan and reduces maintenance costs.

[0033] According to a preferred embodiment, the integrated device is configured with an outer shell including an internal cavity, and the bottom plate of the outer shell is provided with a mounting groove for connecting with a second mounting surface of a mounting bracket, so that the integrated device can be detachably fixed to the mounting bracket.

[0034] A mounting slot is provided on the bottom plate of the outer casing for connection with the second mounting surface of the mounting bracket, allowing the integrated device to be detachably fixed to the mounting bracket. This design embodies the advantages of modular design. First, the mounting slot makes the connection between the outer casing and the mounting bracket simpler and faster, greatly facilitating on-site commissioning and maintenance. Second, the detachable fixing method not only allows users to flexibly adjust the position and angle of the device according to actual needs, but also facilitates the inspection or upgrading of internal components without worrying about damaging the overall structure. Furthermore, this modular design helps improve the system's scalability and flexibility; for example, new functional modules can be added or existing components replaced in the future to adapt to changing application scenarios and technological advancements.

[0035] According to a preferred embodiment, the housing of the integrated device includes two opposing side plates and a working plate disposed between the two side plates. The working plate has a plurality of through holes spaced apart, so that the linear array camera and laser filler of the image acquisition module can be installed in a set on these through holes. The through holes on the working plate can be positioned based on the field of view angle of the linear array camera, so that imaging of the tunnel lining surface other than the track area can be achieved while ensuring that the field of view ranges of adjacent linear array cameras at least partially overlap.

[0036] This design fully considers the practical needs and optical characteristics of image acquisition. First, the positions of the through-holes are carefully arranged based on the field of view of the line scan cameras, ensuring at least partial overlap between the fields of view of adjacent line scan cameras, thus achieving seamless coverage of the tunnel lining surface except for the track area. This design not only improves the integrity of image acquisition but also avoids information loss due to blind spots. Second, the integrated installation of the line scan cameras and laser illuminators ensures their coordinated operation, providing sufficient illumination during image capture and ensuring image quality is not limited by lighting conditions. Furthermore, by rationally planning the position and spacing of the through-holes, the overall layout of the equipment can be optimized, reducing volume and improving space utilization. Finally, this design allows the equipment to adapt to different types of tunnel environments, whether it's tunnel cross-sections of varying widths or complex internal structures, enabling the acquisition of high-quality image data and providing a solid foundation for subsequent analysis and processing. Attached Figure Description

[0037] Figure 1 This is a flowchart of the detection method provided by the present invention;

[0038] Figure 2 This is a schematic diagram illustrating the setting of the second basic information of the image acquisition module according to a preferred embodiment of the present invention;

[0039] Figure 3 This is an image schematic diagram of different types of tunnel appearance ancillary facilities and equipment provided by the present invention being collected;

[0040] Figure 4 This is a schematic diagram of the process for detecting the detachment of communication signal cables provided by the present invention;

[0041] Figure 5 This is a hardware connection diagram of the detection system provided by the present invention;

[0042] Figure 6 This is a schematic diagram of the mounting bracket of the detection system provided by the present invention;

[0043] Figure 7 This is a schematic diagram of the integrated device of the detection system provided by the present invention;

[0044] Figure 8 This is a schematic diagram of the imaging range of the image acquisition module provided by the present invention.

[0045] List of reference numerals

[0046] 100: Carrier vehicle; 200: Mounting bracket; 211: First mounting surface; 212: Second mounting surface; 300: Integrated equipment; 310: Outer shell; 320: Base plate; 321: Mounting groove; 330: Side plate; 340: Working plate; 341: Through hole; 342: Image acquisition module; 343: Line scan camera; 344: Field of view; 345: Laser filler; 350: Handle; 400: Processing module; 500: Tunnel lining; 600: Auxiliary facilities and equipment; 610: Distribution box; 620: Broadcast loudspeaker; 630: Communication signal cable; 640: Cable tray; 650: Instrument; 660: Signal device; 670: Signal device box; 680: Socket box; 690: CCTV equipment system. Detailed Implementation

[0047] The following is a detailed explanation with reference to the accompanying drawings.

[0048] Example 1

[0049] like Figure 1 As shown, this invention discloses a method for detecting the detachment of a two-stage tunnel apparent communication signal cable 630, which may include the following steps:

[0050] S1. Use the image acquisition module 342 to dynamically acquire original images of the tunnel's surface auxiliary facilities and equipment 600;

[0051] S2. Preprocess the raw image acquired by the image acquisition module 342;

[0052] S3. Based on the preprocessed image data, classify the tunnel's apparent ancillary facilities and equipment 600;

[0053] S4. Detect the slope of the communication signal cable 630 to detect the detachment of the communication signal cable 630.

[0054] Preferably, in step S1, the image acquisition module 342 installed on the train (e.g., an electric multiple unit) can be used to achieve digital imaging of the tunnel lining surface 500 during vehicle operation. Even when the train is running at high speeds (e.g., 80 km / h or 120 km / h), the image acquisition module 342 can still acquire images of the tunnel's surface auxiliary facilities and equipment 600. This overcomes the problems of low efficiency, high cost, and poor inspection safety associated with manual inspections or using an electric integrated inspection platform with an inspection system at a speed of 15 km / h. To ensure the quality and reliability of the image data, the image acquisition module 342 can select a suitable camera. Specifically, the image acquisition module 342 can use a high-resolution, low-noise industrial-grade camera with a wide dynamic range, and can be customized according to the characteristics of the tunnel environment (e.g., lighting conditions, humidity, temperature). These cameras can adapt to the relatively dim environment inside the tunnel and maintain image stability and clarity during high-speed movement. Furthermore, considering the camera's field of view coverage and installation stability, the image acquisition module 342 can be installed at the front or rear of the train to ensure its field of view covers the entire tunnel's apparent ancillary facilities and equipment 600. In addition, protective measures for the camera, such as waterproofing, dustproofing, and shock-resistant design, need to be considered to cope with the complex environment inside the tunnel.

[0055] Preferably, due to the poor lighting conditions inside the tunnel, the supplementary lighting unit of the image acquisition module 342 is particularly important. The supplementary lighting unit may include LED strips or other types of light sources, which can provide uniform and sufficient illumination, enabling the camera to acquire high-quality images even in low-light conditions. The brightness and angle of the supplementary lighting unit need to be adjusted according to the actual situation to avoid problems such as overexposure or shadows affecting image quality.

[0056] Preferably, after the image acquisition module 342 is installed, the first basic information such as the camera's exposure and gain parameters, and the fill light parameters of the fill light source can be adjusted. Second basic information such as the trigger mode and trigger parameters can also be set, so that when the train (hereinafter referred to as the carrier vehicle 100) equipped with the image acquisition module 342 moves on the track, the image acquisition module 342 can receive pulse signals to trigger the acquisition. Further, the trigger mode determines the rules for initiating the image acquisition operation. Different trigger modes can be associated with corresponding trigger parameters. Common trigger modes include time triggering and distance triggering. Other triggering methods may include event triggering. The trigger parameters are the trigger condition data set for the selected trigger mode. When the trigger parameters corresponding to the trigger mode reach the set trigger condition data, the rules for initiating the image acquisition operation are met, and the image acquisition module 342 can be driven to perform image acquisition.

[0057] For example, such as Figure 2 As shown, for the time-triggered trigger mode, a preset trigger time (e.g., 300ms) can be input so that the image acquisition module 342 can be driven to acquire images after each preset trigger time interval; for the distance-triggered trigger mode, in addition to inputting a preset trigger distance (e.g., 750mm), the encoder frequency (i.e., the number of pulses output per revolution) and the wheel circumference can also be input to calculate the distance traveled by the carrier vehicle 100, thereby accurately determining whether the carrier vehicle 100 has reached the preset trigger distance.

[0058] Preferably, based on the pre-set first and second basic information, the image acquisition module 342 can periodically acquire original images of the tunnel's apparent auxiliary facilities and equipment 600 as the carrier vehicle 100 moves. The auxiliary facilities and equipment 600 may include a junction box 610, a loudspeaker 620, a communication signal cable 630, a cable tray 640, an instrument 650, a signal controller 660, a signal controller box 670, a socket box 680, a CCTV equipment system 690, etc.

[0059] Preferably, in step S2, the raw images acquired by the image acquisition module 342 are often affected by various factors, such as uneven lighting, motion blur, and noise, which can reduce the accuracy of subsequent analysis. Therefore, a series of preprocessing operations must be performed on these image data to improve image quality and feature recognizability. Preferably, the preprocessing process may include image correction, noise removal, and image contrast enhancement to improve image quality and feature recognizability.

[0060] In a tunnel environment, due to changes in the speed of the vehicle 100 and the varying curve radii of the track, the acquired original images may exhibit problems such as tilting, stretching, or compression. Furthermore, the angle setting of the image acquisition module 342 may also cause image distortion. To ensure that all images accurately reflect the actual scene and provide a consistent basis for subsequent feature extraction, geometric correction of the original images is necessary first. Preferably, affine transformation can be used for image geometric correction in step S2. Affine transformation is a linear mapping method that can be represented as matrix multiplication and is suitable for describing geometric transformations such as translation, rotation, scaling, and shearing. By calculating the correspondence between specific points, appropriate transformation parameters can be determined, thereby converting the original image into a standard view that meets expectations. Preferably, algorithms such as SIFT (Scale-Invariant Feature Transform) or ORB (Oriented Fast and Rotated BRIEF) can be used to find representative feature points in the original image and match them with corresponding points in a standard template. Based on the matching results, the optimal affine transformation matrix is ​​solved using the least squares method or other optimization algorithms. This matrix will be used to adjust the position, orientation, and size of the entire image to make it as close as possible to the ideal state. Finally, the calculated transformation matrix is ​​applied to the original image to complete the geometric correction. Simultaneously, the quality of the corrected image needs to be evaluated to check for any distortion, and the parameters are adjusted as needed until satisfactory results are achieved.

[0061] Despite the use of high-quality cameras and lighting, images captured during high-speed movement inevitably contain some noise components, such as randomly distributed bright spots or dark spots. This noise not only affects visual quality but also interferes with subsequent feature detection and classification. Therefore, it is necessary to take effective noise reduction measures to clean up the image data. Preferably, Gaussian filtering can be used for noise removal in step S2. Gaussian filtering is a spatial filter that smooths the image by weighted averaging of neighboring pixels, where the weights are determined by a two-dimensional Gaussian function. This method can effectively reduce high-frequency noise while preserving edge information, making it particularly suitable for processing images in natural scenes. The performance of the filter depends on the selection of two main parameters: its core size and standard deviation. A larger core size can eliminate noise more thoroughly but may also lead to image blurring; a smaller core size may not be able to filter noise sufficiently. Therefore, these two parameters need to be flexibly adjusted according to the specific application scenario to achieve the best balance. Considering that different types of noise are distributed at different spatial scales, a multi-scale Gaussian filtering strategy can be adopted, that is, first use a filter with a larger core size to process low-frequency noise, and then gradually reduce the core size to remove high-frequency noise, ultimately obtaining a clean and clear image. To avoid artifacts at image edges, methods such as mirror reflection, periodic expansion, or constant padding can be selected to ensure the consistency and integrity of the filtering operation.

[0062] The lighting conditions inside tunnels are complex and variable, with some areas exhibiting strong light and shadow contrasts while others appear dim and unclear. This uneven lighting distribution makes images difficult to use directly for analysis, especially when target objects are of similar color. Therefore, image contrast enhancement is necessary to highlight details and facilitate subsequent feature recognition. Preferably, in step S2, the Retinex algorithm can be used for image contrast enhancement. The Retinex algorithm separates the reflection component (i.e., the true color of the object) and the illumination component in the image, and then recombines them to improve the overall contrast. The Retinex algorithm can include single-scale Retinex (SSR) and multi-scale Retinex (MSR). Considering the diversity of lighting conditions inside tunnels, an adaptive parameter adjustment strategy can be adopted. For example, the weights of each scale can be dynamically adjusted according to the brightness level of different areas, or the appropriate algorithm version can be automatically selected based on the image content to ensure the best enhancement effect. After completing the Retinex processing, histogram equalization, gamma correction, and other techniques can be further applied to fine-tune the overall tone of the image, maintaining its original style characteristics while achieving good visual effects.

[0063] Preferably, in step S3, the YOLO v5m algorithm can be used to classify the tunnel's apparent ancillary facilities and equipment 600, to identify devices such as junction boxes 610, loudspeakers 620, communication signal cables 630, cable trays 640, instruments 650, signal controllers 660, signal controller boxes 670, socket boxes 680, and CCTV equipment systems 690. Figure 3 As shown. Furthermore, during the training phase, the collected images of various tunnel appearance ancillary facilities and equipment 600 can be labeled accordingly to serve as the training set for the model; during the testing phase, the model will automatically assign the corresponding label after recognizing the equipment.

[0064] Preferably, based on a comprehensive consideration of detection accuracy, model size, and detection speed, this invention selects YOLO v5m as the base model and introduces an SE module into the backbone layer of YOLO v5m. The SE module includes two processes: compression and excitation.

[0065] Preferably, the compression process uses global average pooling to compress the spatial dimension of the feature map from H×W×C to 1×1×C, so that the features of each channel are aggregated into a global statistic. In this way, the statistic of each channel contains global spatial information, equivalent to obtaining the receptive field of the entire image. This embedding of global information helps the model capture global contextual information in the image, thereby improving the accuracy of image recognition. The specific formula is as follows:

[0066]

[0067] Among them, u C (i,j) represents the element in the i-th row and j-th column of the C-th channel in the input feature map U, where H and W are the height and width of the feature map, respectively, and C represents the number of channels; after compression, the feature map is compressed into a 1×1×C vector, with each channel corresponding to a global average value; z C This represents the compressed feature vector.

[0068] Preferably, the activation process uses two fully connected layers to predict and adaptively recalibrate the importance of each channel. Further, the first fully connected layer reduces the dimensionality of the 1×1×C feature vector to 1×1×C×Ratio, and the second fully connected layer increases it back to 1×1×C, where Ratio is a scaling parameter used to reduce the number of channels and decrease computational cost. Through this activation process, the model learns the dependencies between different channels and assigns a weight to each channel, representing the channel's importance to the final task.

[0069] Preferably, the activation process may include two fully connected layers and a nonlinear activation function to learn and generate a weight vector for one channel. The specific formula is as follows:

[0070] y C =σ(W2·δ(W1·z) C )),

[0071] Among them, z C This is the compressed feature vector, where W1 and W2 are the weights of the two fully connected layers, δ is the ReLU activation function, σ is the Sigmoid activation function, and y... C It is the output of the excitation operation, i.e., the channel weight vector.

[0072] Channel weight vectors can be used to weight each channel of the original feature map, recalibrating the original features along the channel dimension. This recalibration helps the model focus on more informative features and suppress less important features, thereby improving image recognition performance. The specific formula is as follows:

[0073]

[0074] in, This represents the weighted feature map, y C This represents the channel weight vector for the C-th channel. The new feature map is then fed into the next layer of the network for subsequent feature extraction and image recognition tasks.

[0075] By following the above steps, the detection accuracy can be improved, enabling the identification of tunnel surface ancillary facilities and equipment 600, such as junction box 610, loudspeaker 620, communication signal cable 630, cable tray 640, instrument 650, signal machine 660, signal machine box 670, socket box 680, and CCTV equipment system 690.

[0076] For example, suppose the feature map X has a size of 2×2×3, then it is a three-dimensional tensor, where the first dimension is the height (i.e., H=2), the second dimension is the width (i.e., W=2), and the third dimension is the number of channels (i.e., C=3), as shown below:

[0077]

[0078] Here, X consists of three 2×2 matrices, each representing a feature of a channel.

[0079] After compression by the SE module, the feature map of each channel is globally averaged to obtain a 1×1×3 feature vector z:

[0080]

[0081] Next, the importance weight of each channel is learned through the activation process. Assume that the channel weight vector y obtained after the fully connected layer and activation function is as follows:

[0082]

[0083] Now, multiply the channel weight vector obtained from the activation operation channel by each channel of the original feature map to obtain the weighted feature map X':

[0084]

[0085] This reveals that for the weighted feature map X', the value of each channel is adjusted according to its importance. This new feature map will be used in subsequent layers of the network to help the model better identify features in the image.

[0086] Preferably, the images of various types of tunnel appearance auxiliary facilities and equipment 600 identified in step S3 can be stored in corresponding databases. For example, the images of communication signal cables 630 can be stored in a communication signal cable library so that in step S4, the Hough straight line detection method can be used to detect each communication signal cable 630 in the communication signal cable library.

[0087] Preferably, in step S4, edge detection can be performed on the input image to determine pixels that may belong to a straight line, and a parameter space can be defined. Edge detection can be performed using algorithms such as Canny and Sobel. For example, the Canny algorithm can achieve good signal-to-noise ratio and sub-pixel accuracy. Preferably, in the Hough transform, a straight line can be defined by two parameters: ρ (the distance from the origin to the line) and θ (the angle between the line and the positive x-axis). Therefore, for each edge point in the image, all possible combinations of ρ and θ can be calculated, and these combinations define the position of the line in the parameter space. Further, an accumulator array is created. The accumulator array is a two-dimensional array whose horizontal coordinate corresponds to θ and its vertical coordinate corresponds to ρ. Each cell corresponds to a point (ρ, θ) in the parameter space, used to store the count of each possible straight line. If multiple edge points correspond to the same combination of ρ and θ, the count of this combination will increase, indicating that multiple edges may belong to the same straight line. By finding local maxima in the accumulator array, the corresponding ρ and θ combinations represent the detected straight lines. Then, based on the location of these local maxima, the corresponding ρ and θ values ​​are returned, which can be used to reconstruct the straight lines in the original image, i.e., the specific location of the cable. For example, the conversion from parameter space back to image space can be achieved using the transformation relationship between polar coordinates and Cartesian coordinates.

[0088] When using the above method to detect straight lines in an image, it is unaffected by line rotation or scaling; that is, regardless of how the line is tilted or its length changes within the image, including lines that are nearly horizontal, vertical, or diagonal, it can be accurately detected. The Hough line detection method handles image noise well, even in situations with noise points after edge detection, low image contrast, or even partially occluded or broken images, it can still detect complete straight lines. Furthermore, the Hough transform allows for control of detection sensitivity by adjusting the accumulator threshold, thus striking a balance between the number of lines detected and accuracy.

[0089] Preferably, in step S4, a cable slope threshold can be set to detect the detachment of the communication signal cable 630 by combining the cable slope threshold and the abnormality of the cable tray 640. Figure 4 As shown. Preferably, in this invention, the cable slope threshold can be set to 0.3. This is because, through analysis of a large amount of image data, it was found that when the communication signal cable 630 is detached, its absolute slope value is always above 0.3. Therefore, if the absolute value of the cable slope is greater than 0.3, it is judged as a suspected detachment. Preferably, the cable bracket 640 is usually used to fix the communication signal cable 630, and these cable brackets 640 should maintain a certain position and posture, that is, the position of the cable bracket 640 relative to the tunnel wall or track is fixed. Further, for the cable bracket 640 associated with the suspected detached communication signal cable 630, the image of the corresponding cable bracket 640 is extracted from the cable bracket library generated in step S3, so as to determine whether the cable bracket 640 associated with the suspected detached communication signal cable 630 has experienced abnormal conditions such as positional displacement, posture change and / or structural loss by comparing it with the standard image of the cable bracket 640. If the above abnormal conditions exist, it can be determined that the communication signal cable 630 is detached. This method is simple and easy to implement, requires no additional algorithms, and has high accuracy.

[0090] Example 2

[0091] This embodiment is a further improvement on embodiment 1, and repeated content will not be described again.

[0092] This invention discloses a two-stage tunnel apparent communication signal cable 630 detachment detection system. The detection system can perform the detection method as described in Example 1 to achieve intelligent identification of tunnel apparent communication signal cable 630 detachment defects.

[0093] Preferably, such as Figure 5As shown, the detection system of the present invention may include: an image acquisition module 342, used to dynamically acquire original images of the tunnel surface ancillary facilities and equipment 600; and a processing module 400, used to preprocess the original images acquired by the image acquisition module 342, and then classify the tunnel surface ancillary facilities and equipment 600 based on the preprocessed image data, and detect the slope of the communication signal cable 630, thereby realizing the detection of the detachment of the communication signal cable 630.

[0094] Preferably, such as Figure 6 and Figure 7 As shown, the detection system of the present invention can be configured with an integrated device 300 externally mounted on a carrier vehicle 100 and a mounting bracket 200 for connecting the integrated device 300 to the carrier vehicle 100. The integrated device 300 can be connected to the front or rear of the carrier vehicle 100 via the mounting bracket 200. The mounting bracket 200 can be configured with a Z-shaped structure to form two parallel but non-coplanar mounting surfaces. Further, the first mounting surface 211, which has a relatively larger area, is used to connect with the body of the carrier vehicle 100, and the second mounting surface 212, which has a relatively smaller area, is used to connect with the integrated device 300.

[0095] Preferably, the image acquisition module 342 can be housed within the integrated device 300. The integrated device 300 may also include a synchronization controller for generating control signals and an industrial control computer for data processing. Preferably, the synchronization controller can be used to adjust the first basic information and / or the second basic information, and can generate a trigger signal when a trigger condition is met, enabling the image acquisition module 342 to complete the acquisition of the original image. Preferably, the processing module 400 can be integrated into the industrial control computer to achieve synchronous data acquisition and processing within the integrated device 300, or it can be housed in an external server host that communicates with the industrial control computer. The server host may, for example, be located on the carrier vehicle 100. Furthermore, a power supply unit may be installed on the carrier vehicle 100 to provide a stable power supply for the operation of the integrated device 300.

[0096] Preferably, such as Figure 7As shown, the integrated device 300 may have a housing portion 310, wherein the base plate 320 of the housing portion 310 is provided with a mounting groove 321 for connecting with the second mounting surface 212, so that the integrated device 300 can be detachably fixed to the mounting bracket 200. Further, the housing portion 310 may include two opposing side plates 330 and a working plate 340 disposed between the two side plates 330, wherein the two sides of the working plate 340 are respectively fitted to the edges of the two side plates 330, and the two ends of the working plate 340 are fitted to the edges of the base plate 320, to form a closed structure with an internal cavity. Preferably, the working plate 340 may have a plurality of through holes 341, so that the camera and supplementary light of the image acquisition module 342 can achieve image acquisition and supplementary lighting through these through holes 341. Preferably, the camera of the image acquisition module 342 can be configured as a (high-definition) line scan camera 343, and the fill light can be configured as a laser fill light 345. The line scan camera 343 and the laser fill light 345 can be combined as a set, so that each through-hole 341 on the work plate 340 can be equipped with a set of line scan cameras 343 and laser fill lights 345, thereby achieving cooperative use. Preferably, the integrated device 300 of the present invention can be equipped with eight sets of line scan cameras 343 and laser fill lights 345, that is, including eight line scan cameras 343 and eight laser fill lights 345. These eight sets of line scan cameras 343 and laser fill lights 345 can be arranged on the work plate 340 at different intervals with different orientations. Figure 8 As shown, each line array camera 343 has a field of view of 45°, and the overlap angle of the field of view 344 of adjacent line array cameras 343 is approximately 10°, so that the eight line array cameras 343 can achieve imaging of the tunnel (lining surface) within a range of 290°, that is, excluding the track area within a range of 70° below the tunnel, so as to cover most of the tunnel's apparent ancillary facilities and equipment 600.

[0097] Preferably, such as Figure 7 As shown, the work plate 340 of the integrated device 300 may also be provided with a handle 350 for easy handling or gripping by the user. The handle 350 can be set in a way that does not obstruct the optical path of the line scan camera 343 and the laser filler 345.

[0098] It should be noted that the specific embodiments described above are exemplary. Those skilled in the art can devise various solutions inspired by the disclosure of this invention, and these solutions all fall within the scope of this invention and its protection. Those skilled in the art should understand that this specification and its accompanying drawings are illustrative and do not constitute a limitation on the claims. The scope of protection of this invention is defined by the claims and their equivalents. This specification contains multiple inventive concepts; phrases such as "preferred" or "according to a preferred embodiment" indicate that the corresponding paragraph discloses an independent concept. The applicant reserves the right to file divisional applications based on each inventive concept. Throughout the text, the feature introduced by "preferred" is only an optional mode and should not be construed as mandatory. Therefore, the applicant reserves the right to abandon or delete relevant preferred features at any time.

Claims

1. A method for detecting the detachment of communication signal cables, characterized in that, It includes the following steps: S1. Using an image acquisition module (342) mounted on a carrier vehicle (100), when the carrier vehicle (100) moves at high speed on the track, original images of the tunnel appearance auxiliary facilities and equipment (600) are dynamically acquired by multiple line array cameras (343) in such a way that the field of view (344) of adjacent line array cameras (343) at least partially overlap. S2. Preprocess the original image acquired by the image acquisition module (342); S3. Based on the preprocessed image data, YOLO v5m is used as the basic model. In the backbone layer of YOLO v5m, an SE module containing compression and excitation processes is introduced to classify the tunnel appearance ancillary facilities and equipment (600) in order to identify different types of equipment and store the tunnel appearance ancillary facilities and equipment (600) images in the corresponding databases according to the types of equipment contained therein. S4. Use the Hough straight line detection method to detect each communication signal cable (630) in the communication signal cable library to identify the straight segments of the communication signal cable (630) and detect the slope of the communication signal cable (630). If the slope exceeds the preset threshold, it is judged as suspected detachment. Further extract the image of the corresponding cable tray (640) from the cable tray library generated in step S3. By comparing it with the standard image of the cable tray (640), it is determined whether the cable tray (640) associated with the communication signal cable (630) has undergone positional shift, posture change and / or structural loss. If there is an abnormality, it is confirmed as detachment.

2. The detection method according to claim 1, characterized in that, When the carrier vehicle (100) moves on the track, the image acquisition module (342) can respond to the received pulse signal to achieve acquisition triggering based on the preset second basic information including the trigger mode and trigger parameters. The trigger mode includes time triggering and distance triggering.

3. The detection method according to claim 1 or 2, characterized in that, The preprocessing of the original image acquired by the image acquisition module (342) includes using affine transformation for image geometric correction, using Gaussian filtering for noise removal, and using the Retinex algorithm for image contrast enhancement, so as to improve image quality and feature recognizability.

4. The detection method according to claim 1, characterized in that, The images of each communication signal cable (630) in the communication signal cable library are used as input images. By performing edge detection on the input images and applying Hough transform, the straight segments of the communication signal cable (630) are identified by accumulating votes in the parameter space. Then, the specific position of the cable is reconstructed based on the local maximum value in the accumulator array, thereby determining the slope of the communication signal cable (630) in the input image.

5. A communication signal cable detachment detection system capable of performing the detection method as described in any one of claims 1 to 4, characterized in that, It includes: Image acquisition module (342) is used to dynamically acquire raw images of the tunnel's apparent ancillary facilities and equipment (600); The processing module (400) is used to preprocess the original image acquired by the image acquisition module (342), and then classify the tunnel surface auxiliary facilities and equipment (600) based on the preprocessed image data, and detect the slope of the communication signal cable (630) to realize the detection of the detachment of the communication signal cable (630).

6. The detection system according to claim 5, characterized in that, It is equipped with an integrated device (300) externally mounted on a carrier vehicle (100) and a mounting bracket (200) for connecting the integrated device (300) to the carrier vehicle (100). The image acquisition module (342) is installed inside the integrated device (300). The mounting bracket (200) includes two non-coplanar mounting surfaces. The first mounting surface (211) is used to connect with the carrier vehicle (100), and the second mounting surface (212) is used to connect with the integrated device (300).

7. The detection system according to claim 6, characterized in that, The integrated device (300) is configured with a housing portion (310) including an internal cavity. The bottom plate (320) of the housing portion (310) is provided with a mounting groove (321) for connecting with a second mounting surface (212) of the mounting bracket (200), so that the integrated device (300) can be detachably fixed to the mounting bracket (200).

8. The detection system according to claim 7, characterized in that, The housing (310) of the integrated device (300) includes two opposing side plates (330) and a working plate (340) disposed between the two side plates (330). The working plate (340) has a plurality of through holes (341) spaced apart, so that the line scan camera (343) and laser filler (345) of the image acquisition module (342) can be installed in these through holes (341) in a set manner. The through holes (341) on the working plate (340) can be positioned based on the field of view of the line scan camera (343) so that imaging of the tunnel lining (500) surface, excluding the track area, can be achieved while ensuring that the field of view (344) of adjacent line scan cameras (343) at least partially overlaps.

Images