Tunnel section digital sampling device for urban rail passenger car and working method of tunnel section digital sampling device
By installing a digital sampling device for tunnel sections on urban rail buses, the camera module and AI processing are used to achieve all-weather and high-frequency detection of tunnel environment status, the problems of low efficiency and insufficient timeliness of existing detection methods are solved, and the design is lightweight, Removable and industrially optimized appearance.
Patent Information
- Application Number
- CN202411210073.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-30
- Publication Date
- 2024-11-15
AI Technical Summary
The existing tunnel environmental status detection method is inefficient, manual inspections rely on manpower and are not frequent, engineering vehicle inspection can only be carried out during operation, and the data timeliness and overallness are insufficient.
Design a digital sampling device for tunnel sections for urban rail passenger cars, installed in the front of the head hood of the electric bus, above the hook, and below the wiper. It uses multiple camera modules, linear lasers and linear array cameras for online real-time sampling, combined with AI Processing realizes all-weather high-frequency detection.
All-weather and high-frequency detection of tunnel environmental status is realized, the detection efficiency and timeliness and globality of data are improved, the device design is lightweight, miniaturized, and symmetrical, with a hierarchical detachable design and industrially optimized appearance.
Smart Images

Figure CN118960681A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of urban rail transit equipment, and in particular relates to a tunnel section digital sampling device for an urban rail passenger car and a working method thereof. Background Art
[0002] The operation lines of urban rail transit include a large number of underground lines, which are mainly composed of tunnels and their auxiliary structures, providing space, lighting, gas, energy, network, etc. required for vehicle operation. A good tunnel environment is an important part of ensuring the safe operation of urban rail transit, and it is very necessary to detect the tunnel environment status during the operation process.
[0003] However, the existing detection of tunnel environmental status mostly adopts two methods: manual inspection or detection by engineering vehicles equipped with equipment. Among them, the manual inspection method requires a lot of manpower, and manual inspection often requires the use of professional auxiliary tools to observe the overall situation of the tunnel section. The detection results need to be judged based on the experience of the operator, and the inspection efficiency is relatively low. Engineering vehicles are equipped with professional tools, which reduces the dependence on manpower. However, due to the operation of urban rail, operations can only be carried out during the operation window period, and the attendance frequency is often not high. The sampling intervals for the above two inspection methods for tunnel environmental status monitoring are generally long, and the sampling results are also the status of the object when it is not in operation. The data provided for tunnel environmental status detection are insufficient in timeliness and globality. Summary of the invention
[0004] The purpose of the present invention is to overcome the above problems existing in the prior art, and to provide a tunnel section digital sampling device for urban rail passenger cars and a working method thereof, using an urban rail mainline electric passenger car as the carrying object, and installing the tunnel section digital sampling device in the middle position of the front of the vehicle driver's cab hood, above the coupler, and below the wiper. When the device is in mainline operation, the tunnel section can be digitally sampled online and in real time, and the sampled data can be analyzed to realize all-weather and high-frequency detection of the tunnel environment status.
[0005] In order to achieve the above technical objectives and the above technical effects, the present invention is implemented through the following technical solutions:
[0006] The embodiment of the first aspect of the present invention provides a tunnel cross-section digital sampling device for urban rail passenger cars, comprising a bottom plate and a shell sealed and connected to the top of the bottom plate, wherein a plurality of camera modules are arranged on the bottom plate, the plurality of camera modules are distributed at the edge of the bottom plate, and the shell is sleeved on the outside of the plurality of camera modules;
[0007] The camera module includes a module housing, an area array camera, a line laser and a linear array camera. The area array camera, the line laser and the linear array camera are all arranged outside the module housing. A plurality of viewing windows are opened on the housing, and the viewing windows are used to provide a shooting field of view for the area array camera, the line laser and the linear array camera.
[0008] Furthermore, an anti-drop rope is arranged in the shell, one end of the anti-drop rope is fixedly installed in the shell, the other end of the anti-drop rope is fixedly installed with an end head, and the other end of the end head is fixedly connected with a mounting bracket.
[0009] Furthermore, a fixing hole is provided on the edge of the top surface of the base plate, a notch is provided on the outer surface of the shell, a bolt fixing groove is provided at the bottom of the notch, a locking bolt is arranged in the bolt fixing groove, one end of the locking bolt passing through the bolt fixing groove is located in the fixing hole, and one end of the locking bolt passing through the fixing hole is threadedly connected with a locking nut.
[0010] Furthermore, a wiring hole is provided in the center of the lower bottom surface of the bottom plate, and a heat shrink tubing fixing member is provided in the wiring hole;
[0011] The bottom surface of the bottom plate is also provided with a connecting flange, and the inner hole of the connecting flange is cocentric with the wiring hole;
[0012] An inner sealing groove and an outer sealing groove are also provided with the inner hole center of the connecting flange as the center of the circle, the inner sealing groove and the outer sealing groove are both concentric with the wiring hole, and the outer diameter of the inner sealing groove is smaller than the inner diameter of the outer sealing groove;
[0013] O-type sealing rings are arranged in the inner sealing groove and the outer sealing groove.
[0014] Furthermore, the bottom surface of the bottom plate and the outer surface of the shell are provided with a plurality of heat dissipation modules, and the heat dissipation modules include a plurality of weight-reducing windows arranged in parallel;
[0015] The shell is provided with a plurality of reinforcing ribs;
[0016] The top edge of the bottom plate is also integrally formed with a lug, which is snap-fitted into the shell.
[0017] Furthermore, the number of the camera modules is three and they are distributed in an isosceles triangle.
[0018] Furthermore, two fixing brackets are arranged outside the camera module, a camera fixing position is arranged below the fixing brackets, and the fixing brackets are fixed to the camera fixing position by bolt connection;
[0019] The camera fixing position is integrally formed and arranged on the top of the bottom plate;
[0020] The camera module is fixed between two fixing brackets by means of bolt connection.
[0021] Furthermore, the area array camera, line laser and line array camera are arranged on the same side of the module housing;
[0022] The center point of the area array camera, the center point of the line array camera and the center point of the line laser form a right triangle;
[0023] The center point of the area array camera and the center point of the line array camera are located on one of the right-angled sides of the right triangle, and the center point of the line array camera and the center point of the line laser are located on the other right-angled side of the right triangle.
[0024] Furthermore, a plurality of reinforcement members are evenly distributed outside the module housing;
[0025] The module housing is also provided with an electric control component, a heat dissipation structure and an equipment electrical interface;
[0026] The electric control component is used to control the area array camera, line laser and line array camera;
[0027] The heat dissipation structure is used for heat dissipation of area array cameras, line lasers, line array cameras and electronic control components.
[0028] The embodiment of the second aspect of the present invention provides a working method of a tunnel section digital sampling device for a city rail passenger car, the method comprising the following steps:
[0029] S1: Sampling device installation
[0030] The tunnel section digital sampling device for the urban rail passenger car is installed at the middle position of the front of the driver's cab hood, above the coupler and below the wiper of the urban rail passenger car;
[0031] S2: Image capture
[0032] When the vehicle is running, the square wave signal generated by the axle end speed sensor is transmitted to the linear array camera, triggering the linear array camera to shoot. The shooting position is the image of the intersection of the line laser and the tunnel surface. The linear array camera obtains a line of image each time it shoots.
[0033] S3: Multiple lines of images are stitched together to form a picture data file
[0034] The linear array camera sets the shooting working frequency in combination with the signal frequency of the speed sensor, and splices and processes the captured multiple lines of images to form a continuous, visible picture data file of the tunnel surface;
[0035] S4: Image data preprocessing and neural network processing
[0036] The image data file is transferred to the AI board, and then goes through basic image preprocessing such as denoising and filtering by the image algorithm program, and then image cropping and normalization are performed to obtain the recognition image data. The recognition image data is then sent to the deep neural network for matrix calculation, and features are extracted and matched to obtain output. Then, NMS non-maximum suppression and threshold are used to filter out false matches and analyze the preliminary abnormal status.
[0037] S5: Filtering image algorithm false positives
[0038] The area array camera and the line laser convert the distance from the photographed tunnel surface to the tunnel center through the principle of triangulation to obtain distance parameters. The preliminary abnormal state combined with the distance parameters can filter out the image algorithm false alarm data;
[0039] Among them, the image algorithm false alarm data includes image shooting false alarm data and processing abnormality false alarm data;
[0040] S6: Output abnormal image
[0041] The identified abnormal image of the tunnel surface is output, and watermarks of vehicle speed information, station information, and vehicle location information are added to the image to generate an abnormal image with fault point information.
[0042] Compared with the prior art, the present invention has the following beneficial effects:
[0043] 1. The present invention provides a tunnel section digital sampling device for urban rail passenger cars and a working method thereof. The tunnel section digital sampling device is installed in the middle position of the front of the vehicle driver's cab hood, above the coupler, and below the wiper. When the vehicle is operating on the main line, the device can perform online and real-time digital sampling of the tunnel section, and the sampling data can be analyzed to realize all-weather and high-frequency detection of the tunnel environment status.
[0044] 2. The present invention adopts a lightweight, miniaturized, and symmetrical design. The vehicle body structure is fully considered when arranging the device structure, and the length dimension reserves a working buffer space for the coupling. The installation position of the device design is close to the geometric center of the tunnel, so that each group of cameras in the device can capture the largest possible area of the tunnel under the same viewing angle; at the same time, the number of cameras is also reduced, so that the device can be lightweight and miniaturized. After the camera is installed symmetrically and centered, the captured picture is also symmetrical, which reduces the difficulty of picture splicing and data processing in the software work, takes into account the requirements of efficient output of the final state from the perspective of raw data acquisition, and optimizes the structural design.
[0045] 3. In the present invention, a hierarchical detachable design is adopted. The tunnel section digital sampling device and the vehicle body are connected by a deformed flange, and two groups of bolt holes are evenly distributed on the flange surface; the connection between the device and the vehicle body is provided with a sealing structure. In order to ensure the sealing effect, an O-ring is designed on both the inner and outer sides of the bolt hole distribution circle. Bolts are used to connect the device bottom plate and the camera module, and between the bottom plate and the outer shell. The wiring harness in the device passes through the middle wire hole to enter the vehicle. A sealing groove is provided at the wire harness passing position, and a heat shrink tubing with adhesive is used to complete the sealing of the wire passing position.
[0046] 4. Industrial design optimization of the external structure of the device in the present invention. The external structure design of the device not only meets the mechanical installation requirements, but also fully considers integration into the vehicle appearance, and adopts a full curved surface structure design. The bottom plate uses the space of the non-contact sealing position to set an oblong structure window to reduce the weight of the equipment. A viewing window is added to the side of the outer shell. The viewing window is a spatial oblong structure, which reduces the weight, improves the heat dissipation efficiency of the device, and increases the curved surface elements of the device; a square structure is added to the front end of the outer shell to strengthen the strength of the force-bearing surface of the equipment; the bolt fixing structure of the outer shell adopts a hidden design, which is convenient for the disassembly operation of the equipment and does not affect the overall shape.
[0047] 5. The device of the present invention is provided with three camera modules, each of which adopts a combination of a linear array camera, a planar array camera and a linear laser, that is, a combination of a 2D camera, a 3D camera and a linear laser. The viewing angles of the three camera modules can completely cover the cross-section except the vehicle contact surface; the 2D camera and the 3D camera are triggered by the vehicle shaft end speed sensor signal to ensure the synchronization of the camera operation; the positions of the 2D camera and the 3D camera are the tunnel surface illuminated by the linear laser, and the linear laser adopts non-visible light with adjustable frequency to ensure the clarity of the image taken and the safety of the equipment during operation. BRIEF DESCRIPTION OF THE DRAWINGS
[0048] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:
[0049] Figure 1 It is a structural explosion diagram of the tunnel section digital sampling device for urban rail passenger cars provided by the present invention;
[0050] Figure 2 It is a structural explosion diagram of the tunnel section digital sampling device for urban rail passenger cars provided by the present invention;
[0051] Figure 3 It is a flow chart of the working method of the tunnel section digital sampling device provided by the present invention.
[0052] Among them, the figure markings are: 1. Base plate; 101. Fixing hole; 102. Connecting flange; 103. Heat shrink tube fixing piece; 104. Lug; 105. Camera fixing position; 106. Weight reduction window; 107. Inner sealing groove; 108. Outer sealing groove; 2. Shell; 201. Bolt fixing groove; 202. Viewing angle window; 3. Camera module; 301. Area array camera; 302. Line laser; 303. Line array camera; 304. Electronic control component; 305. Reinforcement; 306. Heat dissipation structure; 307. Equipment electrical interface; 4. Fixing bracket; 5. O-ring; 6. Anti-slip rope; 601. Terminal; 602. Mounting bracket. DETAILED DESCRIPTION
[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0054] In the description of the present invention, it should be understood that the terms "opening", "upper", "lower", "thickness", "top", "middle", "length", "inside", "all around" and the like indicating orientation or positional relationship are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the components or elements referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.
[0055] Example 1
[0056] like Figure 1 to Figure 2 As shown, a tunnel section digital sampling device for urban rail passenger cars comprises:
[0057] A bottom plate 1 and a shell 2 sealed and connected to the top of the bottom plate 1. Three camera modules 3 are arranged on the bottom plate 1. The three camera modules 3 are distributed at the edge of the bottom plate 1 and are distributed in an isosceles triangle.
[0058] The device housing 2 is designed as a thin-walled part, which has the characteristics of integration and lightness;
[0059] An anti-slip rope 6 is provided in the housing 2, one end of the anti-slip rope 6 is fixedly installed in the housing 2, the other end of the anti-slip rope 6 is fixedly installed with an end 601, and the other end of the end 601 is fixedly connected with a mounting bracket 602;
[0060] The equipment is provided with an anti-slip rope 6, one end of which is fixed inside the equipment and the other end is fixed to the vehicle body structure. The two ends of the anti-slip rope 6 are fixed by crimping, which reduces the space of the end 601, and at the same time, it comes with a mounting bracket 602 for hanging on the vehicle body, which is convenient for equipment layout;
[0061] Two fixing brackets 4 are arranged outside the camera module 3, and a camera fixing position 105 is arranged below the fixing brackets 4. The fixing brackets 4 are fixed to the camera fixing position 105 by bolt connection.
[0062] The camera fixing position 105 is integrally formed and arranged on the top of the bottom plate 1;
[0063] The camera module 3 is fixed between two fixing brackets 4 by bolt connection;
[0064] Each camera module 3 is fixed on the bottom plate 1 by two fixing brackets 4. The positioning accuracy of the camera is high, providing a basic structure for data collection. Since the three camera modules 3 are distributed in an isosceles triangle on the bottom plate 1, the camera modules 3 symmetrically arranged on the left and right sides of the bottom plate 1 are installed on the same horizontal plane, keeping the line laser on the same horizontal plane, and the appearance is symmetrical; within the viewing angle of the top camera module 3, the laser lines of the camera modules on the left and right sides will appear. In order to ensure the independence of the light source, the installation surface of the top camera module 3 is different from the other two groups of camera modules 3; at the same time, the arrangement of the three groups of camera modules 3 takes into account the operating space, so that the design of the equipment is minimized and lightweight. This part is connected and fixed to the vehicle body through the fixed hole 101 in the middle of the bottom plate 1. The connection process has an anti-slip structure, that is, an anti-slip rope 6 and a sealing structure, that is, an O-ring 5, which assists the digital sampling device in meeting the anti-slip and sealing requirements. Finally, the housing 2 of the digital sampling device is installed on the bottom plate 1 to complete the assembly of the entire equipment. The disassembly process is the reverse of the assembly, and both disassembly and assembly are relatively convenient;
[0065] A fixing hole 101 is provided at the top edge of the bottom plate 1, a notch is provided on the outer surface of the shell 2, a bolt fixing groove 201 is provided at the bottom of the notch, a locking bolt is provided in the bolt fixing groove 201, one end of the locking bolt passing through the bolt fixing groove 201 is located in the fixing hole 101, and one end of the locking bolt passing through the fixing hole 101 is threadedly connected with a locking nut; since the bolt fixing groove 201 is provided at the bottom of the notch, the locking bolt adopts such a hidden design, which is convenient for the overall disassembly operation of the device without affecting the overall shape;
[0066] That is, the base plate 1 and the housing 2 are connected and fixed by locking bolts and locked by locking nuts;
[0067] A wiring hole is provided in the center of the bottom surface of the bottom plate 1, and a heat shrink tube fixing member 103 is provided in the wiring hole;
[0068] There is a channel for wires and ropes in the middle, and a heat shrink sleeve fixing piece 103 is provided for fixing the heat shrink sleeve. Three layers of grooves are provided to allow the colloid to flow out of the groove during the heat shrinking process. The sleeve has a concave-convex structure here, so as to achieve reliable waterproofing;
[0069] A connecting flange 102 is also provided on the bottom surface of the bottom plate 1, and the inner hole of the connecting flange 102 is cocentric with the wiring hole;
[0070] On the deformed 145 connecting flange 102, 8 holes with a diameter of 11 mm are arranged unevenly and symmetrically, and connected to the vehicle body, so that the connection strength is high;
[0071] An inner sealing groove 107 and an outer sealing groove 108 are formed with the inner hole of the connecting flange 102 as the center. The inner sealing groove 107 and the outer sealing groove 108 are both concentric with the wiring hole. The outer diameter of the inner sealing groove 107 is smaller than the inner diameter of the outer sealing groove 108.
[0072] O-rings 5 are provided in both the inner sealing groove 107 and the outer sealing groove 108;
[0073] An O-ring 5 is provided in the sealing groove; after the equipment base plate is assembled with the vehicle, the protection level requirement of IP67 can be achieved.
[0074] The bottom surface of the bottom plate 1 and the outer surface of the shell 2 are also provided with multiple groups of heat dissipation modules, and the heat dissipation modules include multiple weight-reducing windows 106 arranged in parallel; multiple reinforcing ribs are also provided on the shell 2, specifically, the outer shell side wall, transition part and front end of the shell 2 are all provided with reinforcing ribs; the reinforcing ribs are used to strengthen the strength of the force-bearing surface of the device;
[0075] A plurality of weight-reducing windows 106 are provided to reduce the overall weight of the digital sampling device while achieving a heat dissipation function, which is beneficial to the normal operation of internal components;
[0076] The top edge of the bottom plate 1 is also integrally formed with a lug 104, which is snap-fitted into the housing 2;
[0077] There are lugs 104 on the left and right sides of the bottom plate, which can ensure that the housing can be placed on the bottom plate after the fastening bolts are released. Human intervention is required to separate the housing and the bottom plate to prevent the equipment from falling suddenly during the disassembly process;
[0078] The camera module 3 includes a module housing, an area array camera 301, a line laser 302, and a line array camera 303. The area array camera 301, the line laser 302, and the line array camera 303 are all arranged outside the module housing. A plurality of viewing angle windows 202 are provided on the housing 2. The viewing angle windows 202 are used to provide a shooting field of view for the area array camera 301, the line laser 302, and the line array camera 303. The viewing angle windows 202 are arc-shaped structures, which are oblong in space so as not to interfere with the shooting of the area array camera 301, the line laser 302, and the line array camera 303.
[0079] The area array camera 301, the line laser 302 and the line array camera 303 are arranged on the same side of the module housing;
[0080] The center point of the area array camera 301, the center point of the line array camera 303 and the center point of the line laser 302 form a right triangle;
[0081] The center point of the area array camera 301 and the center point of the line array camera 303 are located on one of the right-angled sides of the right triangle, and the center point of the line array camera 303 and the center point of the line laser 302 are located on the other right-angled side of the right triangle;
[0082] Since each camera module has an area array camera 301, a line laser 302, and a line array camera 303, the three camera modules can capture a fan angle of >270°, which can effectively sample the tunnel surface;
[0083] The combination of the area array camera 301 and the line laser 302 can calculate the distance of the sampling object through the principle of triangulation; the combination of the line laser 302 and the line array camera 303 can obtain a high-definition image of the sampling object;
[0084] A plurality of reinforcement members 305 are evenly distributed outside the module housing;
[0085] The module housing is also provided with an electric control component 304, a heat dissipation structure 306 and an equipment electrical interface 307;
[0086] Three camera modules 3 are provided, and the three camera modules 3 are arranged in an isosceles triangle, that is, a symmetrical design is adopted. Each camera module 3 adopts a combination of a linear array camera 303, a planar array camera 301 and a linear laser 302, that is, a combination of a 2D camera, a 3D camera and a linear laser. The viewing angles of the three camera modules 3 can completely cover the cross-section except the vehicle contact surface. Both the 2D camera and the 3D camera are triggered by the vehicle shaft end speed sensor signal to ensure the synchronization of the camera operation. The shooting positions of the 2D camera and the 3D camera are the tunnel surface irradiated by the linear laser. The linear laser adopts non-visible light with adjustable frequency to ensure the clarity of the shooting image and the safety of the equipment during operation.
[0087] In actual use, the device can be installed in the middle position between the front of the driver's cab hood, above the hook and below the wiper. Since the vehicle structure has been fully considered when arranging the device, this installation position allows each camera in the camera module 3 to capture the largest possible area of the tunnel within the same viewing angle. At the same time, due to this arrangement, the number of cameras can be reduced to a certain extent, making the device lightweight and miniaturized.
[0088] This embodiment provides a tunnel section digital sampling device for urban rail passenger cars, which uses urban rail mainline electric passenger cars as the carrying object, and installs the tunnel section digital sampling device in the middle position of the front of the vehicle driver's cab hood, above the coupler, and below the wiper. When the device is in mainline operation, the tunnel section can be digitally sampled online and in real time, and the sampled data can be analyzed to realize all-weather and high-frequency detection of the tunnel environment status.
[0089] Example 2
[0090] like Figure 3 As shown, this embodiment provides a working method of a tunnel section digital sampling device for a city rail passenger car, the method comprising the following steps:
[0091] S1: Sampling device installation
[0092] The tunnel cross-section digital sampling device for urban rail passenger cars is installed in the middle of the front of the driver's cab hood, above the coupler, and below the wiper of the urban rail passenger car; the data of the area array camera 301 and the line array camera 303 are transmitted using the Gigabit Ethernet communication protocol based on the UDP protocol, which defines the camera control and configuration as well as the data type transmitted to ensure the integrity and reliability of data transmission. The operation of the line laser 302 is controlled by the RS485 communication protocol, and the data signal adopts the differential transmission method, which can meet the control frequency requirements and has strong anti-interference ability;
[0093] S2: Image capture
[0094] When the vehicle is running, the square wave signal generated by the axle end speed sensor is transmitted to the linear array camera 303, triggering the linear array camera 303 to shoot. The shooting position is the image at the intersection of the line laser 302 and the tunnel surface. The linear array camera 303 obtains a line of image each time it shoots.
[0095] During the operation of the vehicle, the speed sensor at the end of the axle plays a key role and is responsible for real-time monitoring and measuring the speed of the vehicle. Specifically, when the vehicle is running, the speed sensor at the end of the axle generates a square wave signal. This square wave signal has the characteristics of constant amplitude, adjustable duty cycle, adjustable phase and good stability, making it very suitable for speed measurement and transmission. The generated square wave signal is then transmitted to the line array camera 303, triggering the line array camera 303 to shoot. The line array camera is a camera that uses a line array sensor to collect images. Its working principle is to scan the object to be photographed through the line array sensor, and then convert the scanned signal into a digital image. The line array camera 303 captures the image of the intersection of the line laser 302 and the tunnel surface. The line array camera 303 obtains a line of image each time it shoots.
[0096] S3: Multiple lines of images are stitched together to form a picture data file
[0097] The linear array camera 303 sets the shooting working frequency in combination with the signal frequency of the speed sensor, and splices and processes the captured multiple lines of images to form a continuous, visible picture data file of the tunnel surface;
[0098] Specifically, during the shooting process, the linear array camera 303 continuously shoots multiple lines of images at a working frequency that matches the vehicle speed. Since these images are captured based on the signal frequency of the speed sensor, the intervals between them are uniform and accurate, which is convenient for subsequent image stitching processing; at the same time, the data collected by the three groups of linear array cameras 303 are stitched together, and the overall viewing angle of the stitched images covers the upper surface area of the tunnel, forming a continuous and visible picture data file of the tunnel surface;
[0099] S4: Image data preprocessing and neural network processing
[0100] The image data file is transferred to the AI board, and after basic image preprocessing such as denoising and filtering by the image algorithm program, the image is cropped and normalized to obtain the recognition image data;
[0101] Identify the image data and then use the deep neural network to perform matrix calculations, extract and match features to get output, and then use NMS non-maximum suppression and thresholding to filter out false matches and analyze the preliminary abnormal state;
[0102] The specific process of obtaining the recognition image data is as follows: During the operation of the vehicle, due to various environmental factors and the limitations of the equipment itself, the image data file may contain noise, which will interfere with the quality and clarity of the image data file. The denoising algorithm can effectively reduce or eliminate these noises and improve the clarity of the image data file; after the denoising process is completed, filtering is performed to help eliminate high-frequency noise and details in the image while retaining the main features of the image; cropping is then performed to select the target area from the original image and remove redundant information according to specific needs or goals; finally, after normalization, the pixel values of the image data file are mapped to a specific range, usually scaling the pixel values to between 0 and 1; thereby obtaining the recognition image data;
[0103] The specific process of analyzing the initial abnormal state is as follows:
[0104] S41: Based on the feature learning and expression capabilities of deep neural networks, complex feature information is extracted from the recognition image data. Through multi-layer convolution, pooling and full connection operations, the neural network can learn the key features in the recognition image data. In the feature extraction stage, the deep neural network will generate a feature matrix, which contains the representation of the key features in the image. These key features may be indicators of cracks, deformations or other abnormal phenomena on the tunnel surface. By comparing the feature matrices between different recognition image data, the similarities and differences between them can be determined;
[0105] S42: Use NMS (non-maximum suppression) and threshold filtering techniques to eliminate false matches. NMS suppresses the matching results with lower response in the local area and only retains the matches with the highest response, thereby reducing repeated and redundant matches. Threshold filtering filters out the results with lower matching degrees according to the set threshold and only retains the matches with higher matching degrees.
[0106] S43: After eliminating the wrong matches in step S42, the matching results are analyzed to obtain the output of the preliminary abnormal status, which may include information such as the abnormal location, type, severity, etc., providing a basis for subsequent abnormal processing and decision-making;
[0107] S5: Filtering image algorithm false positives
[0108] The area array camera 301 and the line laser 302 convert the distance from the photographed tunnel surface to the tunnel center through the principle of triangulation to obtain distance parameters. The initial abnormal state is combined with the distance parameters to filter out the false alarm data of the image algorithm.
[0109] Among them, the image algorithm false alarm data includes image shooting false alarm data and processing abnormality false alarm data;
[0110] The line laser 302 emits a laser beam, which is irradiated onto the tunnel surface to form a laser line. The area array camera 301 is responsible for capturing the projection of the laser line on the tunnel surface. Since the positions of the area array camera 301 and the line laser 302 are known, the distance from the tunnel surface to the tunnel center can be calculated by measuring the position of the laser line in the image taken by the area array camera 301 using the triangulation principle.
[0111] Among them, the principle of triangulation is based on the geometric relationship of triangle similarity;
[0112] After obtaining these distance parameters, they can be combined with preliminary abnormal status data for more in-depth analysis. Since image algorithms may be affected by factors such as lighting, shadows, and noise when processing images, resulting in false positives, combining distance parameters can effectively filter out these false positive data.
[0113] S6: Output abnormal image
[0114] Output the identified abnormal image of the tunnel surface, and add watermarks of vehicle speed information, station information, and vehicle location information to the image to generate an abnormal image of the fault point information;
[0115] The abnormal image output may be directly displayed on a display screen, or the image may be saved to a local or remote server for subsequent viewing and analysis.
[0116] This embodiment provides a working method of a digital sampling device for a tunnel section for a city rail passenger car, which provides a reference for the use of the digital sampling device for a tunnel section for a city rail passenger car.
[0117] The present invention provides a tunnel section digital sampling device for urban rail passenger cars and a working method thereof. The tunnel section digital sampling device is installed in the middle position of the front of the vehicle driver's cab hood, above the coupler, and below the wiper, using an urban rail mainline electric passenger car as the carrying object. When the vehicle is in mainline operation, the device can perform online and real-time digital sampling of the tunnel section, and the sampling data is analyzed to realize all-weather and high-frequency detection of the tunnel environment status.
[0118] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.
[0119] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.
Claims
1. A tunnel section digital sampling device for urban rail passenger cars, characterized in that: It comprises a base plate (1) and a shell (2) sealedly connected to the top of the base plate (1), a plurality of camera modules (3) are arranged on the base plate (1), the plurality of camera modules (3) are distributed at the edge of the base plate (1), and the shell (2) is sleeved on the outside of the plurality of camera modules (3); The camera module (3) comprises a module housing, an area array camera (301), a line laser (302) and a line array camera (303); the area array camera (301), the line laser (302) and the line array camera (303) are all arranged outside the module housing; a plurality of viewing angle windows (202) are provided on the housing (2); the viewing angle windows (202) are used to provide a shooting field of view for the area array camera (301), the line laser (302) and the line array camera (303).
2. The tunnel cross-section digital sampling device for urban rail passenger cars according to claim 1, characterized in that: An anti-slip rope (6) is arranged in the housing (2), one end of the anti-slip rope (6) is fixedly installed in the housing (2), the other end of the anti-slip rope (6) is fixedly installed with an end head (601), and the other end of the end head (601) is fixedly connected to a mounting bracket (602).
3. The tunnel section digital sampling device for urban rail passenger cars according to claim 1, characterized in that: A fixing hole (101) is provided on the top edge of the bottom plate (1), a notch is provided on the outer surface of the shell (2), a bolt fixing groove (201) is provided at the bottom of the notch, a locking bolt is arranged in the bolt fixing groove (201), one end of the locking bolt passes through the bolt fixing groove (201) and is located in the fixing hole (101), and one end of the locking bolt passes through the fixing hole (101) and is threadedly connected to a locking nut.
4. The tunnel cross-section digital sampling device for urban rail passenger cars according to claim 1, characterized in that: A wiring hole is provided in the center of the lower bottom surface of the base plate (1), and a heat shrink sleeve fixing member (103) is provided in the wiring hole; The bottom surface of the base plate (1) is also provided with a connecting flange (102), and the inner hole of the connecting flange (102) is cocentric with the wiring hole; An inner sealing groove (107) and an outer sealing groove (108) are formed with the inner hole of the connecting flange (102) as the center of the circle. The inner sealing groove (107) and the outer sealing groove (108) are both concentric with the wiring hole. The outer diameter of the inner sealing groove (107) is smaller than the inner diameter of the outer sealing groove (108). O-type sealing rings (5) are arranged in both the inner sealing groove (107) and the outer sealing groove (108).
5. The tunnel section digital sampling device for urban rail passenger cars according to claim 1, characterized in that: The bottom surface of the base plate (1) and the outer surface of the shell (2) are also provided with a plurality of heat dissipation modules, wherein the heat dissipation modules include a plurality of weight-reducing windows (106) arranged in parallel; The shell (2) is provided with a plurality of reinforcing ribs; The top edge of the bottom plate (1) is also integrally formed with a lug (104), and the lug (104) is snap-fitted into the shell (2).
6. The tunnel section digital sampling device for urban rail passenger cars according to claim 1, characterized in that: The number of the camera modules (3) is three and they are distributed in an isosceles triangle.
7. The tunnel cross-section digital sampling device for urban rail passenger cars according to claim 1, characterized in that: Two fixing brackets (4) are arranged outside the camera module (3), a camera fixing position (105) is arranged below the fixing brackets (4), and the fixing brackets (4) are fixed to the camera fixing position (105) by means of bolt connection; The camera fixing position (105) is integrally formed and arranged on the top of the bottom plate (1); The camera module (3) is fixed between two fixing brackets (4) by means of bolt connection.
8. The tunnel section digital sampling device for urban rail passenger cars according to claim 1, characterized in that: The area array camera (301), the line laser (302) and the line array camera (303) are arranged on the same side of the module housing; The center point of the area array camera (301), the center point of the line array camera (303) and the center point of the line laser (302) form a right triangle; The center point of the area array camera (301) and the center point of the line array camera (303) are located on one of the right-angled sides of the right triangle, and the center point of the line array camera (303) and the center point of the line laser (302) are located on the other right-angled side of the right triangle.
9. The tunnel section digital sampling device for urban rail passenger cars according to claim 1, characterized in that: A plurality of reinforcement members (305) are evenly distributed on the outside of the module housing; The module housing is also provided with an electric control component (304), a heat dissipation structure (306) and an equipment electrical interface (307); The electric control component (304) is used to control the area array camera (301), the line laser (302) and the line array camera (303); The heat dissipation structure (306) is used for dissipating heat of the area array camera (301), the line laser (302), the line array camera (303) and the electric control component (304).
10. A working method of the tunnel section digital sampling device for urban rail passenger cars according to any one of claims 1 to 9, characterized in that: The method comprises the following steps: S1: Sampling device installation The tunnel section digital sampling device for the urban rail passenger car is installed at the middle position of the front of the driver's cab hood, above the coupler and below the wiper of the urban rail passenger car; S2: Image capture When the vehicle is running, the square wave signal generated by the axle end speed sensor is transmitted to the linear array camera 303, triggering the linear array camera 303 to shoot. The shooting position is the image at the intersection of the line laser 302 and the tunnel surface. The linear array camera 303 obtains a line of image each time it shoots. S3: Multiple lines of images are stitched together to form a picture data file The linear array camera 303 sets the shooting working frequency in combination with the signal frequency of the speed sensor, and splices and processes the captured multiple lines of images to form a continuous, visible picture data file of the tunnel surface; S4: Image data preprocessing and neural network processing The image data file is transferred to the AI board, and then goes through basic image preprocessing such as denoising and filtering by the image algorithm program, and then image cropping and normalization are performed to obtain the recognition image data. The recognition image data is then sent to the deep neural network for matrix calculation, and features are extracted and matched to obtain output. Then, NMS non-maximum suppression and threshold are used to filter out false matches and analyze the preliminary abnormal status. S5: Filtering image algorithm false positives The area array camera 301 and the line laser 302 convert the distance from the photographed tunnel surface to the tunnel center through the principle of triangulation to obtain distance parameters. The preliminary abnormal state combined with the distance parameters can filter out the image algorithm false alarm data; Among them, the image algorithm false alarm data includes image shooting false alarm data and processing abnormality false alarm data; S6: Output abnormal image The identified abnormal image of the tunnel surface is output, and watermarks of vehicle speed information, station information, and vehicle location information are added to the image to generate an abnormal image with fault point information.
