Quality rapid detection system of elastic curing ballast bed composite unit block

Abstract

The invention discloses a rapid quality detection system for an elastic curing ballast bed composite unit block. The rapid quality detection system comprises an overturning demolding assembly, a detection assembly and an image processing module, the overturning demolding assembly comprises a first rack, overturning mechanisms located on the two sides of the interior of the first rack and a demolding mechanism arranged on the top of the first rack. A first image acquisition device is arranged at the top end in the first rack; the detection assembly comprises a second rack, a moving platform located at the top end in the second rack, a second image acquisition device located on the moving platform, and third image acquisition devices located on the two sides in the second rack. The method is used for solving the problems that in the prior art, the detection efficiency of quality detection of the elastic curing track bed composite unit block is low, and missing detection or misjudgment of defects is easily caused by manual detection, and the purposes of comprehensively and automatically collecting the surface data of the elastic curing track bed composite unit block and improving the efficiency and accuracy of surface defect detection are achieved.

Description

Technical Field

[0001] The present invention relates to the field of rapid detection of elastic cured ballast composite unit blocks, and particularly to a quality rapid detection system for elastic cured ballast composite unit blocks. Background Art

[0002] The elastic cured ballast is a new type of track structure that combines the elastic advantages of traditional ballasted tracks with the integrity, stability, and less maintenance characteristics of ballastless tracks. It forms a composite unit block structure by pouring polyurethane mixture on the top surface of the ballast bed, infiltrating into the pores of the ballast, foaming, bonding, and curing. This technology has been successfully applied in some railway projects, showing good engineering performance. With the development of railway track construction towards high efficiency and high quality, further promotion and application of the elastic cured ballast technology need to break through the existing technical bottlenecks, and innovative research in the detection link has become a key direction.

[0003] The quality inspection of elastic cured ballast composite unit blocks mainly relies on traditional manual methods, which have prominent problems such as long time consumption, high labor intensity, and difficulty in ensuring the accuracy of inspection results, and cannot meet the urgent needs of railway track construction for high efficiency and high quality. Moreover, traditional inspection methods are not only difficult to evaluate key quality indicators such as the uniformity, compactness, and durability of unit blocks in real time and comprehensively, but may also lead to missed detection or misjudgment of defects due to human errors, affecting the overall performance and service life of the ballast bed. Summary of the Invention

[0004] The present invention provides a quality rapid detection system for elastic cured ballast composite unit blocks to solve the problems in the prior art such as low detection efficiency and easy missed detection or misjudgment of defects in the quality inspection of elastic cured ballast composite unit blocks, and to achieve the purpose of comprehensively and automatically collecting the surface data of elastic cured ballast composite unit blocks and improving the efficiency and accuracy of surface defect detection.

[0005] The present invention is achieved through the following technical solutions:

[0006] A quality rapid detection system for elastic cured ballast composite unit blocks includes a flipping and demolding component, a detection component, and an image processing module;

[0007] The flipping and demolding component includes a first rack, flipping mechanisms located on both sides inside the first rack, and a demolding mechanism arranged on the top of the first rack; a first image acquisition device is arranged at the inner top end of the first rack;

[0008] The detection component includes a second rack, a moving platform located at the inner top end of the second rack, a second image acquisition device located on the moving platform, and third image acquisition devices located on both sides inside the second rack;

[0009] The image processing module is signal-connected to the first image acquisition device, the second image acquisition device, and the third image acquisition device.

[0010] While existing technologies offer a limited number of devices / systems for sleeper inspection, conventional sleepers lack elastic unit blocks, limiting their inspection focus to structural elements such as embedded reinforcing bars. Even the few surface inspections available only examine the top rail-bearing surface. Furthermore, the manufacturing process for elastic-cured track bed composite unit blocks differs significantly from that of conventional sleepers. The foaming and solidification of the polyurethane composite material makes demolding more difficult, thus demanding higher standards for surface defect detection. This is one of the core reasons why existing technologies for quality inspection of elastic-cured track bed composite unit blocks rely heavily on manual labor, resulting in low efficiency. Additionally, the inverted production method—with the sleeper below and the unit block above after solidification—further hinders rapid inspection.

[0011] To overcome the aforementioned problems, this application proposes a rapid quality inspection system for elastically cured track bed composite unit blocks. This system uses a flipping demolding assembly to demold and flip the cured composite unit blocks. During this process, a first image acquisition device acquires a bottom image of the composite unit block, thus overcoming the deficiency in existing technologies that completely ignore the detection of bottom defects in the composite unit blocks. Furthermore, the flipping demolding assembly enables simultaneous demolding and bottom inspection, significantly improving the production efficiency of the composite unit blocks.

[0012] The composite unit block, having undergone demolding and bottom inspection via the flipping demolding assembly, is transferred to the inspection assembly with its front side facing upwards (i.e., the rail-supporting surface facing upwards). The second image acquisition device is moved to a suitable position via a moving platform to capture a top-down image of the composite unit block. Simultaneously, third image acquisition devices located on either side capture images of the side walls of the composite unit block. The image data acquired by the first, second, and third image acquisition devices are all transmitted to the image processing module, facilitating the identification of surface defects in the composite unit block using image recognition technology, thereby achieving automatic detection.

[0013] It can be seen that this application abandons the traditional manual inspection approach, overcoming the problems of long inspection time, high labor intensity, low accuracy of inspection results, and strong subjectivity in existing technologies. Moreover, compared with the existing technology that only inspects the rail bearing surface, this application can effectively inspect the rail bearing surface, bottom surface, and side surface of the elastic curing track bed composite unit block, significantly improving the comprehensiveness of the inspection. It can not only accurately measure the various dimensions of the track bed composite unit block to ensure that it meets the design standards, but also identify defects on the surface of the track bed composite unit block such as gaps and missing pieces, providing strong technical support for ensuring the quality of the track bed.

[0014] Furthermore, the system also includes a transfer device for moving the tested elastic-cured track bed composite unit block between the flipping and demolding assembly and the testing assembly. This solution uses the transfer device to move the tested elastic-cured track bed composite unit block between different workstations. The transfer device can be implemented using any existing mobile equipment, such as a carrier trolley, conveyor, or conveyor belt.

[0015] Furthermore, a burr trimming station is provided between the flipping demolding component and the detection component, and a material unloading station is provided on the side of the detection component away from the flipping demolding component.

[0016] After the demolding process is completed by the flipping demolding component, the composite unit block is transported to the burr trimming station. At the burr trimming station, the burrs generated during the demolding process are manually trimmed. After trimming, the composite unit block is transported to the inspection component. After the inspection is completed, it is transported to the unloading station. Based on the inspection results of the inspection component, qualified and unqualified composite unit blocks are sorted and transferred at the unloading station.

[0017] Furthermore, the flipping mechanism includes a clamp for holding the mold of the elastic curing track bed composite unit block and a rotating device for driving the clamp to rotate, the rotating device being installed within the first frame.

[0018] The flipping mechanisms on both sides inside the first frame simultaneously clamp the two ends of the elastic curing track bed composite unit block mold through fixtures, and then drive the two fixtures to rotate synchronously by 180° through the rotating device, thus flipping the inverted mold to the upright position. The first frame is used to provide a station for installing and fixing the rotating device.

[0019] Furthermore, the first frame slides longitudinally within the second frame, and the second frame is fixedly connected to the first frame; the flipping mechanism also includes a first linear drive device for driving the second frame to perform linear reciprocating motion laterally along the first frame, and a second linear drive device for driving the first frame to perform lifting motion within the second frame.

[0020] In this design, the first linear drive device drives the second frame to move laterally, so that when flipping is required, the clamps move inward to contact the end of the mold for clamping; when flipping is not required, the clamps move outward to disengage from the mold. After the mold is clamped by the clamps on both sides, the second linear drive device lifts the first frame, causing the clamps on both sides and the composite unit block to rise together, making way for the flipping of the composite unit block and avoiding the flipping action required by its bottom being restricted.

[0021] Furthermore, it also includes a guide rod disposed on the top of the first frame, the axis of the guide rod being parallel to the driving direction of the first linear drive device; the second linear drive device is slidably engaged with the guide rod.

[0022] When the first linear drive device drives the second frame to move laterally, the first frame and the second linear drive device move synchronously. In order to improve the synchronicity of the lateral movement, the second linear drive device is slidably engaged with the guide rod to avoid the second linear drive device and the first frame moving laterally relative to the second frame, which is beneficial to improving the stability and reliability of the flipping mechanism.

[0023] Furthermore, the clamp includes a turntable connected to the output end of the rotating device, two positioning cylinders disposed on the surface of the turntable, a slide rail disposed on the surface of the turntable, a clamping member slidably engaged on the slide rail, and a third linear drive device for driving the clamping member to slide on the slide rail; the two positioning cylinders are symmetrically distributed on both sides of the turntable; the line connecting the two positioning cylinders is perpendicular to the long axis of the slide rail; and the long axis of the slide rail extends radially along the turntable.

[0024] For the production mold of the elastic curing track bed composite unit block, both end faces have two outwardly protruding columnar handles for gripping and handling. In this solution, the two positioning cylinders on the turntable match the two columnar handles on the same side end face of the mold. When clamping the mold is required, the clamping component is first slid radially outward to the open state via a third linear drive device. Then, the turntable is moved inward as a whole, so that the two positioning cylinders are respectively fitted over the two columnar handles (i.e., the two columnar handles are respectively inserted into the two positioning cylinders). Then, the clamping component is slid radially inward via the third linear drive device until the clamping component firmly clamps the side wall of the mold, after which the flipping operation can be performed. It can be seen that the fixture in this solution clamps the mold through a three-point clamping method, which has extremely strong clamping stability. Moreover, the inherent structure of the mold itself is utilized during the clamping process, which significantly reduces the structural complexity of the fixture and reduces the difficulty of driving and controlling the fixture.

[0025] Furthermore, the demolding mechanism includes a fourth linear drive device with its output end facing downwards, the output end of which is connected to a push head, the surface of which is covered with a flexible layer.

[0026] For the production mold of the elastic curing track composite unit block, after the mold is flipped, the top has several (usually two) holes, with the fourth linear drive device corresponding to each hole. The fourth linear drive device moves the push head downward, inserts it into the corresponding hole, and continues to apply downward force to achieve the demolding operation of the composite unit block. The surface of the push head is covered with a flexible layer to avoid damage to the composite unit block; the flexible layer can be made of any existing flexible material.

[0027] Furthermore, the detection assembly also includes a fifth linear drive device disposed on the top of the second frame with its output end facing downwards, and the output end of the fifth linear drive device is connected to the inkjet printer.

[0028] After the testing process is completed, the inkjet printer is started to print relevant information on the composite unit block according to the preset format, such as the test results, manufacturing time, shift, mold number, etc., thereby significantly improving the functionality of this application.

[0029] Furthermore, the second image acquisition device is a depth camera. Utilizing the principle of structured light projection, the depth camera can acquire the three-dimensional spatial information of the track bed composite unit block, accurately measure its key dimensional parameters such as width, length, and height, and simultaneously clearly capture the subtle features of the track bed composite unit block surface, providing abundant data for surface defect detection.

[0030] Compared with the prior art, the present invention has at least the following advantages and beneficial effects:

[0031] 1. This invention provides a rapid quality inspection system for elastically cured track bed composite unit blocks. It abandons the traditional manual inspection approach, overcoming the problems of long inspection times, high labor intensity, low accuracy, and strong subjectivity inherent in existing technologies. Furthermore, compared to existing technologies that only inspect the rail-bearing surface, this application can effectively inspect the rail-bearing surface, bottom surface, and sides of the elastically cured track bed composite unit blocks, significantly improving the comprehensiveness of the inspection. It can not only accurately measure the various dimensions of the track bed composite unit blocks to ensure they meet design standards, but also identify surface defects such as gaps and missing pieces, providing strong technical support for ensuring track bed quality.

[0032] 2. The present invention provides a rapid quality inspection system for elastically cured track bed composite unit blocks, which is designed with a dedicated flipping mechanism, which is conducive to completing automated inspection operations quickly and efficiently.

[0033] 3. The present invention provides a rapid quality inspection system for elastic curing track bed composite unit blocks. It features a dedicated clamping structure that clamps the mold using a three-point clamping method, resulting in extremely strong clamping stability. Furthermore, the system utilizes the inherent structure of the mold itself during clamping, significantly reducing the structural complexity of the clamp and lowering the difficulty of driving and controlling the clamp. Attached Figure Description

[0034] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, do not constitute a limitation thereof. In the drawings:

[0035] Figure 1 This is a schematic diagram of a specific embodiment of the present invention;

[0036] Figure 2 This is a front view of the flipping demolding component in a specific embodiment of the present invention;

[0037] Figure 3 This is a schematic diagram of the flipping and demolding component during the clamping process in a specific embodiment of the present invention;

[0038] Figure 4 This is a schematic diagram of the flipping and demolding component during the flipping process in a specific embodiment of the present invention;

[0039] Figure 5 This is a schematic diagram of the flipping and demolding component after flipping in a specific embodiment of the present invention;

[0040] Figure 6 This is a schematic diagram of the demolding process of the flipping demolding component in a specific embodiment of the present invention;

[0041] Figure 7 This is a schematic diagram of the mold removal component taking out the mold in a specific embodiment of the present invention;

[0042] Figure 8 is a front view of the turntable in a specific embodiment of the present invention;

[0043] Figure 9 This is a schematic diagram of the detection component in a specific embodiment of the present invention.

[0044] The attached diagram shows the markings and corresponding component names:

[0045] 1-Flipping demolding assembly, 11-First frame, 12-Clamping device, 13-Rotating device, 14-First frame, 15-Second frame, 16-First linear drive device, 17-Second linear drive device, 18-Guide rod, 19-Fourth linear drive device, 110-Push head;

[0046] 2-Detection component, 21-Second rack, 22-Mobile platform, 23-Second image acquisition device, 24-Third image acquisition device;

[0047] 3-Transfer device, 4-Deburring station, 5-Unloading station, 6-Elastic curing track bed composite unit block, 7-Mold, 8-Columnar handle;

[0048] 121-Turntable, 122-Positioning cylinder, 123-Clamping component, 124-Slide rail. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments and accompanying drawings. The illustrative embodiments and descriptions of this invention are for explaining the invention only and are not intended to limit the invention. In the description of this application, it should be understood that terms such as "front," "rear," "left," "right," "upper," "lower," "vertical," "horizontal," "high," "low," "inner," and "outer," indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description. They do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as limiting the scope of protection of this application.

[0050] Example 1:

[0051] like Figure 1 The system shown is a rapid quality inspection system for elastically cured track bed composite unit blocks, including a flipping and demolding component 1, an inspection component 2, and an image processing module;

[0052] The flipping and demolding assembly includes a first frame 11, flipping mechanisms located on both sides inside the first frame 11, and a demolding mechanism disposed on the top of the first frame 11; a first image acquisition device is disposed at the top inside the first frame 11.

[0053] The detection component 2 is as follows Figure 9 As shown, it includes a second frame 21, a mobile platform 22 located at the top of the inside of the second frame 21, a second image acquisition device 23 located on the mobile platform 22, and a third image acquisition device 24 located on both sides inside the second frame 21.

[0054] The image processing module is signal-connected to the first image acquisition device, the second image acquisition device 23, and the third image acquisition device 24.

[0055] It also includes a transfer device 3 for transferring the tested elastically cured track bed composite unit block between the flipping demolding component 1 and the detection component 2.

[0056] A burr trimming station 4 is provided between the flipping demolding component 1 and the detection component 2, and a material unloading station 5 is provided on the side of the detection component 2 away from the flipping demolding component 1.

[0057] In this embodiment, the demolding mechanism includes a fourth linear drive device 19 with its output end facing downward. The output end of the fourth linear drive device 19 is connected to a push head 110, and the surface of the push head 110 is covered with a flexible layer.

[0058] In a more preferred embodiment, the detection component 2 further includes a fifth linear drive device disposed on the top of the second frame 21 with its output end facing downward, and the output end of the fifth linear drive device is connected to the inkjet printer.

[0059] In a more preferred embodiment, the second image acquisition device 23 is a depth camera, employing 3D structured light technology to obtain depth image data; furthermore, both the first image acquisition device and the third image acquisition device 24 can be high-resolution industrial cameras. The acquired images and measurement data are rapidly transmitted to the image processing module. Based on image processing algorithms, this data undergoes in-depth processing and analysis, enabling precise measurement of various dimensions of the track bed blocks to ensure compliance with design standards. It also allows for the identification of surface defects on the track bed blocks, such as gaps and spalling, providing strong technical support for ensuring the quality of the track bed blocks.

[0060] In a more preferred embodiment, the first image acquisition device is mounted on the crossbeam of the first frame 11. However, the object distance here is relatively limited, only about 500mm, making it difficult to obtain a clear and complete image of the bottom of the unit block. Therefore, in this embodiment, three cameras can be used to capture images of the bottom of the unit block from different angles. Through reasonable parameter settings and image stitching technology, omnidirectional image acquisition of the bottom of the unit block can be achieved, providing sufficient data support for subsequent bottom defect detection.

[0061] The detection method in this embodiment includes the following steps:

[0062] S1. The mold and the undemolded elastic cured track bed composite unit blocks inside it are transported together to the first frame 11 in the production state; the production state means that the rail bearing surface is facing down, that is, the sleeper is below and the polyurethane track bed block is above.

[0063] S2. Disassemble the top template at this time, and obtain the bottom image data of the elastic curing track bed composite unit block through the first image acquisition device;

[0064] S3. The mold is flipped by the flipping mechanism so that the elastic curing track bed composite unit block is placed upright, that is, the rail bearing surface is facing up.

[0065] S4. Demolding is performed through a demolding mechanism to obtain independent elastically cured track bed composite unit blocks;

[0066] S5. The elastic curing track bed composite unit block is transferred to the rough edge trimming station 4 by the transfer device 3, and the rough edges are trimmed manually.

[0067] S6. The elastic curing track bed composite unit block is transferred to the second frame 21 by the transfer device 3; the second image acquisition device 23 is adjusted to a suitable position by the moving platform 22, the suitable position being a position that can completely acquire the top image data of the elastic curing track bed composite unit block; the top surface image data of the elastic curing track bed composite unit block is acquired by the second image acquisition device 23; the end surface image data of the elastic curing track bed composite unit block is acquired by the third image acquisition device 24.

[0068] S7. The fifth linear drive device drives the inkjet printer to descend to a set height above the elastic curing track composite unit block, and inkjet prints the set information on the surface of the elastic curing track composite unit block.

[0069] S8. The elastic curing track bed composite unit block is transferred to the unloading station 5 via the transfer device 3.

[0070] Example 2:

[0071] A rapid quality testing system for elastically cured track bed composite unit blocks, based on Example 1, such as... Figures 2 to 7 As shown, the flipping mechanism includes a clamp 12 for holding the mold of the elastic curing track bed composite unit block and a rotating device 13 for driving the clamp 12 to rotate. The rotating device 13 is installed in the first frame 14.

[0072] The first frame 14 slides longitudinally within the second frame 15, and the second frame 15 is fixedly connected to the first frame 11. The flipping mechanism also includes a first linear drive device 16 for driving the second frame 15 to perform linear reciprocating motion laterally along the first frame 11, and a second linear drive device 17 for driving the first frame 14 to perform lifting and lowering motion within the second frame 15.

[0073] It also includes a guide rod 18 disposed on the top of the first frame 11, the axis of the guide rod 18 being parallel to the driving direction of the first linear drive device 16; the second linear drive device 17 is slidably engaged with the guide rod 18.

[0074] like Figure 8 As shown, the clamp 12 in this embodiment includes a turntable 121 connected to the output end of the rotating device 13, two positioning cylinders 122 disposed on the surface of the turntable 121, a slide rail 124 disposed on the surface of the turntable 121, a clamping member 123 slidably engaged on the slide rail 124, and a third linear drive device for driving the clamping member 123 to slide on the slide rail 124; the two positioning cylinders 122 are symmetrically distributed on both sides of the turntable 121; the line connecting the two positioning cylinders 122 is perpendicular to the long axis of the slide rail 124; and the long axis of the slide rail 124 extends radially along the turntable 121.

[0075] The rotating device 13 uses a motor, such as a stepper motor or a servo motor; the motors on both sides can be controlled to move in tandem.

[0076] In a more preferred embodiment, a cross slide for adjusting the position of the fourth linear drive device 19 may be provided on the top of the first frame 11.

[0077] In this embodiment, please refer to the process of flipping the mold using the flipping mechanism. Figures 3 to 6 Specifically, it includes the following steps:

[0078] S301, in the initial state as follows Figure 3 As shown;

[0079] S302, Adjust the turntable 121 so that the two positioning cylinders 122 are respectively aligned with the two columnar handles 8 on the same side end face of the mold; Drive the clamping member 123 to move radially outward along the slide rail 124 through the third linear drive device, so that the clamping member 123 slides radially outward to the open state.

[0080] S303, the first linear drive device 16 drives the second frame 15 and the first frame 14 to move inward synchronously, so that the two positioning cylinders 122 on the same side are respectively sleeved on the two columnar handles 8; then the third linear drive device drives the clamping member 123 to move radially inward along the slide rail 124, so that the clamping member 123 firmly clamps the top of the mold 7; the state at this time is as follows. Figure 4 As shown;

[0081] S304, the second linear drive devices 17 on both sides drive the first frames 14 on both sides to rise synchronously, and then control the turntables 121 on both sides to rotate 180° in the same direction, flipping the mold 7; this process is as follows Figure 5 As shown;

[0082] S305, the second linear drive devices 17 on both sides drive the first frames 14 on both sides to descend synchronously, placing the mold 7 on the worktable or transfer device below; at this time, the state is as follows Figure 6 As shown;

[0083] S306. After that, release the clamping of the mold 7, and drive the second frame 15 and the first frame 14 to move laterally outward to reset via the first linear drive device 16.

[0084] In this embodiment, please refer to the demolding process performed by the demolding mechanism. Figures 7 to 9 Specifically, it includes the following steps:

[0085] S401, Position the pusher head 110 at the bottom of the fourth linear drive unit 19 directly opposite the hole at the top of the mold; the state at this time is as follows. Figure 7 As shown;

[0086] S402, The fourth linear drive device 19 causes the push head 110 to move downward and insert into the corresponding hole, and continues to apply downward force until the demolding operation is completed;

[0087] S403, the first linear drive device 16 drives the second frame 15 and the first frame 14 to move inward synchronously, so that the two positioning cylinders 122 on the same side are respectively fitted over the two columnar handles 8; then the third linear drive device drives the clamping member 123 to move radially inward along the slide rail 124, so that the clamping member 123 firmly clamps the top of the mold 7; the second linear drive devices 17 on both sides drive the first frames 14 on both sides to rise synchronously, lifting the mold 7; at this time, the state is as follows. Figure 8 As shown;

[0088] S404, Transfer the elastically cured track bed composite unit block, and flip mold 7 again. The state at this time is as follows: Figure 9 As shown; simply lower mold 7 and move mold 7 away.

[0089] Example 3:

[0090] A rapid quality inspection system for elastically cured track bed composite unit blocks, based on any of the above embodiments, wherein the second image acquisition device 23 employs a three-dimensional laser scanner capable of acquiring point cloud data; the scanned point cloud data is sent to the image processing module and processed using the following method:

[0091] Step 1: Point cloud data preprocessing to prepare high-quality point cloud data for subsequent feature extraction and parameter calculation. This specifically includes:

[0092] Step 11, Point Cloud Data Loading: Obtain the 3D point cloud data file, parse the XYZ coordinate information in the file, and load the point cloud data into memory and store it as an N×3 numerical matrix, where N represents the number of points and 3 represents the 3D coordinate dimension.

[0093] Step 12, Point Cloud Denoising and Downsampling: Statistical outlier removal is performed on the loaded original point cloud. Noise points are removed based on the neighborhood distribution characteristics of the point cloud. Then, the point cloud density is reduced by voxel downsampling method, which improves processing efficiency while maintaining shape features. The processed point cloud is stored as a new numerical matrix.

[0094] Step 13, Point Cloud Region Extraction: Based on the characteristics of the sleeper structure, extract the highest part of the point cloud along the Z-axis (in this embodiment, the highest 1 / 4 of the point cloud with the highest Z coordinate is taken), which is defined as the top point cloud; the top point cloud includes key features such as the rail bearing surface and the pre-embedded sleeve, and the extracted regional point cloud is stored separately for subsequent processing.

[0095] Step 2, Rail bearing surface plane fitting and parameter calculation: Determine the spatial position and orientation of the sleeper's rail bearing surface to provide a reference plane for subsequent measurements. Specifically, this includes:

[0096] Step 21, Plane Model Fitting: The extracted top point cloud is fitted with a plane using the RANSAC algorithm. The optimal plane model parameters [a, b, c, d] are calculated through iterative optimization, where (a, b, c) represent the plane normal vector and d represents the plane constant term. This plane model represents the sleeper bearing surface, which is defined as the fitted bearing plane.

[0097] Step 22, In-plane point extraction: Based on the fitted plane model, calculate the distance from all points to the plane, and identify points whose distance is less than a threshold (in this embodiment, the threshold is 0.5 mm) as in-plane points of the rail bearing surface. Store these in-plane points of the rail bearing surface as a subset of the point cloud for subsequent calculation of rail bearing surface parameters.

[0098] Step 23: Calculation of rail bearing surface parameters: Based on the fitted rail bearing plane and in-plane point data, calculate the key parameters of the rail bearing surface. The key parameters of the rail bearing surface include:

[0099] Rail bearing surface height: Calculate the vertical distance from the center of a point within the rail bearing surface to the lowest point of the entire point cloud;

[0100] Rail bearing surface inclination angle: Calculate the angle between the rail bearing surface normal vector and the Z-axis, reflecting the levelness of the rail bearing surface;

[0101] Angle between the rail bearing surface and the ZY plane: Calculate the angle between the rail bearing surface normal vector and the X-axis, reflecting the lateral inclination of the rail bearing surface. These parameters are stored and output in floating-point form.

[0102] Step 3, Point Cloud Projection and Rasterization: Convert the 3D point cloud into a 2D image to facilitate the detection of pre-embedded sleeve holes using image processing methods. Specifically, this includes:

[0103] Step 31, Plane Projection Transformation: Project the points in the track plane onto the fitted track plane, and convert the three-dimensional points into two-dimensional UV coordinates through coordinate transformation to establish the mapping relationship from three-dimensional space to two-dimensional plane. The projected two-dimensional point set is stored as an N×2 matrix.

[0104] Step 32, 2D gridding: Divide the projected 2D point cloud region into uniform grid units, determine the grid resolution based on the point cloud density (0.5 mm / pixel in this embodiment), and mark each grid unit as occupied or unoccupied based on whether it contains point cloud data, generating a binary image matrix.

[0105] Step 33, Image Storage and Representation: Store the gridded binary image as a two-dimensional matrix, where 1 indicates that there is a point cloud at that location (occupied) and 0 indicates that there is no point cloud (idle). This image is used for subsequent hole detection processing.

[0106] Step 4: Detection and Positioning of Embedded Sleeve Holes: Identify and position the embedded sleeve holes on the sleeper bearing surface, and calculate the center-to-center distance between the holes. Specifically, this includes:

[0107] Step 41, Hole Contour Detection: Apply a hole detection algorithm to the gridded binary image to identify the background area surrounded by foreground pixels, use a contour tracking algorithm to extract the boundary contours of all holes, and store the contour point set as a coordinate sequence.

[0108] Step 42, Ellipse Fitting and Center Location: Perform ellipse fitting on the detected hole boundary contours, use the least squares method to calculate the optimal ellipse parameters, thereby determining the center coordinates of each hole (in the UV coordinate system), and store the hole center coordinates as a two-dimensional vector.

[0109] Step 43, 3D coordinate restoration: Using the mapping relationship established in step 31, the center coordinates of the hole in the UV coordinate system are back-projected into the 3D space to obtain the 3D coordinates of the hole center. The 3D coordinates of the hole centers corresponding to the two pre-embedded sleeves are stored as 3D vectors respectively.

[0110] Step 44, Hole Spacing Calculation: Calculate the Euclidean distance between the centers of the two pre-embedded sleeve holes, which is the key parameter for the pre-embedded sleeve spacing. This parameter is stored and output in floating-point form.

[0111] Step 5: Calculation of the distance between the edges of the gauge blocks: Measure the distance between the edges of the gauge blocks to ensure that the sleeper dimensions meet the design requirements. This specifically includes:

[0112] Step 51: Offset plane construction: Offset a set distance (28mm in this embodiment) along the direction of the normal vector of the rail bearing surface to construct a parallel plane. This offset plane represents the plane position of the edge of the rail gauge baffle. Store the plane equation parameters after offset.

[0113] Step 52, plane intersection point extraction: Calculate the intersection points of the original point cloud and the offset plane, and extract the point cloud that is less than the threshold (0.3mm in this embodiment). These intersection points represent the edge positions of the gauge baffle and are stored as a subset of the point cloud.

[0114] Step 53, Left and right edge clustering: Based on the sign of the U coordinate of the intersection point in the plane coordinate system, the intersection point is divided into two clusters, left and right, representing the left and right track gauge baffle edge point sets respectively. The density clustering algorithm is used to optimize the edge point set and remove outliers.

[0115] Step 54: Spacing Statistical Calculation: Perform distance statistical calculations on the edge point sets of the left and right gauge baffles, including center distance, minimum distance, average distance, median distance, and standard deviation. These statistics comprehensively describe the spacing characteristics of the gauge baffle edges and are stored in the form of floating-point number sets.

[0116] Step 6: Calculation of the inclined angle of the gauge baffle: Measure the inclination angle of the gauge baffle to ensure that the inclined surface design meets the specifications. Specifically, this includes:

[0117] Step 61, Sloping surface point cloud extraction: Using the left and right track gauge baffle edge point sets obtained in step 53 as seed points, extract the neighboring point cloud (the neighboring radius is 1.0 mm in this embodiment) from the original point cloud. These point clouds represent the sloping surface area of ​​the track gauge baffle and are stored as left and right sloping surface point sets respectively.

[0118] Step 62, Plane Fitting of the Sloping Surfaces: Perform plane fitting on the point clouds of the left and right sloping surfaces using the RANSAC algorithm to obtain the plane equation parameters of the left and right sloping surfaces. These parameters are stored as plane model vectors.

[0119] Step 63, Calculate the angle between the left and right inclined planes and the rail bearing surface. This is achieved by calculating the angle between the normal vectors of the two planes. The inclination angles of the left and right inclined planes are obtained, stored and output as floating-point numbers.

[0120] Step 7, Result Storage and Output: Persistently store all measurement results and provide visual output for quality verification. Specifically, this includes:

[0121] Step 71, Parameter Result Storage: Store all calculated parameters (rail surface height, inclination angle, hole spacing, baffle spacing, slope angle, etc.) in a structured text format, including parameter name, value and unit, to facilitate subsequent quality traceability and analysis.

[0122] Step 72, Intermediate Process Storage: Save the intermediate results of key processing stages, including denoised point clouds, points in the rail bearing surface, hole outlines, edge point sets, etc., in text format for process verification and debugging.

[0123] Step 73, Visualization Output: Generate visualization results of the detection process, including point cloud rendering, plane fitting diagram, hole detection effect diagram, edge point distribution diagram, etc., stored in image format to provide intuitive quality inspection basis.

[0124] The advantages of the above processing method are:

[0125] (1) Traditional methods typically use the entire point cloud or other threshold segmentation to fit the plane, which is easily affected by point clouds from non-rail-bearing surfaces such as the side and bottom surfaces. This method innovatively selects the 1 / 4 portion with the highest Z coordinate in the point cloud for RANSAC plane fitting. This ensures that the fitted plane strictly represents the actual working surface of the sleeper's rail-bearing platform, greatly improving the fitting accuracy and anti-interference ability, and is the basis for all subsequent dimensional calculations.

[0126] (2) This method projects the three-dimensional point cloud onto the fitted track surface plane, transforms it into two-dimensional UV coordinates, and dynamically and adaptively determines the size of the gridded voxels based on the point cloud density to generate a high-precision binary image. Therefore, this method transforms the problem of detecting complex three-dimensional shapes into a more reliable two-dimensional image processing problem, significantly reducing computational complexity and providing a stable and accurate input for subsequent hole recognition.

[0127] (3) In the binarized image, this method does not rely directly on the traditional contour search, but first identifies all background areas, and then accurately locates the "hole" by judging whether the background area is completely surrounded by the foreground. This method can effectively eliminate interference such as grooves and edge damage, and accurately identify the real bolt holes specifically for the hole features of the sleeper pre-embedded sleeve.

[0128] (4) This method is based on the fitted plane equation and simultaneously calculates the absolute height, flatness (tilt angle), and angle relative to the ZY reference plane of the rail bearing surface. This not only provides the spatial position of the rail bearing surface, but also comprehensively describes its attitude orientation, which is crucial for ensuring the smoothness of the rail after laying. All these parameters are derived from a single fitting step, making it extremely efficient.

[0129] (5) This method does not directly measure the distance between the centers of two holes in three-dimensional space (which is susceptible to point cloud defects and noise). Instead, it first fits the center of an ellipse on the outline of each target two-dimensional projected hole, and then uses the projection transformation relationship to map the center point back to three-dimensional space to calculate the distance. This method bypasses the instability of direct fitting in three-dimensional space, and has higher accuracy and stronger robustness.

[0130] (6) Construct a virtual measurement plane 28mm above the rail bearing surface and obtain the intersection line between this plane and the sleeper point cloud. Separate the intersection points into left and right clusters using a clustering algorithm, and use multiple statistical measures such as center distance, minimum distance, mean, and median to characterize the baffle spacing, rather than a single value. This provides a more comprehensive manufacturing quality assessment and can effectively avoid random errors.

[0131] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0132] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Additionally, the term "connection" as used herein, unless otherwise specified, can refer to a direct connection or an indirect connection via other components.

Claims

1. A system for rapid quality detection of a unit of elastic cured track bed composite, characterized in that, The device comprises a turnover demolding assembly (1), a detection assembly (2) and an image processing module. The turnover demolding assembly comprises a first rack (11), turnover mechanisms located at both sides of the inside of the first rack (11), and a demolding mechanism arranged on the top of the first rack (11); a first image acquisition device is arranged at the top end of the inside of the first rack (11). The detection assembly (2) comprises a second rack (21), a moving platform (22) located at the top end of the inside of the second rack (21), a second image acquisition device (23) located on the moving platform (22), and third image acquisition devices (24) located at both sides of the inside of the second rack (21). The image processing module is signal connected with the first image acquisition device, the second image acquisition device (23) and the third image acquisition device (24).

2. The system for rapid quality detection of a unit block of a flexible solidified road bed according to claim 1, wherein, The device further comprises a transfer device (3) for transferring the measured elastic solidified track bed composite unit block between the turnover demolding assembly (1) and the detection assembly (2).

3. The system for rapid quality detection of a unit block of a flexible solidified road bed of claim 1, wherein, An edge trimming station (4) is arranged between the turnover demolding assembly (1) and the detection assembly (2), and a discharging station (5) is arranged on the side of the detection assembly (2) away from the turnover demolding assembly (1).

4. The system for rapid quality detection of a unit block of a flexible solidified road bed of claim 1, wherein, The turnover mechanism comprises a clamp (12) for clamping the mold of the elastic solidified track bed composite unit block, and a rotating device (13) for driving the clamp (12) to rotate, wherein the rotating device (13) is installed in a first frame (14).

5. The system for rapid quality detection of a unit block of a resiliently cured road bed of claim 4, wherein, The first frame (14) is longitudinally slidingly fitted in a second frame (15), and the second frame (15) is fixedly connected with the first rack (11); the turnover mechanism further comprises a first linear driving device (16) for driving the second frame (15) to make linear reciprocating motion transversely along the first rack (11), and a second linear driving device (17) for driving the first frame (14) to make lifting motion in the second frame (15).

6. The system for rapid quality detection of a unit block of a resiliently cured road bed of claim 5, wherein, The device further comprises a guide rod (18) arranged on the top of the first rack (11), wherein the axis of the guide rod (18) is parallel to the driving direction of the first linear driving device (16); and the second linear driving device (17) is slidingly fitted with the guide rod (18).

7. The system for rapid quality detection of a unit block of a flexible solidified road bed of claim 4, wherein, The clamp (12) comprises a rotating disc (121) connected with the output end of the rotating device (13), two positioning cylinders (122) arranged on the surface of the rotating disc (121), a sliding rail (124) arranged on the surface of the rotating disc (121), a clamping piece (123) slidingly fitted on the sliding rail (124), and a third linear driving device for driving the clamping piece (123) to slide on the sliding rail (124); the two positioning cylinders (122) are symmetrically distributed on both sides of the rotating disc (121); the line connecting the two positioning cylinders (122) is perpendicular to the long axis of the sliding rail (124); and the long axis of the sliding rail (124) extends radially along the rotating disc (121).

8. The system for rapid quality detection of a unit block of a flexible solidified road bed of claim 1, wherein, The demolding mechanism comprises a fourth linear driving device (19) with an output end downward, the output end of the fourth linear driving device (19) is connected with a pushing head (110), and the surface of the pushing head (110) is covered with a flexible layer.

9. The system for rapid quality detection of a unit block of a flexible solidified road bed of claim 1, wherein, The detection assembly (2) further comprises a fifth linear driving device provided on the top of the second rack (21) and having an output end downward, and the output end of the fifth linear driving device is connected with a code spraying device.

10. The system for rapid quality detection of a unit block of a flexible solidified road bed of claim 1, wherein, The second image acquisition device (23) is a depth camera.

Images