Dimension measurement method for bogie and handheld device

Through preprocessing and feature extraction based on two-dimensional and three-dimensional images, combined with handheld equipment, the problems of low efficiency and insufficient accuracy of bogie size measurement in the prior art are solved, and fast and accurate on-site detection is achieved, reducing the risk of accidents .

CN118913104APending Publication Date: 2024-11-08CSR CHENGDU
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411247798.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-06
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

The prior art has low efficiency and accuracy when measuring bogie sizes, which are affected by the lighting environment and surface conditions, and require disassembly and transport, affecting the measurement efficiency.

Method used

The dimension measurement method based on preprocessing, feature extraction and weighted average calculation of two-dimensional and three-dimensional images is adopted, and the on-site inspection is carried out in combination with hand-held devices, including data acquisition, storage and analysis modules, to achieve fast and accurate dimensional measurement.

Benefits of technology

It improves detection accuracy and efficiency, can directly measure on site without dismantling the bogie, reducing the probability of accidents.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118913104A_ABST
    Figure CN118913104A_ABST
Patent Text Reader

Abstract

The invention discloses a size measurement method for a bogie and handheld equipment, and the method comprises the steps: S1, carrying out the preprocessing operation based on a two-dimensional image of the bogie; s2, performing preprocessing operation based on the three-dimensional image of the bogie acquired by the point cloud; s3, performing feature extraction based on the preprocessed two-dimensional image to obtain a first group of comparison data; s4, performing feature extraction based on the preprocessed three-dimensional image to obtain a second group of comparison data; and S5, based on the first group of comparison data and the second group of comparison data, calculating a weighted average value to obtain final size data. According to the invention, the two-dimensional image and the three-dimensional image are subjected to feature extraction to obtain the comparison data, and then weighting is carried out to obtain the final size data, so that the detection precision and the detection efficiency are ensured; meanwhile, all the unit modules are integrated into handheld equipment, operation is easy and convenient, the bogie can be directly detected on site, the bogie does not need to be disassembled or transported to other places, and work efficiency is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the field of vehicle structure detection, and in particular to a dimension measurement method and a handheld device for a bogie. Background Art

[0002] The bogie is a key component of rail vehicles or subway vehicles used to support the car body, install wheels and transmit traction. By measuring the size of the bogie, potential dimensional errors can be discovered and corrected in a timely manner, avoiding component wear, breakage and other failures caused by dimensional problems, and reducing the probability of accidents.

[0003] In the prior art, manual measurement, laser measurement technology and ultrasonic detection technology are often used to measure the size of the bogie, but these solutions all have certain defects. Manual measurement relies on the operating experience of the inspector and has low efficiency; laser measurement technology has high requirements on the lighting environment. If there is strong light interference or the target surface can cause strong reflection, it will affect the accuracy of the measurement result; ultrasonic detection technology has certain requirements on the surface conditions of the detection target. If its surface is rough or there is a coating, it will affect the propagation and reflection of the ultrasonic wave, thereby affecting the accuracy of the measurement result. In addition, laser measurement technology and ultrasonic detection technology often require the bogie to be disassembled in advance and then transported to the detection site, which affects the measurement efficiency. Summary of the invention

[0004] In view of this, the present invention proposes a dimension measurement method for a bogie, comprising:

[0005] S1: Perform preprocessing operations based on the two-dimensional image of the bogie;

[0006] S2: preprocessing operation based on the three-dimensional image of the bogie acquired by point cloud;

[0007] S3: performing feature extraction based on the two-dimensional image after the preprocessing operation to obtain a first set of comparison data;

[0008] S4: performing feature extraction based on the three-dimensional image after the preprocessing operation to obtain a second set of comparison data;

[0009] S5: Based on the first set of comparison data and the second set of comparison data, a weighted average is calculated to obtain final size data.

[0010] According to a preferred embodiment, in S1, the preprocessing operation includes graying and denoising.

[0011] According to a preferred embodiment, in S2, the preprocessing operation includes phase unpacking and three-dimensional reconstruction.

[0012] According to a preferred embodiment, in S3, the first set of comparison data includes: bogie thickness a1, bogie diameter a2 and suspension clearance a3 between the bogie frame and the vehicle body.

[0013] According to a preferred embodiment, in S4, the second set of comparison data includes: bogie thickness b1, bogie diameter b2 and suspension clearance b3 between the bogie frame and the vehicle body.

[0014] According to a preferred implementation, in S5, the specific calculation expression is: D_i=(w_ai)*ai+(w_bi)*bi, wherein D_i represents the weighted dimension value of the i-th feature dimension, w_ai represents the weight parameter of feature ai, and w_bi represents the weight parameter of feature bi.

[0015] According to a preferred embodiment, the bogie-oriented dimension measurement method further comprises the steps of:

[0016] S6: Generate a measurement report based on the final dimension data.

[0017] Accordingly, the present invention also proposes a handheld device for measuring dimensions of a bogie, which is used to implement the above method, comprising:

[0018] Data acquisition module, used to acquire two-dimensional images and three-dimensional images, including two acquisition modes: frontal shooting and oblique shooting;

[0019] A data storage module, used for creating archives for the two-dimensional image and the three-dimensional image, storing the original image, feature data and final size data;

[0020] A data analysis and reporting module, used to implement preprocessing, feature extraction and weighted average of the two-dimensional image and the three-dimensional image;

[0021] The data acquisition module, data storage module and data analysis and reporting module are integrated into a handheld device.

[0022] According to a preferred embodiment, the handheld device for measuring dimensions of the bogie further includes a network expansion interface.

[0023] According to a preferred embodiment, the data acquisition module includes a built-in mask layer and a line layer for assisting shooting.

[0024] The present invention extracts features from two-dimensional images and three-dimensional images respectively to obtain comparison data, and then weights them to obtain final size data, thereby ensuring detection accuracy and detection efficiency. At the same time, each unit module is integrated into a handheld device, which is easy to operate and can directly detect the bogie on site without disassembling the bogie or transporting it to other locations, thereby improving work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a flow chart of a dimension measurement method for a bogie of the present invention;

[0026] Figure 2 It is a module diagram of a handheld device for measuring dimensions of a bogie according to the present invention. DETAILED DESCRIPTION

[0027] The bogie is a key component of rail vehicles or subway vehicles used to support the car body, install wheels and transmit traction. By measuring the size of the bogie, potential dimensional errors can be discovered and corrected in a timely manner, avoiding component wear, breakage and other failures caused by dimensional problems, and reducing the probability of accidents.

[0028] In the existing technology, manual measurement, laser measurement technology and ultrasonic detection technology are often used to measure the size of the bogie, but these solutions all have certain defects, such as: manual measurement depends on the operating experience of the inspector and is inefficient; laser measurement technology has high requirements for the lighting environment. If there is strong light interference or the target surface can cause strong reflection, it will affect the accuracy of the measurement result; ultrasonic detection technology has certain requirements for the surface conditions of the detection target. If the surface is rough or coated, it will affect the propagation and reflection of the ultrasonic wave, thereby affecting the accuracy of the measurement result. In addition, laser measurement technology and ultrasonic detection technology often require the bogie to be disassembled in advance and then transported to the detection site, which affects the measurement efficiency.

[0029] In view of this, the present invention proposes a dimension measurement method and a handheld device for a bogie. In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0030] A dimension measurement method for a bogie, comprising:

[0031] S1: Perform preprocessing operations based on the two-dimensional image of the bogie;

[0032] S2: preprocessing operation based on the three-dimensional image of the bogie acquired by point cloud;

[0033] S3: performing feature extraction based on the two-dimensional image after the preprocessing operation to obtain a first set of comparison data;

[0034] S4: performing feature extraction based on the three-dimensional image after the preprocessing operation to obtain a second set of comparison data;

[0035] S5: Based on the first set of comparison data and the second set of comparison data, a weighted average is calculated to obtain final size data.

[0036] Among them, the three-dimensional image collected by the point cloud includes three-dimensional coordinates and additional attribute information:

[0037] Three-dimensional coordinates: Each point cloud data point contains its coordinate information in three-dimensional space, that is, its position in the X, Y, and Z directions.

[0038] In addition to basic 3D coordinates, point cloud data can also carry additional information about the properties of the point: color information, reflectivity / intensity, and timestamp.

[0039] The present invention extracts features from two-dimensional images and three-dimensional images respectively to obtain comparison data, and then performs weighting to obtain final size data, thereby ensuring detection accuracy and detection efficiency.

[0040] According to a preferred embodiment, in S1, the preprocessing operation includes graying and denoising.

[0041] Wherein, the grayscale step includes:

[0042] Reading the image: First, read the raw color image using the image processing library.

[0043] Application formula: Grayscale conversion is usually achieved through weighted average method. The conversion formula is: Grayscale value = 0.299×R+0.587×G+0.114×B, where R, G, and B represent the pixel values ​​of the red, green, and blue channels, respectively. Traverse each pixel of the image, apply the above formula to calculate its grayscale value, and replace the original pixel value, save and output the processed grayscale image

[0044] The denoising step adopts Gaussian filtering, including:

[0045] Determine the filter size: Use a smaller Gaussian filter, such as a 3x3 kernel, to retain more details.

[0046] Apply Gaussian filtering: Perform a convolution operation on the grayscale image based on a Gaussian kernel of a selected size, with the largest weight at the center and gradually decreasing towards the surrounding areas.

[0047] Adjust the standard deviation: A smaller standard deviation means a sharper filtering effect, while a larger standard deviation leads to a smoother filtering effect. It can be adjusted according to the actual processing situation and the denoised image is output after completion.

[0048] According to a preferred embodiment, in S2, the preprocessing operation includes phase unpacking and three-dimensional reconstruction.

[0049] Phase unwrapping is a process used in phase measurement profilometry to recover a continuous phase image from a wrapped phase image. Due to physical or technical limitations, the directly measured phase values ​​are usually wrapped in the interval [-π,π] or [0,2π), which is called the wrapped phase. The purpose of phase unwrapping is to convert these wrapped phase values ​​into their true, continuous phase values.

[0050] In the present invention, after obtaining the wrapped phase image through interferometry or phase measurement technology, the phase is gradually unwrapped along a predetermined row path, while considering the phase difference between adjacent pixels. Specifically, starting from the starting point, the phase difference between adjacent pixels is gradually integrated along the selected path. During the integration process, it is necessary to consider the periodicity of the phase difference, that is, when the phase difference exceeds 2π or is less than -2π, corresponding adjustments need to be made to ensure the continuity of the integration; during the integration process, discontinuities may be encountered due to noise, phase mutations or measurement errors. These discontinuities will destroy the continuity assumption of the phase difference, resulting in errors in the integration results. In order to deal with these problems, phase quality diagrams, residual detection, branch cutting and other methods can be used to identify and correct discontinuities.

[0051] After phase unpacking, the bogie is reconstructed in 3D, including surface reconstruction and model optimization:

[0052] Surface reconstruction includes connecting points in the point cloud data into a triangular mesh to form a surface model of the object, and reconstructing the surface of the object by solving implicit equations, preferably using a signed distance function.

[0053] Model optimization includes optimizing the reconstructed three-dimensional model, such as smoothing the surface, removing redundant parts, etc.

[0054] Phase unwrapping and three-dimensional reconstruction are important preprocessing operations in bogie dimensional measurement, and are of great significance for improving measurement accuracy and efficiency.

[0055] According to a preferred embodiment, in S3, the first group of comparison data includes: bogie thickness a1, bogie diameter a2 and suspension clearance a3 between the bogie frame and the car body; According to a preferred embodiment, in S4, the second group of comparison data includes: bogie thickness b1, bogie diameter b2 and suspension clearance b3 between the bogie frame and the car body.

[0056] The measurement of bogie thickness is the basis for evaluating whether the overall dimensions of the bogie meet the design requirements. By comparing the bogie thickness, i.e., a1 and b1, the wear of the bogie during use can be detected, as well as whether there are dimensional changes caused by material fatigue or external forces.

[0057] The measurement of bogie diameter reflects the roundness and coaxiality of its wheelset. By comparing the diameter data a2 and b2, problems such as wheelset wear and deformation can be discovered in a timely manner.

[0058] Suspension clearance is a key parameter to ensure a safe distance between the bogie and the car body. Appropriate clearance can prevent collision and damage caused by vibration or impact during train operation. By comparing the suspension clearance a3 and b3 between the bogie frame and the car body, the wear or adjustment status of the suspension system can be found.

[0059] The bogie thickness, diameter and suspension clearance are selected as comparative data, which can be better correlated to the structural integrity, dynamic performance of the bogie and the safety of train operation.

[0060] According to a preferred implementation, in S5, the specific calculation expression is: D_i=(w_ai)*ai+(w_bi)*bi, wherein D_i represents the weighted dimension value of the i-th feature dimension, w_ai represents the weight parameter of feature ai, and w_bi represents the weight parameter of feature bi.

[0061] The two-dimensional image provides a plan view of the bogie, making it easy to observe its outline, shape and surface features; while the three-dimensional image provides richer spatial information, including height, depth and the relative position of each component. Combining the two and performing weighted calculations can make full use of their respective advantages to obtain more comprehensive and accurate measurement results; for some factors that may cause errors, such as image resolution, lighting conditions, measurement angles, etc., these errors can be balanced through weighted calculations to improve the overall measurement accuracy.

[0062] The design of weighting coefficients w_ai and w_bi should take into account the following criteria:

[0063] Data quality: Evaluate the data quality of 2D and 3D images, including resolution, clarity, noise level, etc. High-quality data should be given a higher weight.

[0064] Information complementarity: Analyze the complementarity of 2D and 3D images in providing information. If a parameter is difficult to measure accurately in 2D images but easy to obtain in 3D images, the weight of the parameter in 3D images should be increased accordingly.

[0065] According to a preferred embodiment, the bogie-oriented dimension measurement method further comprises the steps of:

[0066] S6: Generate a measurement report based on the final dimension data.

[0067] The measurement report includes the original 2D and 3D image data, as well as related data such as the measurement date, measurement environment, etc., and the final dimension data after preprocessing and weighted average. In addition, at the end of the measurement report, targeted maintenance suggestions can be made based on the problem diagnosis results, including repairs, replacement of parts, adjustment of parameters, etc.

[0068] Accordingly, the present invention proposes a handheld device for measuring dimensions of a bogie, which is used to implement the above-mentioned method for measuring dimensions of a bogie, comprising:

[0069] Data acquisition module, used to acquire two-dimensional images and three-dimensional images, including two acquisition modes: frontal shooting and oblique shooting;

[0070] A data storage module, used for creating archives for the two-dimensional image and the three-dimensional image, storing the original image, feature data and final size data;

[0071] The data analysis and reporting module is used to implement preprocessing, feature extraction and weighted average of the two-dimensional image and the three-dimensional image.

[0072] The data acquisition module selects a 2D plus 3D camera device to acquire two-dimensional and three-dimensional images, and can be easy to hold and convenient to acquire.

[0073] At the same time, each unit module is integrated into a handheld device, which is easy to operate and can be used to inspect the bogie directly on site without disassembling the bogie or transporting it to other locations, thereby improving work efficiency.

[0074] Furthermore, the bogie-oriented dimension measurement handheld device also includes a network expansion interface for accessing a vehicle depot management handheld device.

[0075] Furthermore, the data acquisition module includes a built-in mask layer and a line layer for assisting shooting.

[0076] The above is a detailed introduction to a bogie-oriented dimension measurement method and a handheld device provided in an embodiment of the present invention. The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same and similar parts between the various embodiments can be referenced to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description. It should be pointed out that for ordinary technicians in this technical field, without departing from the principle of the present invention, several improvements and modifications can be made to the present invention, and these improvements and modifications also fall within the scope of protection of the claims of the present invention.

[0077] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0078] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

Claims

1. A dimension measurement method for bogies, It is characterized in that include: S1: Perform preprocessing operations based on the two-dimensional image of the bogie; S2: preprocessing operation based on the three-dimensional image of the bogie acquired by point cloud; S3: performing feature extraction based on the two-dimensional image after the preprocessing operation to obtain a first set of comparison data; S4: performing feature extraction based on the three-dimensional image after the preprocessing operation to obtain a second set of comparison data; S5: Based on the first set of comparison data and the second set of comparison data, a weighted average is calculated to obtain final size data.

2. The dimension measurement method for a bogie according to claim 1, It is characterized in that In S1, the preprocessing operation includes grayscale and denoising.

3. The dimension measurement method for a bogie according to claim 1, It is characterized in that In S2, the preprocessing operation includes phase unwrapping and three-dimensional reconstruction.

4. The dimension measurement method for a bogie according to claim 1, It is characterized in that In S3, the first set of comparison data includes: bogie thickness a1, bogie diameter a2 and suspension clearance a3 between the bogie frame and the vehicle body.

5. The dimension measurement method for a bogie according to claim 1, It is characterized in that In S4, the second set of comparison data includes: bogie thickness b1, bogie diameter b2 and suspension clearance b3 between the bogie frame and the vehicle body.

6. The dimension measurement method for a bogie according to claim 1, It is characterized in that In S5, the specific calculation expression is: D_i=(w_ai)*ai+(w_bi)*bi, wherein D_i represents the weighted dimension value of the i-th feature dimension, w_ai represents the weight parameter of feature ai, and w_bi represents the weight parameter of feature bi.

7. The dimension measurement method for a bogie according to claim 1, It is characterized in that Also includes the steps: S6: Generate a measurement report based on the final dimension data.

8. A handheld device for measuring dimensions of a bogie, It is characterized in that include: Data acquisition module, used to acquire two-dimensional images and three-dimensional images, including two acquisition modes: frontal shooting and oblique shooting; A data storage module, used for creating archives for the two-dimensional image and the three-dimensional image, storing the original image, feature data and final size data; A data analysis and reporting module, used to implement preprocessing, feature extraction and weighted average of the two-dimensional image and the three-dimensional image; The data acquisition module, data storage module and data analysis and reporting module are integrated into a handheld device.

9. The handheld device for measuring dimensions of a bogie according to claim 8, It is characterized in that The handheld device for measuring dimension of the bogie also includes a network expansion interface.

10. The handheld device for measuring dimensions of a bogie according to claim 8, It is characterized in that The data acquisition module includes a built-in mask layer and a line layer for assisting shooting.