Track line preview curvature measurement method, device and medium

Abstract

The present invention relates to a method, device, and medium for measuring the preview curvature of a track line. The method comprises the following steps: acquiring a track line image, generating an initial lateral offset in the image after the track line bends, and correcting the initial lateral offset to obtain the actual lateral offset; calculating the detected preview curvature of the track based on the actual lateral offset, and correcting it using a preview hysteresis to obtain the actual preview curvature. Compared with the prior art, the present invention uses preview hysteresis to correct the detected preview curvature, achieving a more accurate measurement of the actual preview curvature.

Description

Technical Field

[0001] The present invention belongs to the technical field of track line curvature measurement, and relates to a track line preview curvature measurement method, equipment and medium. Background Art

[0002] Real-time tracking of track curvature provides a key input parameter for many active rail vehicle control technologies. Previewing the curvature by visually capturing an image of the track ahead is one effective way to address signal lag.

[0003] A search of Chinese publication number CN112810664A revealed a method for detecting track curvature based on vehicle posture sensing. This method uses sensors to detect vehicle posture and calculates the track curvature based on the relationship between the posture and track geometry. However, this method requires a large number of sensors and cannot detect the curvature of the track ahead in advance.

[0004] A search of Chinese publication number CN110929662A revealed a method for detecting track curvature based on visual images. This method processes the track image and calculates the track curvature ahead of the vehicle through geometric calculations, but does not provide a method for correcting the spatial position. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for measuring the preview curvature of a track line, so as to overcome the problem that the detected preview curvature is not accurate enough due to the lack of spatial position correction.

[0006] The purpose of the present invention can be achieved by the following technical solutions:

[0007] The present invention provides a method for measuring the curvature of a track line preview, comprising the following steps:

[0008] Acquire the track line image, and the initial lateral offset generated by the track line bending in the image is corrected to obtain the actual lateral offset;

[0009] Based on the actual lateral offset, the detection preview curvature of the line is calculated, and the preview hysteresis is used to correct it to obtain the actual preview curvature. The calculation formula of the preview hysteresis is:

[0010]

[0011] Among them, l m is the preview lag, l AB is the vehicle distance, c l is the preview distance l BC The ratio of the vehicle's fixed distance.

[0012] Furthermore, the detection preview curvature is obtained by calculation, and the calculation formula is:

[0013]

[0014] Among them, k is the detection preview curvature, l BC is the preview distance, l AB is the vehicle distance, l CC′ It is the lateral offset caused by the bending of the track centerline.

[0015] Furthermore, the actual preview curvature is obtained by a correction formula, which is:

[0016] k * (s) = k(s + l m )

[0017] Where s is the driving distance, l m is the preview lag.

[0018] Furthermore, the correction comprises the following steps:

[0019] Obtaining the initial lateral offset generated in the image by a visual image method;

[0020] The initial lateral offset is corrected by trapezoidal correction to obtain a corrected lateral offset;

[0021] The corrected lateral offset is multiplied by the calibrated determination coefficient to obtain the actual lateral offset.

[0022] Furthermore, the visual image method is to capture an image of the track centerline after bending at the front preview point through a camera and pre-process the image.

[0023] Furthermore, the preprocessing includes grayscale processing and edge recognition.

[0024] Furthermore, the trapezoidal correction is used to make the normal direction of the image coincide with the shooting direction of the camera.

[0025] Furthermore, the determination coefficient is obtained through a mapping relationship between the actual physical object size and the image pixels.

[0026] The present invention provides an electronic device, comprising a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the method when executing the program.

[0027] The present invention provides a computer-readable storage medium having a computer program stored thereon, and the method described above is implemented when the program is executed by a processor.

[0028] Compared with the prior art, the present invention has the following characteristics:

[0029] 1. The present invention obtains the actual lateral offset through correction, and uses the preview hysteresis to correct the obtained detection preview curvature, thereby achieving a more accurate measurement of the actual preview curvature.

[0030] 2. The present invention predicts the curvature of the track line through visual images, overcoming the time lag problem caused by the traditional curvature detection method based on vehicle posture sensors.

[0031] 3. The present invention performs trapezoidal correction on the image, which solves the problem that the camera illumination direction does not coincide with the normal direction of the actual imaging area, and further improves the measurement accuracy of the actual preview curvature. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a schematic diagram of the physical model of the present invention;

[0033] Figure 2 Schematic diagram of trapezoidal correction in the physical model of the present invention;

[0034] Figure 3 This is a schematic diagram comparing the actual preview curvature of a certain line obtained by the present invention with the actual preview curvature of the line;

[0035] Figure 4 Schematic diagram of the flow of the track line preview curvature measurement method in Example 1. DETAILED DESCRIPTION

[0036] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0037] Example 1

[0038] The present invention proposes a method for measuring the curvature of a track line. Figure 4 , including the following steps:

[0039] S1. Acquire an image. After the track is bent, an initial lateral offset is generated in the image. The actual lateral offset is obtained through correction.

[0040] S2. Calculate the detection preview curvature of the line based on the actual lateral offset;

[0041] S3. Based on the detected preview curvature, the actual preview curvature is obtained by correcting the preview hysteresis. The calculation formula of the preview hysteresis is:

[0042]

[0043] Among them, l m is the preview lag, lAB is the vehicle distance, c l is the preview distance l BC The ratio of the vehicle's fixed distance.

[0044] Specifically, the detection preview curvature is obtained by calculation, and the calculation formula is:

[0045]

[0046] Among them, k is the detection preview curvature, l BC is the preview distance, l AB is the vehicle distance, l CC′ It is the lateral offset in the image caused by the bending of the track centerline.

[0047] The present invention utilizes the preview hysteresis to correct the detected preview curvature, thereby achieving more accurate measurement of the actual preview curvature.

[0048] Specifically, the actual preview curvature is obtained by a correction formula, which is:

[0049] k * (s) = k(s + l m )

[0050] Where s is the driving distance, l m is the preview lag.

[0051] Specifically, the correction includes the following steps:

[0052] Obtaining the initial lateral offset generated in the image by a visual image method;

[0053] The initial lateral offset is corrected by trapezoidal correction to obtain a corrected lateral offset;

[0054] The corrected lateral offset is multiplied by the calibrated determination coefficient to obtain the actual lateral offset.

[0055] Specifically, the visual image method is to capture an image of the track centerline after bending at the front preview point through a camera and pre-process the image.

[0056] Specifically, the preprocessing includes grayscale processing and edge recognition.

[0057] The present invention predicts the curvature of the track line through visual images, overcoming the time lag problem caused by the traditional curvature detection method based on vehicle posture sensors.

[0058] Specifically, the trapezoidal correction is used to make the normal direction of the image coincide with the shooting direction of the camera.

[0059] The present invention performs trapezoidal correction on the image, solves the problem that the camera illumination direction does not coincide with the normal direction of the actual imaging area, and further improves the measurement accuracy of the actual preview curvature.

[0060] Specifically, the determination coefficient is obtained through a mapping relationship between the actual physical object size and the image pixels.

[0061] In a specific embodiment, see Figure 1 , is a schematic diagram of the physical model of the present invention, including:

[0062] In this model, the front and rear bogies of the rail vehicle are abstracted as points B and A respectively. The on-board camera captures the track line image at a certain angle on the top of the vehicle body. Point N is the lower edge point in the image, point C is the midpoint of the image, point F is the upper edge point of the image, and point C* is the actual position point on the line corresponding to the actual preview curvature after spatial correction.

[0063] See Figure 2 , perform trapezoidal correction on the image. After correction, the normal direction of the image coincides with the shooting direction. At this time, it can be equivalent to the camera shooting directly above point C; further, a plane rectangular coordinate system is established with point B (the center of the front bogie) as the origin, and a top view of the physical model is given. In the figure, point C' is the point on the track centerline with the same longitudinal coordinate as the midpoint C of the camera's field of view.

[0064] Based on the lateral offset, the analytical equation of the track curvature under ideal conditions is given by geometric deduction, where k is the curvature, l BC is the preview distance, l AB is the vehicle distance, l CC′ It is the lateral offset caused by the bending of the track centerline, specifically:

[0065]

[0066] The lateral offset is calculated after camera calibration.

[0067] The spatial position correction formula of the actual preview curvature is given, where l m is the preview lag, c l It is the ratio of preview distance to vehicle distance, specifically:

[0068]

[0069] Figure 3 The following figure compares the actual curvature of a line obtained using this method with the actual preview curvature of the line. Line 1 is the actual curvature of the line, and line 2 is the actual preview curvature. As can be seen from the figure, the actual preview curvature is ahead of the actual curvature in space. The amount of advance is the difference between the preview distance and the spatial position correction, specifically:

[0070] l pre =l BC -l m

[0071] In this embodiment: the vehicle distance l AB =15.7m, preview distance l BC = 50m, so the actual preview curvature is the spatial advance of the actual curvature

[0072] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0073] Example 2

[0074] The present invention also provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the program, all the methods described in Example 1 are implemented.

[0075] The above description of the embodiments is intended to facilitate understanding and use of the invention by those skilled in the art. It will be apparent that those skilled in the art can readily make various modifications to these embodiments and apply the general principles described herein to other embodiments without requiring inventive effort. Therefore, the present invention is not limited to the above-described embodiments. Improvements and modifications made by those skilled in the art based on the disclosure of the present invention, without departing from the scope of the present invention, should be within the scope of protection of the present invention.

Claims

1. A method for measuring the curvature of a track preview, characterized in that: The following steps are involved: Acquire the track line image, and the initial lateral offset generated by the track line bending in the image is corrected to obtain the actual lateral offset; Based on the actual lateral offset, the detection preview curvature of the line is calculated, and the preview hysteresis is used to correct it to obtain the actual preview curvature. The calculation formula of the preview hysteresis is: in, is the preview lag, Determine the distance between vehicles. Preview distance The ratio of the vehicle distance to the vehicle The detection preview curvature is obtained by calculation, and the calculation formula is: in, To detect the preview curvature, is the preview distance, Determine the distance between vehicles. is the lateral offset caused by the bending of the track centerline, The actual preview curvature is obtained by a correction formula, which is: Where s is the driving distance, is the preview lag.

2. A method for measuring the curvature of a track preview according to claim 1, characterized in that: The correction comprises the following steps: Obtaining the initial lateral offset generated in the image by a visual image method; The initial lateral offset is corrected by trapezoidal correction to obtain a corrected lateral offset; The corrected lateral offset is multiplied by the calibrated determination coefficient to obtain the actual lateral offset.

3. A method for measuring the curvature of a track preview according to claim 2, characterized in that: The visual image method is to capture an image of the track centerline after bending at the front preview point through a camera and pre-process the image.

4. A method for measuring the curvature of a track preview according to claim 3, characterized in that: The preprocessing includes grayscale processing and edge recognition.

5. The method for measuring the curvature of a track preview according to claim 2, wherein: The trapezoidal correction is used to make the normal direction of the image coincide with the shooting direction of the camera.

6. The method for measuring the curvature of a track preview according to claim 2, wherein: The determination coefficient is obtained through the mapping relationship between the actual physical object size and the image pixels.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 6 is implemented.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program for executing the method according to any one of claims 1 to 6.

Images