Car coupler centering state video detection method, device and system and storage medium

Abstract

The invention discloses a car coupler centering state video detection method, device and system and a storage medium, and the method comprises the steps: S1, obtaining double-mark angular point coordinate values of different deflection angles of a car coupler offline, and fitting a trajectory curve through a polynomial; and S2, collecting a coupler image on line, extracting double angular points of the coupler through an angular point and straight line detection algorithm, and calculating the centering state of the coupler. By the adoption of the technical scheme, the method has the advantages of being high in detection real-time performance and capable of effectively improving the marshalling efficiency and operation safety of the train.

Description

Technical Field

[0001] The present invention belongs to the technical field of rail transit intelligent detection, and in particular relates to a method and device, a system, and a storage medium for video detection of a coupler centering state. Background Art

[0002] The online train unmarshaling technology enables trains to complete coupling and unmarshaling tasks at any position on the main line, which can improve the flexibility and operability of train marshaling operations and save manpower, material resources and time costs; this technology can enable flexible marshaling and mixed operation of lines, improve the efficiency of urban rail transit operations, enhance residents' travel experience, and save energy and reduce emissions. In the application theory of online train unmarshaling technology, the core requirement is to detect the status of the couplers between trains anytime and anywhere, determine the real-time status of the couplers, and then determine whether the train has the conditions for coupling and unmarshaling based on the coupler status.

[0003] At present, the operation of coupler centering is mainly carried out through manual assistance on the ground, which requires the driver to get off the vehicle to check the status of the coupler. The manually assisted centering method has low automation, low efficiency, high cost, and high site requirements, and it is difficult to meet the growing flexible marshaling needs of urban rail transit. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a method and device, system and storage medium for video detection of coupler centering status, which has the characteristics of high real-time detection and can effectively improve the train formation efficiency and operation safety.

[0005] To achieve the above object, the present invention adopts the following technical solution:

[0006] A method for video detection of a coupler centering state, comprising:

[0007] S1, offline acquisition of the coordinate values ​​of the double-marked corner points of the coupler at different deflection angles, and obtaining the trajectory curve of the double-marked corner points by polynomial fitting;

[0008] S2. Extract the double corner points of the coupler based on the online collected coupler image; and calculate the coupler centering state based on the trajectory curves of the double corner points of the coupler and the double marked corner points.

[0009] Preferably, S1 comprises:

[0010] The video detection system collects m images of the coupler with different deflection angles offline and performs distortion correction;

[0011] Extract different deflection angles θ of the coupler on the corrected image i The coordinate values ​​of the double-marked corner points a and b

[0012] Calculate the average distance between the two labeled corners The formula is as follows:

[0013]

[0014] Substitute the coordinate values ​​of the double-marked corner points into the polynomial and calculate the fitting coefficients w0, w1, ..., w by the least squares method. n-1 ,w n , get the trajectory curve f of the double-marked corner point a (x) and f b (x), the fitting curve equation is as follows:

[0015] f(x)=w0+w1·x+w2·x 2 +w n-1 ·x n-1 +w n ·x n .

[0016] Preferably, in S2, the coupler image is collected online, and the double corner points of the coupler are extracted by using a corner point and line detection algorithm.

[0017] Preferably, S2 comprises:

[0018] Use the trajectory curve of the double-marked corner points to screen out the corner point set P' of corner points a and b a and P' b , the screening formula is as follows:

[0019]

[0020]

[0021] Among them, (x i,0 ,f a (x i,0 )),(x i,1 ,f b (x i,1 )) is the trajectory curve f a (x) and f b (x) and the corner point (x i ,y i ) is the coordinate value with the shortest distance, and d1 is the distance error threshold from the corner point to the trajectory curve;

[0022] The pixel points whose gradient amplitude of the corrected image exceeds the preset threshold are defined as anchor points. A pixel chain is constructed based on the anchor point connection and a straight line fitting is performed to generate a straight line segment set, and the line segment endpoint coordinate set L = [(x L,j ,y L,j ),(x R,j ,y R,j )], j = 1,…,w;

[0023] Using the line segment endpoint coordinate set L and the corner point set P' of corner points a and b a and P' b , locate the coordinates of the double-marked corner points a and b of the coupler (x a ,y a ) and (x b ,y b ), the screening formula is as follows:

[0024]

[0025] Among them, d2 is the distance error threshold between the endpoint and the corner point of the line segment set;

[0026] Based on the double-marked corner points of the coupler, the deflection angle θ of the coupler is calculated. Δc ,Right now

[0027]

[0028] The present invention also provides a video detection device for the centering state of a coupler, comprising:

[0029] The first processing module is used to obtain the coordinate values ​​of the double-marked corner points of the coupler at different deflection angles offline, and obtain the trajectory curve of the double-marked corner points through polynomial fitting;

[0030] The second processing module is used to extract the double corner points of the coupler based on the online collected coupler image; and calculate the coupler centering state based on the trajectory curves of the double corner points of the coupler and the double marked corner points.

[0031] Preferably, the second processing module is used to collect the coupler image online and extract the double corner points of the coupler by using a corner point and straight line detection algorithm.

[0032] An embodiment of the present invention also provides a video detection system for the centering status of a coupler, comprising: a memory and a processor, wherein the memory stores a computer program executed by the processor, and the computer program executes a video detection method for the centering status of a coupler when executed by the processor.

[0033] An embodiment of the present invention further provides a storage medium, on which a computer program is stored, and when the computer program is run, the method for video detection of the centering state of a coupler is executed.

[0034] Due to the limitations of stains and rust on the coupler, shooting light, environment and other conditions, the accurate identification of feature points is affected. The present invention uses straight line detection and corner point detection at the same time, combined with double-marker point trajectory curve matching technology, to achieve high-reliability train coupler alignment real-time detection based on video recognition technology. The present invention has a wide range of applications and can significantly improve the level of automation of flexible train marshaling and promote the development and application of online train coupling and uncoupling. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0036] Figure 1 It is a flow chart of the method for video inspection of the centering state of a coupler of the present invention;

[0037] Figure 2 This is a schematic diagram of the installation location of the video detection system. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0040] Embodiment 1:

[0041] like Figure 1 As shown, an embodiment of the present invention provides a method for video detection of a coupler centering state, comprising:

[0042] S1, offline acquisition of the coordinate values ​​of the double-marked corner points of the coupler at different deflection angles, and fitting the trajectory curve through a polynomial;

[0043] S2. Collect the coupler image online, extract the double corner points of the coupler through the corner point and line detection algorithm, and calculate the coupler alignment status.

[0044] As an implementation of the embodiment of the present invention, the step S1 specifically includes the following steps:

[0045] S11. The video detection system is installed on the front surface of the vehicle, located on the longitudinal plane of the center of the coupler. The shooting direction is parallel to the coupler in the centered state, and the shooting picture can fully capture the coupler;

[0046] S12, the video detection system collects m images of the coupler at different deflection angles offline and performs distortion correction;

[0047] S13, extracting different deflection angles θ of the coupler on the corrected imagei The coordinate values ​​of the double-marked corner points a and b

[0048] S14. Calculate the average distance between double-marked corner points The formula is as follows:

[0049]

[0050] Substitute the coordinate values ​​of the double-marked corner points into the polynomial and calculate the fitting coefficients w0, w1, ..., w by the least squares method. n-1 ,w n , get the trajectory curve f of the double-marked corner point a (x) and f b (x), the fitting curve equation is as follows:

[0051] f(x)=w0+w1·x+w2·x 2 +w n-1 ·x n-1 +w n ·x n

[0052] Specifically, in step S11, the video detection system is installed at a position such as Figure 2 As shown, the video detection system 1 is installed on the surface of the vehicle head 3, located on the longitudinal plane at the center of the coupler 2, and the shooting direction is parallel to the coupler 2 in the centered state. The system shooting picture can completely capture the coupler 2.

[0053] Specifically, in step S12, the video detection system collects m images of the coupler with different deflection angles offline and performs distortion correction, the radial distortion coefficients are -0.32, 0.11, and the tangential distortion is not considered. Specifically, in step S22, t = 0.04, R0 = 0.05*R max (x, y); d1 is the distance error threshold from the corner point to the trajectory curve, d1=5.

[0054] As an implementation of the embodiment of the present invention, the step S2 specifically includes the following steps:

[0055] S21, the video detection system collects the coupler image online and performs distortion correction;

[0056] S22, defining the pixel points whose gradient response values ​​R(x, y) of the corrected image exceed the preset threshold R0 as corner points, extracting the corner point set P, and the calculation formula is as follows:

[0057] P={(x i ,y i )|R(x i ,y i )>R0}

[0058] R(x,y)=G(x) 2 ·G(y) 2 -(G(x)G(y)) 2 -t·(G(x) 2 +G(y) 2 ) 2

[0059] Wherein, G(x) is the horizontal gradient amplitude of the pixel point (x, y) in the gradient map of the corrected image, G(y) is the vertical gradient amplitude of the pixel point (x, y) in the gradient map of the corrected image, and t is an empirical constant.

[0060] Use the trajectory curve of the double-marked corner points to screen out the corner point set P' of corner points a and b a and P' b , the screening formula is as follows:

[0061]

[0062]

[0063] Among them, (x i,0 ,f a (x i,0 )),(x i,1 ,f b (x i,1 )) is the trajectory curve f a (x) and f b (x) and the corner point (x i ,y i ) is the coordinate value with the shortest distance, and d1 is the distance error threshold from the corner point to the trajectory curve.

[0064] S23, defining the pixel points whose gradient amplitude of the corrected image exceeds the preset threshold as anchor points, constructing a pixel chain based on the anchor point connection and performing straight line fitting to generate a straight line segment set, and obtaining a line segment endpoint coordinate set L=[(x L,j ,y L,j ),(x R,j ,y R,j )], j = 1,…,w;

[0065] S24, using the line segment endpoint coordinate set L and the corner point set P' of the corner points a and b a and P' b , locate the coordinates of the double-marked corner points a and b of the coupler (x a ,y a ) and (x b ,y b ), the screening formula is as follows:

[0066]

[0067] Wherein, d2 is the distance error threshold between the endpoints and corner points of the line segment set, preferably, d2=5.

[0068] S25, based on the double-marked corner points of the coupler, calculate the deflection angle θ of the coupler Δc .

[0069]

[0070] Embodiment 2:

[0071] The embodiment of the present invention further provides a video detection device for a coupler centering state, comprising:

[0072] The first processing module is used to obtain the coordinate values ​​of the double-marked corner points of the coupler at different deflection angles offline, and obtain the trajectory curve of the double-marked corner points through polynomial fitting;

[0073] The second processing module is used to extract the double corner points of the coupler based on the online collected coupler image; and calculate the coupler centering state based on the trajectory curves of the double corner points of the coupler and the double marked corner points.

[0074] As an implementation of an embodiment of the present invention, the second processing module is used to collect the coupler image online and extract the double corner points of the coupler through a corner point and straight line detection algorithm.

[0075] Embodiment 3:

[0076] An embodiment of the present invention also provides a video detection system for the centering status of a coupler, comprising: a memory and a processor, wherein the memory stores a computer program executed by the processor, and the computer program executes a video detection method for the centering status of a coupler when executed by the processor.

[0077] Embodiment 4:

[0078] An embodiment of the present invention further provides a storage medium, on which a computer program is stored, and when the computer program is run, the method for video detection of the centering state of a coupler is executed.

[0079] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for detecting the centering status of a coupler by video, characterized in that: include: S1, offline acquisition of the coordinate values ​​of the double-marked corner points of the coupler at different deflection angles, and obtaining the trajectory curve of the double-marked corner points by polynomial fitting; S2. Extract the double corner points of the coupler based on the online collected coupler image; and calculate the coupler centering state based on the trajectory curves of the double corner points of the coupler and the double marked corner points.

2. The method for detecting the centering status of a coupler by video according to claim 1, characterized in that: S1 includes: The video detection system collects m images of the coupler with different deflection angles offline and performs distortion correction; Extract different deflection angles θ of the coupler on the corrected image i The coordinate values ​​of the double-marked corner points a and b Calculate the average distance between the two labeled corners The formula is as follows: Substitute the coordinate values ​​of the double-marked corner points into the polynomial and calculate the fitting coefficients w0, w1, ..., w by the least squares method. n-1 , w n , get the trajectory curve f of the double-marked corner point a (x) and f b (x), the fitting curve equation is as follows: f(x)=w0+w1·x+w2·x 2 +w n-1 ·x n-1 +w n ·x n 。 3. The method for detecting the centering state of a coupler by video according to claim 2, characterized in that: In S2, the coupler image is collected online, and the double corner points of the coupler are extracted through the corner point and line detection algorithm.

4. The method for detecting the centering state of a coupler by video according to claim 3, characterized in that S2 include: Use the trajectory curve of the double-marked corner points to screen out the corner point set P′ of corner points a and b a and P' b , the screening formula is as follows: Among them, (x i，0 , f a (x i，0 )),(x i，1 , f b (x i，1 )) is the trajectory curve f a (x) and f b (x) and the corner point (x i ,y i ) is the coordinate value with the shortest distance, and d1 is the distance error threshold from the corner point to the trajectory curve; The pixel points whose gradient amplitude of the corrected image exceeds the preset threshold are defined as anchor points. A pixel chain is constructed based on the anchor point connection and a straight line fitting is performed to generate a straight line segment set, and the line segment endpoint coordinate set L = [(x L，j ,y L，j ), (x R，j ,y R，j )], j = 1, ..., w; Using the line segment endpoint coordinate set L and the corner point set P' of corner points a and b a and P' b , locate the coordinates of the double-marked corner points a and b of the coupler (x a ,y a ) and (x b ,y b ), the screening formula is as follows: Among them, d2 is the distance error threshold between the endpoint and the corner point of the line segment set; Based on the double-marked corner points of the coupler, the deflection angle θ of the coupler is calculated. Δc ,Right now 5. A video detection device for the centering state of a coupler, characterized in that: include: The first processing module is used to obtain the coordinate values ​​of the double-marked corner points of the coupler at different deflection angles offline, and obtain the trajectory curve of the double-marked corner points through polynomial fitting; The second processing module is used to extract the double corner points of the coupler based on the online collected coupler image; and calculate the coupler centering state based on the trajectory curves of the double corner points of the coupler and the double marked corner points.

6. The video detection device for the centering state of a coupler according to claim 5, characterized in that: The second processing module is used to collect the coupler image online and extract the double corner points of the coupler through the corner point and line detection algorithm.

7. A video detection system for coupler centering status, characterized in that: include: A memory and a processor, wherein the memory stores a computer program executed by the processor, and when the computer program is executed by the processor, the method for video detection of the centering state of a coupler as claimed in any one of claims 1 to 3 is executed.

8. A storage medium, characterized in that: The storage medium stores a computer program, which, when running, executes the method for video detection of the coupler centering state as claimed in any one of claims 1 to 3.

Images