An interpretable method for identifying the state of catenary droppers

Abstract

The present invention proposes an interpretable method for identifying the state of catenary suspension strings, which relates to the technical field of rail transit and includes: reading the images of the inspection of catenary suspension strings, sequentially reducing the images to form a multi-resolution image pyramid, performing convolution operations on each layer of images in the image pyramid, extracting the upper and lower endpoint coordinates of the suspension strings, establishing a first-order Bezier curve, performing an order elevation operation on the Bezier curve to obtain the control point coordinate values of the elevated Bezier curve, and optimizing until the average value of the Euclidean distance difference between the control point coordinate values before and after optimization is less than 1. Then, calculate the judgment index of the suspension string state based on the current control point coordinates, identify the suspension string state based on the judgment index, and output the identification result of the suspension string state; The present invention realizes the interpretable judgment of the slack and fracture fault points of catenary suspension strings, which has practical significance for becoming a key basic technology for catenary video inspection.

Description

Technical Field

[0001] The present invention relates to the technical field of rail transit, and in particular, to a method for identifying the state of catenary suspension strings with interpretability. Background Art

[0002] China's high-speed railway system adopts an electrification technology route, and locomotives use pantographs to obtain traction current from the catenary to drive vehicles. Since the catenary and pantograph systems are weak links in the entire rail transit system, they are vulnerable to the influence of complex environmental factors. Catenary faults will cause significant economic losses and safety risks to the railway transportation system, and their abnormal states are the focus of railway safety inspections. In the catenary system, suspension strings are key components that fix the contact wire to the carrier cable in the chain catenary system. Their functions are to increase the suspension points of the contact wire in the span, improve the sag and elasticity of the contact wire, adjust the height of the contact wire relative to the rail surface, and improve the working quality of the contact wire. However, due to the extensive and far-reaching high-speed railway lines in China, there are a large number of catenary suspension strings, with various passing and stationary states. Manual inspections have problems such as long inspection cycles, low efficiency, and easy omission of inspections, and it is difficult to meet the major application requirements of accurately and efficiently identifying the states of a large number of suspension strings. There is an urgent need for an intelligent inspection technology for catenary suspension string images that integrates operation and detection.

[0003] Early research on the service state monitoring technology of catenaries mainly focused on measuring the key structural geometric features of catenaries, such as the pull-out value and guide height of the contact wire, etc., to indirectly judge whether the measured catenary meets the design standards. However, these methods are difficult to detect early defects of catenaries in a timely manner, such as slight relaxation of suspension strings and hard bends of contact wires; nor can they obtain explanations for the causes of defects and fault points, such as identifying foreign objects on the contact wire. As an outdoor open device, the catenary suspension string wire faces various unpredictable risks. The structural changes caused by its early defects are not significant, but are hidden in the local shape information of the components. The main faults of suspension strings include current burns, broken current-carrying rings, loose clamps, stress-free relaxation and detachment of suspension strings, and so on.

[0004] With the wide use of image sensors, many researchers have conducted a large number of studies on catenary fault recognition for catenary images in recent years. However, with the in-depth promotion of recognition technologies mainly based on deep learning, the following problems have gradually emerged and attracted the attention of researchers. The problems existing in the recognition technology mainly based on deep learning mainly include: 1. A large amount of labeled data is required, but the labeling of these data consumes a lot of manpower; 2. A considerable number of fault samples are needed, but in practice, these fault samples may be difficult to obtain; 3. The currently widely used rectangular box annotation or pixel annotation is not suitable for catenary applications. Taking the rectangular box as an example, in the image area annotated by the rectangular box, in addition to the suspension clamp itself, it also includes a large part of the background and other electrical equipment. The area ratio of the suspension clamp pixels in the rectangular box is usually less than 5%, while pixel annotation has the problems of overly complex annotation and heavy annotation workload.

[0005] Based on this, the present application proposes an interpretable method for identifying the state of catenary suspension clamps to solve the above problems. Summary of the Invention

[0006] The purpose of the present invention is to provide an interpretable method for identifying the state of catenary suspension clamps, which can realize interpretable judgment of the relaxation and fracture fault points of catenary suspension clamp wires.

[0007] The technical solution of the present invention is as follows:

[0008] In a first aspect, the present application provides an interpretable method for identifying the state of catenary suspension clamps, which includes the following steps:

[0009] S1. Read the inspection image of the catenary suspension clamp wire, and sequentially reduce the image to form a multi-resolution image pyramid;

[0010] S2. Perform convolution operations on each layer of the image in the image pyramid to obtain convolution results;

[0011] S3. Extract the convolution results to obtain the upper and lower endpoint coordinates of the suspension clamp wire;

[0012] S4. Establish a first-order Bezier curve according to the upper and lower endpoint coordinates of the suspension clamp wire;

[0013] S5. Perform an order elevation operation on the first-order Bezier curve to obtain the control point coordinate values of the elevated Bezier curve;

[0014] S6. Optimize the control point coordinate values of the Bezier curve, and judge whether the average value of the Euclidean distance difference between the control point coordinate values before and after optimization is less than 1. If so, enter step S7; if not, return to step S5;

[0015] S7. Calculate the judgment index of the suspension string state according to the current control point coordinates;

[0016] S8. Identify the suspension string state based on the judgment index and output the identification result of the suspension string state.

[0017] Furthermore, in step S2, the process of the above convolution operation includes:

[0018]

[0019]

[0020]

[0021] Among them, H(x) represents the Hessian matrix, H xx , H xy and H yy represent the second-order derivative of the image gray level at coordinate x, x represents the pixel coordinate, Λ() represents the evaluation function of the ridge line feature, and λ1 and λ2 are the two eigenvalues of the Hessian matrix.

[0022] Furthermore, in step S5, the process of the above up-degree operation on the Bessel curve includes:

[0023]

[0024]

[0025]

[0026] Among them, s is the coordinate parameter, B represents the sampling coordinate, p represents the coordinate of the control point, n represents the order of the suspension string parameter curve, i represents the serial number of the current control point among all control points, E(p () ) represents the energy function, represents the internal energy function, E ext (p () ) represents the external energy function, E con (p () ) represents the endpoint constraint.

[0027] Furthermore, in step S7, the calculation formula for calculating the judgment index of the suspension string state according to the current control point coordinates includes:

[0028]

[0029] Among them, f loose () represents the discrimination index, p represents the coordinate of the control point, p i represents the coordinate of the i-th control point, n represents the order of the suspension string parameter curve, τ represents the empirical threshold parameter, It represents the ratio between the straight-line distance between the starting point and the ending point of the catenary parameter curve and the total length of the broken line of the control points of the catenary parameter curve.

[0030] Further, in step S8, the formula for identifying the catenary state based on the judgment index includes:

[0031]

[0032] Among them, f broken () represents the catenary state recognition function, I represents the pixel information of the observed image point, B represents the sampling coordinate, s is a parameter, p represents the coordinate of the control point, sup represents the supremum feature of the gray value, inf represents the infimum feature of the gray value, κ represents the empirical threshold parameter, f loose () represents the discrimination index.

[0033] In a second aspect, the present application provides an electronic device, which is characterized by including:

[0034] A memory for storing one or more programs;

[0035] A processor;

[0036] When the above one or more programs are executed by the above processor, an interpretable catenary state recognition method as described in any one of the first aspects above is implemented.

[0037] In a third aspect, the present application provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, an interpretable catenary state recognition method as described in any one of the first aspects above is implemented.

[0038] Compared with the prior art, the present invention has at least the following advantages or beneficial effects:

[0039] (1) The present invention proposes an interpretable catenary state recognition method, which can perform catenary modeling on a Bezier curve with adaptive order, and then further use the catenary line shape information to construct an identification mechanism for catenary line relaxation and breakage faults, which is particularly effective for analyzing the catenary curve shape observed during the driving process and can be applied to the intelligent inspection engineering tasks of catenary facilities;

[0040] (2) The present invention constructs a catenary wire model of a Bessel curve with adaptable order, trains the parameters of the catenary wire model through image gray information to minimize energy, obtains an accurate parametric expression of the catenary wire, and then uses it as the basis to input to the catenary wire relaxation discrimination algorithm and fracture discrimination algorithm to identify the state of the catenary wire. During the modeling process and the recognition process, no manually labeled catenary wire data is required. The process from on-vehicle observation images to catenary wire modeling is fully automated without human intervention, saving a large amount of manpower. The results of model generation can be adjusted and confirmed by the inspection staff of the railway bureau through human-computer interaction, improving the reliability of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the embodiments. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0042] Figure 1 It is a step diagram of a method for identifying the state of a catenary with interpretability according to the present invention;

[0043] Figure 2 It is a schematic structural block diagram of an electronic device according to the present invention.

[0044] Reference numerals: 101, memory; 102, processor; 103, communication interface. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0045] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. Generally, the components of the embodiments of the present application described and illustrated in the drawings here can be arranged and designed in various different configurations.

[0046] Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the present application that is claimed, but merely represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0047] It should be noted that: similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0048] It should be noted that in this text, the term "including" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or device including a series of elements not only includes those elements but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, article, or device. Without further limitation, elements defined by the statement "including..." do not preclude the existence of additional identical elements in the process, method, article, or device including the said elements.

[0049] In the description of the present application, it should also be noted that unless otherwise clearly specified and defined, the terms "arranged" and "connected" should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral connection; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the communication inside two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances.

[0050] The following will describe in detail some embodiments of the present application with reference to the accompanying drawings. Without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0051] Embodiment 1

[0052] Please refer to Figure 1 , Figure 1 which shows a step diagram of a method for identifying the state of catenary suspension strings with interpretability provided by an embodiment of the present application.

[0053] The present application provides a method for identifying the state of catenary suspension strings with interpretability, which includes the following steps:

[0054] S1. Read the images of the inspection of the catenary suspension string wire, and sequentially reduce the images to form a multi-resolution image pyramid;

[0055] S2. Perform convolution operations on each layer of images in the image pyramid to obtain convolution results;

[0056] S3. Extract the convolution results to obtain the upper and lower endpoint coordinates of the suspension string wire;

[0057] S4. Establish a first-order Bézier curve based on the upper and lower endpoint coordinates of the suspension string wire;

[0058] S5. Perform an order-raising operation on the first-order Bézier curve to obtain the control point coordinate values of the order-raised Bézier curve;

[0059] S6. Optimize the coordinate values of the control points of the Bessel curve, and determine whether the average value of the Euclidean distance difference between the coordinate values of the control points before and after optimization is less than 1. If so, proceed to step S7; if not, return to step S5;

[0060] S7. Calculate the judgment index of the suspension string state according to the current control point coordinates;

[0061] S8. Identify the suspension string state based on the judgment index and output the identification result of the suspension string state.

[0062] As a preferred implementation manner, in step S2, the process of the convolution operation includes:

[0063]

[0064]

[0065]

[0066] Among them, H(x) represents the Hessian matrix, H xx , H xy and H yy represent the second-order derivative of the image gray level at the coordinate x, x represents the pixel coordinate, Λ(x) represents the evaluation function of the ridge feature, and λ1 and λ2 are the two eigenvalues of the Hessian matrix.

[0067] As a preferred implementation manner, in step S5, the process of the order elevation operation of the Bessel curve includes:

[0068]

[0069]

[0070]

[0071] Among them, s is the coordinate parameter, B represents the sampling coordinate, p represents the coordinate of the control point, n represents the order of the suspension string parameter curve, i represents the serial number of the current control point among all control points, E(p (n+1) ) represents the energy function, represents the internal energy function, E ext (p (n+1) ) represents the external energy function, E con (p (n+1) ) represents the endpoint constraint.

[0072] As a preferred implementation manner, in step S7, the calculation formula for calculating the judgment index of the suspension string state according to the current control point coordinates includes:

[0073]

[0074] Among them, f loose (p) represents a discrimination index, p represents the coordinates of the control point, p i represents the coordinates of the i-th control point, n represents the order of the suspension string parameter curve, τ represents the empirical threshold parameter, represents the ratio between the straight-line distance between the starting point and the ending point of the suspension string parameter curve and the total length of the control point broken line of the suspension string parameter curve.

[0075] As a preferred implementation manner, in step S8, the formula for identifying the suspension string state based on the judgment index includes:

[0076]

[0077] Among them, f broken () represents the suspension string state recognition function, I represents the pixel information of the observed image point, B represents the sampling coordinate, s is a parameter, p represents the coordinates of the control point, sup represents the supremum feature of the gray value, inf represents the infimum feature of the gray value, κ represents the empirical threshold parameter, f loose (p) represents the discrimination index.

[0078] Embodiment 2

[0079] Please refer to Figure 2 , Figure 2 which is a schematic structural block diagram of an electronic device provided by an embodiment of the present application.

[0080] An electronic device includes a memory 101, a processor 102, and a communication interface 103. The memory 101, the processor 102, and the communication interface 103 are directly or indirectly electrically connected to each other to realize data transmission or interaction. For example, these components can be electrically connected to each other through one or more communication buses or signal lines. The memory 101 can be used to store software programs and modules. The processor 102 executes various functional applications and data processing by executing the software programs and modules stored in the memory 101. The communication interface 103 can be used to communicate with other node devices for signaling or data.

[0081] Among them, the memory 101 can be, but is not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electric Erasable Programmable Read-Only Memory (EEPROM), etc.

[0082] The processor 102 can be an integrated circuit chip with signal processing capabilities. The processor 102 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0083] It can be understood that the structure shown in the figure is only schematic. An interpretable catenary suspension state recognition method may also include more or fewer components than those shown in the figure, or have a different configuration from that shown in the figure. Each component shown in the figure can be implemented using hardware, software, or a combination thereof.

[0084] In the embodiments provided in this application, it should be understood that the disclosed method can also be implemented in other ways. The embodiments described above are merely illustrative. For example, the flowcharts or block diagrams in the accompanying drawings show the possible architectures, functions, and operations of the methods and computer program products according to multiple embodiments of this application. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that marked in the accompanying drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, as well as the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0085] In addition, each functional module in various embodiments of this application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0086] If the above functions are implemented in the form of software functional modules 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 this application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0087] In summary, a method for identifying the state of catenary suspension strings with interpretability provided by the embodiments of the present application reads the images obtained from the inspection of catenary suspension strings, sequentially shrinks the images to form a multi-resolution image pyramid, performs convolution operations on each layer of the images in the image pyramid, extracts the coordinates of the upper and lower endpoints of the suspension strings, establishes a first-order Bézier curve, performs an order elevation operation on the Bézier curve to obtain the coordinate values of the control points of the elevated Bézier curve, and optimizes them until the average value of the Euclidean distance differences between the coordinate values of the control points before and after optimization is less than 1. Then, the judgment index of the suspension string state is calculated based on the current control point coordinates, the suspension string state is identified based on the judgment index, and the suspension string state identification result is output, thereby realizing the interpretable judgment of the slack and breakage fault points of the catenary suspension strings. No manually annotated suspension string data is required in the modeling process and the identification process. The process from on-vehicle observation images to the modeling of suspension strings is fully automated without human intervention, saving a large amount of manpower. The model generation results can be adjusted and confirmed by the inspection staff of the railway bureau through human-computer interaction, improving the reliability of the model.

[0088] The foregoing are only preferred embodiments of the present application and are not intended to limit the present application. For those skilled in the art, various changes and modifications can be made to the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included within the protection scope of the present application.

[0089] For those skilled in the art, it is obvious that the present application is not limited to the details of the above-described exemplary embodiments, and can be implemented in other specific forms without departing from the spirit or basic characteristics of the present application. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present application is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present application. Any reference signs in the claims should not be regarded as limiting the claimed rights.

Claims

1. An interpretable method for identifying the state of catenary droppers, characterized in that Including the following steps: S1. Read the images of the inspection of the catenary suspension wire, and sequentially reduce the images to form a multi-resolution image pyramid; S2. Perform convolution operations on each layer of the image pyramid to obtain convolution results; S3. Extract the convolution results to obtain the upper and lower endpoint coordinates of the suspension wire; S4. Establish a first-order Bézier curve based on the upper and lower endpoint coordinates of the suspension wire; S5. Perform an order elevation operation on the first-order Bézier curve to obtain the control point coordinate values of the elevated Bézier curve; S6. Optimize the control point coordinate values of the Bézier curve, and determine whether the average value of the Euclidean distance difference between the control point coordinate values before and after optimization is less than 1. If so, go to step S7; if not, return to step S5; S7. Calculate the judgment index of the suspension wire state based on the current control point coordinates; S8. Identify the suspension wire state based on the judgment index and output the recognition result of the suspension wire state.

2. The method for identifying the state of catenary suspension strings with interpretability as claimed in claim 1, wherein, In step S2, the process of the convolution operation includes: Among them, H(x) represents the Hessian matrix, H xx , H xy and H yy represent the second-order derivative of the image grayscale at coordinate x, x represents the pixel coordinate, Λ(x) represents the evaluation function of the ridge feature, and λ1 and λ2 are the two eigenvalues of the Hessian matrix.

3. The interpretable catenary suspension state recognition method according to claim 1, characterized in that In step S5, the process of performing the order elevation operation on the Bézier curve includes: Among them, s is the coordinate parameter, B represents the sampling coordinate, p represents the coordinate of the control point, n represents the order of the catenary parameter curve, i represents the serial number of the current control point among all control points, E(p (n+1) ) represents the energy function, represents the internal energy function, E ext (p (n+1) ) represents the external energy function, E con (p (1) ) represents the endpoint constraint.

4. The interpretable catenary suspension state recognition method according to claim 1, wherein In step S7, the calculation formula for calculating the judgment index of the suspension wire state according to the current control point coordinates includes: Among them, f loose () represents the discrimination index, p represents the coordinates of the control point, p i represents the coordinates of the i-th control point, n represents the order of the suspension parameter curve, τ represents the empirical threshold parameter, represents the ratio between the straight-line distance between the starting point and the ending point of the suspension parameter curve and the total length of the control point broken line of the suspension parameter curve.

5. The method for identifying the state of catenary suspension strings with interpretability according to claim 1, characterized in that, In step S8, the formula for identifying the suspension wire state based on the judgment index includes: Among them, f broken () represents the catenary state recognition function, I represents the pixel information of the observed image points, B represents the sampling coordinates, s is a parameter, p represents the coordinates of the control points, sup represents the supremum feature of the gray value, inf represents the infimum feature of the gray value, κ represents the empirical threshold parameter, f loose () represents the discrimination index.

6. An electronic device, characterized in that, includes: A memory for storing one or more programs; A processor; When the one or more programs are executed by the processor, an interpretable catenary suspension wire state recognition method as described in any one of claims 1-5 is implemented.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, an interpretable catenary suspension wire state recognition method as described in any one of claims 1-5 is implemented.

Images