Dynamic detection device and method for catenary wire sag

Abstract

The application discloses a pantograph catenary dropper dynamic detection device and method, wherein the device comprises: a main control module for issuing an image collection command; a signal synchronization control module for obtaining equidistant pulse signals from a photoelectric encoder driven by a wheel and inputting the equidistant pulse signals into a high-definition linear array camera assembly; the high-definition linear array camera assembly is used for shooting a catenary image; an image collection module is used for providing the collected catenary image to an image processing module; the image processing module is used for: identifying the catenary image to obtain image data after identification; analyzing the image data to generate feature data of dropper data and support column data, and performing data fusion processing on the dropper data and the support column data and the processed feature data, and outputting the fused data. The application can improve the definition of the shot image, improve the accuracy of the dropper data and the support column data through image analysis, and thus improve the detection accuracy of the catenary.

Description

Technical Field

[0001] This invention relates to the field of railway catenary testing technology, and in particular to a device and method for dynamic testing of pantograph-catenary droppers. Background Technology

[0002] This section is intended to provide background or context for the embodiments of the invention set forth in the claims. The description herein is not an admission that it is prior art simply because it is included in this section.

[0003] In recent years, with the continuous development of laser technology, image acquisition technology, and image processing technology, the advantages of non-contact measurement technology in catenary inspection have begun to emerge, but some problems also exist. The single binocular measurement method is not intuitive enough. Although sampling according to the sampling interval can estimate the measurement results, omissions are inevitable during the inspection process. Moreover, the out-of-limit data of detection parameters characterizing the current collection quality of the catenary, such as pantograph-catenary contact force and hard points, cannot be verified on-site. Therefore, the analysis of such data needs to be assisted by machine vision. First, it can be used to check the status of catenary components, and second, it can be used to analyze the mechanical characteristics of the catenary structure. Currently, area array cameras or webcams are commonly used to record pantograph-catenary operation videos. However, this method has several shortcomings: First, the system relies on LED lighting, resulting in unclear images at night, which cannot adequately meet the needs of all-day operation. Second, the method of capturing images from the front in the direction of movement makes it difficult to synchronize the mileage of the captured images with measurement systems for catenary geometry and pantograph-catenary dynamic parameters. Adjacent images may have overlapping parts, interfering with data alignment and affecting the positioning of catenary droppers and catenary poles. Third, it has poor resistance to sunlight interference, and existing shooting methods cannot adjust the shooting angle online according to height, resulting in unclear images. Fourth, the catenary suspension device is arranged along the direction of track travel. Traditional pantograph-catenary video monitoring cameras shoot along the direction of movement, and due to the shooting angle, a large portion of the suspension device components are obscured, making it difficult to fully observe the operational status of the suspension device components and providing limited assistance in analyzing the causes of pantograph-catenary contact force, hard points, and other detection exceedances. The current high-speed railway power supply safety monitoring system (6C system) does not yet have a device capable of providing continuous and uninterrupted linear array imaging of the catenary suspension device in operational status.

[0004] Accurately locating structural feature points such as supports and droppers in the contact network inspection data waveform diagram is crucial for statistical analysis of inspection data and on-site location of out-of-limit positions. Current methods rely on regression analysis of inspection data to locate these structural feature points in the waveform diagram. Considering the zigzag arrangement of the contact wire along the line's direction of travel, regression analysis of the pull-out value detection waveform identifies specific inflection points as support locations. Similarly, considering the arrangement of several droppers between adjacent support suspension points (typically tens of meters) to improve the smoothness of the contact wire height, regression analysis of the contact wire height detection waveform identifies roof-like displacement abrupt changes as dropper locations. This data analysis-based support location method typically achieves an accuracy rate of 85%-95%. However, in dynamic contact network inspection, the high-speed sliding of the pantograph across the contact wire and the interaction between the pantograph and the contact wire disrupt the roof-like characteristic of the contact wire height formed by the droppers suspending the contact wire, resulting in lower dropper location accuracy and failing to meet on-site application requirements. Summary of the Invention

[0005] This invention provides a dynamic detection device for catenary droppers to improve the clarity of captured images. Through image analysis, it enhances the accuracy of dropper and support data, thereby improving the accuracy of catenary detection. The device includes:

[0006] The main control module, image acquisition module, high-definition line scan camera assembly, image processing module, and signal synchronization control module; among them,

[0007] The main control module is used to issue image acquisition commands when the high-speed inspection vehicle starts moving.

[0008] The signal synchronization control module is used to: obtain equally spaced pulse signals from the photoelectric encoder driven by the wheel after receiving the image acquisition command, and input the equally spaced pulse signals into the high-definition line scan camera component;

[0009] High-definition line scan camera assembly, used for: capturing images of the overhead contact line after receiving equally spaced pulse signals;

[0010] The image acquisition module is used to: acquire contact network images captured by the high-definition line scan camera assembly and provide them to the image processing module;

[0011] The image processing module is used to: identify the catenary images sent by the image acquisition module using deep learning methods to obtain identified image data, which includes dropper data and support data; analyze the image data, generate feature data for the dropper data and support data based on the analysis results, and preprocess the feature data to obtain processed feature data; perform data fusion processing on the dropper data, support data, and processed feature data to obtain fused catenary dropper data and support data, and output the fused catenary dropper data and support data in the form of graphics and statistical reports.

[0012] This invention also provides a method for dynamic detection of catenary droppers, used to improve the clarity of captured images and enhance the accuracy of dropper and support data through image analysis, thereby improving the accuracy of catenary detection. The method includes:

[0013] The command to acquire images is issued when the high-speed inspection vehicle begins to move;

[0014] According to the image acquisition command, equally spaced pulse signals are obtained from the photoelectric encoder driven by the wheel;

[0015] After obtaining the equally spaced pulse signals, images of the overhead contact line are captured and photographed.

[0016] Deep learning methods are used to identify the acquired images, resulting in identified image data, which includes dropper data and support data. The image data is then analyzed, and feature data for the dropper and support data is generated based on the analysis results. This feature data is then preprocessed to obtain processed feature data. The processed feature data, dropper data, and support data are then fused to obtain fused catenary dropper and support data. Finally, the fused catenary dropper and support data are output in the form of graphics, text, and statistical reports.

[0017] This invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-described dynamic detection method for bow and catenary suspension wires.

[0018] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described dynamic detection method for bow wire and catenary.

[0019] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-described dynamic detection method for bow wire and catenary.

[0020] In this embodiment of the invention, an image acquisition command is issued when the high-speed inspection vehicle begins to move; according to the image acquisition command, equally spaced pulse signals are obtained from the photoelectric encoder driven by the wheels; after obtaining the equally spaced pulse signals, images of the overhead contact line are acquired; the acquired images are identified using deep learning methods to obtain identified image data, which includes dropper data and support data; the image data is analyzed, and feature data of the dropper data and support data is generated based on the analysis results; the feature data is preprocessed to obtain processed feature data; the processed feature data, dropper data, and support data are fused to obtain fused overhead contact line dropper data and support data; the fused overhead contact line dropper data and support data are output in the form of graphics and statistical reports. This improves the clarity of the captured images and enhances the accuracy of the dropper data and support data through image analysis, thereby improving the accuracy of overhead contact line inspection. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0022] Figure 1 This is a schematic diagram of the bow wire dynamic detection device in an embodiment of the present invention;

[0023] Figure 2 This is a schematic diagram of a specific bow and catenary suspension string dynamic detection device in an embodiment of the present invention;

[0024] Figure 3 This is a schematic diagram of the laser ranging principle in an embodiment of the present invention;

[0025] Figure 4 This is a physical image of the bow wire and catenary dynamic detection device in an embodiment of the present invention;

[0026] Figure 5 This is a result diagram of the dynamic detection device for the bow wire and catenary in an embodiment of the present invention;

[0027] Figure 6 This is a waveform diagram of overhead contact line detection obtained using machine vision technology in an embodiment of the present invention. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0029] Figure 1 This is a schematic diagram of the bow and catenary suspension string dynamic detection device in an embodiment of the present invention. The device includes:

[0030] Main control module 01, image acquisition module 04, high-definition line scan camera assembly 03, image processing module 05, signal synchronization control module 02; among which,

[0031] Main control module 01 is used to issue an image acquisition command when the high-speed inspection vehicle starts moving.

[0032] The signal synchronization control module 02 is used to: obtain equidistant pulse signals from the photoelectric encoder driven by the wheel after receiving the image acquisition command, and input the equidistant pulse signals into the high-definition line scan camera component;

[0033] High-definition line scan camera assembly 03 is used to: capture images of the contact wire after receiving equally spaced pulse signals;

[0034] Image acquisition module 04 is used to: acquire contact wire images captured by the high-definition line scan camera assembly and provide them to the image processing module;

[0035] Image processing module 05 is used to: identify the catenary image sent by the image acquisition module using deep learning methods to obtain identified image data, which includes dropper data and support data; analyze the image data, generate feature data of dropper data and support data based on the analysis results, and preprocess the feature data to obtain processed feature data; perform data fusion processing on dropper data, support data, and processed feature data to obtain fused catenary dropper data and support data, and output the fused catenary dropper data and support data in the form of graphics and statistical reports.

[0036] In a specific embodiment, before starting the overhead contact line inspection, the device of the present invention is powered on to complete a self-test and ensure that the hardware is in normal working condition. The host computer acquisition software is opened. Upon first login to the system device, parameters such as the acquisition card and camera should be configured, and information on the section of the line to be inspected should be entered (no further configuration is required upon subsequent logins if the parameters have not changed). The acquisition task is then opened, and the high-speed inspection vehicle is waited for to proceed.

[0037] In a specific embodiment, the high-speed inspection vehicle starts to move, and the main control module issues an image acquisition command. The rotation of the wheels drives the photoelectric encoder to continuously output equally spaced pulse signals to the signal synchronization control module. The two identical pulse signals processed by the signal synchronization control module are respectively input to two high-definition line scan camera components, triggering the cameras to acquire images of the contact network equipment and transmit them to the image acquisition module. The acquired image data is stored on the local hard disk. At the same time as data acquisition, the image processing module identifies and analyzes the acquired contact network image data online.

[0038] like Figure 2 As shown, in one embodiment, a laser ranging module is also included, for:

[0039] Measure the distance between the vehicle roof and the pantograph, and determine the height of the contact wire based on the distance between the vehicle roof and the pantograph;

[0040] Adjust the angle of the high-definition line scan camera assembly according to the height of the contact line.

[0041] like Figure 3 As shown in the specific embodiment, the laser ranging module measures the height of the contact wire in real time. If the contact wire is not within the detection range, the pan-tilt unit automatically adjusts the camera angle. The acquired image can be viewed on the upper computer software to see if it is within the field of view. If the image is slightly deviated, it can also be manually fine-tuned through the software. The laser ranging module is installed on the roof of the vehicle, located on both sides below the pantograph. Because the contact wire covers a large area and the imaging range varies greatly, a single sensor installation position cannot meet the shooting requirements. Therefore, a laser ranging module is needed to measure the distance between the roof of the vehicle and the pantograph. When the distance does not meet the detection requirements, the camera component angle can be adjusted in time by calculating the distance and position to ensure that the contact wire hanger is in the center of the camera's field of view.

[0042] In one embodiment, there are two high-definition line scan camera assemblies, which are respectively disposed on both sides of the contact wire. The two high-definition line scan camera assemblies are used to receive two identical equally spaced pulse signals transmitted by the signal synchronization control module.

[0043] like Figure 4 As shown, 1-post, 2-bar insulator, 3-flat cantilever arm, 4-load-bearing wire, 5-contact wire, 6-positioner, 7-drop string, 8-positioning tube support, 9-positioning tube, 10-single cantilever tube, A-left laser rangefinder, C-left high-definition line scan camera assembly, B-right laser rangefinder, D-right high-definition line scan camera assembly.

[0044] In a specific embodiment, a gimbal is deployed to address sunlight interference, allowing adjustment of the shooting angle of the high-definition line scan camera assembly to select a better angle and effectively avoid sunlight interference. Dual high-definition line scan camera assemblies are used to capture images from both sides of the contact wire. If the image captured on one side is affected by sunlight interference, the image captured on the other side simultaneously also avoids sunlight interference on the overall contact wire image. A fixed-wavelength infrared laser is used, providing a more stable light source. Combined with filters, this further enhances the device's anti-sunlight interference performance. This effectively avoids sunlight interference and significantly improves the contact wire detection effect. Slight image deviations can also be fine-tuned manually.

[0045] like Figure 2 As shown, in one embodiment, a pulse counting module is further included, for:

[0046] The total number of all equally spaced pulse signals from the start to the end of the high-speed inspection vehicle's movement is counted. Based on the total number of equally spaced pulse signals and the preset interval duration of the equally spaced pulse signals, the mileage value of the contact wire pole is determined. The mileage value of the contact wire pole is used for data fusion processing.

[0047] In a specific embodiment, the pulse counting module can count the number of input pulses (each pulse signal input to the linear scan camera triggers the acquisition of one line of images, and each line of images corresponds to an equal distance). The number of pulses N × the distance pulse width L can yield a precise mileage value S, which can be fed back to the mileage synchronization system for the positioning of the catenary support. Each catenary support detection data point has a corresponding number of pulses during processing, thus clearly defining the actual mileage information of the catenary support corresponding to the Nth pulse signal (note: the Nth pulse number is taken as the pulse value at the middle position of the catenary pole). The mileage of the detection line can be progressively subdivided, with the distance to the same infrastructure on each line considered as the scale division value; the smaller the division value, the more accurate the precision. For example, a 1km railway section AB can be divided into 5 zones (each 2000m) with 20 overhead contact line poles (generally 45-55m in straight sections and less than 40m in curved sections; in this example, we calculate 50m). These poles divide the AB section into 5 smaller mileage segments, similar to a scale; the smaller the graduation, the more precise the measurement. X represents the distance between two pulse signals. Therefore, the positioning accuracy of the overhead contact line can reach the millimeter level. Figure 5 As shown.

[0048] In one embodiment, the feature data is preprocessed to obtain processed feature data, including:

[0049] The feature data of the suspension wire data and support data are compared with the known ledger data, and the data are aligned according to the comparison results to obtain the aligned feature data.

[0050] In a specific embodiment, the information of the catenary droppers and catenary supports is located. Based on the located information of the catenary droppers and catenary supports, ledger data is generated. The generated ledger data is compared with the ledger data generated by the catenary geometric parameter detection system and the historical ledger data provided by the railway bureau. Data alignment is performed based on the comparison results.

[0051] In one embodiment, the dropper data includes dropper position data and dropper mileage data; the support data includes support position data and support mileage data.

[0052] In one embodiment, data fusion processing is performed on the dropper data, support data, and processed feature data, including:

[0053] Machine vision technology is used to perform data fusion processing on the waveform diagrams detected by the overhead contact system, the position data of the overhead contact system droppers and the support positions, the mileage data of the overhead contact system droppers and the support mileage data, and the high-definition images of the overhead contact system poles; the high-definition images of the overhead contact system poles are provided by a high-definition line scan camera assembly.

[0054] Figure 6 This is a waveform diagram of the contact wire geometric parameter detection data obtained using machine vision technology. In a specific embodiment, a large amount of normal dropper images, abnormal dropper images, and normal contact wire support images are collected to train the dropper and support positioning network algorithm model. Through repeated parameter tuning and iteration, the algorithm model is optimized to improve the positioning accuracy of contact wire droppers and contact wire supports.

[0055] In one embodiment, a review module is also included, for:

[0056] Based on the fused catenary dropper data and support data, abnormal points in the catenary data are identified. The catenary data includes one or any combination of the following: contact wire height, pantograph-catenary contact force, and hard points.

[0057] The abnormal points in the contact wire data were reviewed based on the high-definition images of the contact wire poles to determine the defects of the contact wire droppers.

[0058] The results of the contact wire dropper defects will be output in the form of graphics and statistical reports.

[0059] In a specific embodiment, by performing deep fusion analysis on the dropper data processed by deep learning, the high-definition dropper images captured by the acquisition module, and the waveforms detected by the catenary detection system, abnormal points in the detection data such as contact wire height, pantograph-catenary contact force, and hard points can be identified. The data can be viewed by browsing the high-definition dropper images, and the operational status of the contact suspension components at abnormal points can be verified, effectively improving the performance and ease of use of the catenary detection system.

[0060] This invention also provides a method for dynamic detection of bow wire and canopy lines, as described in the following embodiments. Since the principle behind this method is similar to that of the dynamic detection device for bow wire and canopy lines, the implementation of this method can refer to the implementation of the dynamic detection device for bow wire and canopy lines; repeated details will not be elaborated further. The method includes:

[0061] The command to acquire images is issued when the high-speed inspection vehicle begins to move;

[0062] According to the image acquisition command, equally spaced pulse signals are obtained from the photoelectric encoder driven by the wheel;

[0063] After obtaining the equally spaced pulse signals, images of the overhead contact line are captured and photographed.

[0064] Deep learning methods are used to identify the acquired images, resulting in identified image data, which includes dropper data and support data. The image data is then analyzed, and feature data for the dropper and support data is generated based on the analysis results. This feature data is then preprocessed to obtain processed feature data. The processed feature data, dropper data, and support data are then fused to obtain fused catenary dropper and support data. Finally, the fused catenary dropper and support data are output in the form of graphics, text, and statistical reports.

[0065] This invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-described dynamic detection method for bow and catenary suspension wires.

[0066] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described dynamic detection method for bow wire and catenary.

[0067] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-described dynamic detection method for bow wire and catenary.

[0068] In this embodiment of the invention, an image acquisition command is issued when the high-speed inspection vehicle begins to move; according to the image acquisition command, equally spaced pulse signals are obtained from the photoelectric encoder driven by the wheels; after obtaining the equally spaced pulse signals, images of the overhead contact line are acquired; the acquired images are identified using deep learning methods to obtain identified image data, which includes dropper data and support data; the image data is analyzed, and feature data of the dropper data and support data is generated based on the analysis results; the feature data is preprocessed to obtain processed feature data; the processed feature data, dropper data, and support data are fused to obtain fused overhead contact line dropper data and support data; the fused overhead contact line dropper data and support data are output in the form of graphics and statistical reports. This improves the clarity of the captured images and enhances the accuracy of the dropper data and support data through image analysis, thereby improving the accuracy of overhead contact line inspection.

[0069] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0070] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0071] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0072] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0073] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A dynamic detection device for bow and canopy suspension strings, characterized in that, include: The main control module, image acquisition module, high-definition line scan camera assembly, image processing module, and signal synchronization control module; among them, The main control module is used to issue image acquisition commands when the high-speed inspection vehicle starts moving. The signal synchronization control module is used to: obtain equally spaced pulse signals from the photoelectric encoder driven by the wheel after receiving the image acquisition command, and input the equally spaced pulse signals into the high-definition line scan camera component; A high-definition line scan camera assembly is used to capture images of the contact network after receiving equally spaced pulse signals. There are two high-definition line scan camera assemblies, which are respectively set on both sides of the contact network. The two high-definition line scan camera assemblies are used to receive two identical equally spaced pulse signals transmitted by the signal synchronization control module. The image acquisition module is used to: acquire contact network images captured by the high-definition line scan camera assembly and provide them to the image processing module; The image processing module is used to: identify the catenary images sent by the image acquisition module using deep learning methods to obtain identified image data, which includes dropper data and support data; analyze the image data, generate feature data for the dropper data and support data based on the analysis results, and preprocess the feature data to obtain processed feature data; perform data fusion processing on the dropper data, support data, and processed feature data to obtain fused catenary dropper data and support data, and output the fused catenary dropper data and support data in the form of graphics and statistical reports; The dropper data includes dropper position data and dropper mileage data; the support data includes support position data and support mileage data. Data fusion processing is performed on the dropper data, support data, and processed feature data, including: Machine vision technology is used to perform data fusion processing on the waveform diagrams detected by the overhead contact system, the position data of the overhead contact system droppers and the support positions, the mileage data of the overhead contact system droppers and the support mileage data, and the high-definition images of the overhead contact system poles; the high-definition images of the overhead contact system poles are provided by a high-definition line scan camera assembly.

2. The apparatus as claimed in claim 1, characterized in that, It also includes a laser ranging module, used for: Measure the distance between the vehicle roof and the pantograph, and determine the height of the contact wire based on the distance between the vehicle roof and the pantograph; Adjust the angle of the high-definition line scan camera assembly according to the height of the contact line.

3. The apparatus as described in claim 1, characterized in that, It also includes a pulse counting module, used for: The total number of all equally spaced pulse signals from the start to the end of the high-speed inspection vehicle's movement is counted. Based on the total number of equally spaced pulse signals and the preset interval duration of the equally spaced pulse signals, the mileage value of the contact wire pole is determined. The mileage value of the contact wire pole is used for data fusion processing.

4. The apparatus as claimed in claim 1, characterized in that, The feature data is preprocessed to obtain the processed feature data, including: The feature data of the suspension wire data and support data are compared with the known ledger data, and the data are aligned according to the comparison results to obtain the aligned feature data.

5. The apparatus as claimed in claim 1, characterized in that, It also includes a review module, used for: Based on the fused catenary dropper data and support data, abnormal points in the catenary data are identified. The catenary data includes one or any combination of the following: contact wire height, pantograph-catenary contact force, and hard points. The abnormal points in the contact wire data were reviewed based on the high-definition images of the contact wire poles to determine the defects of the contact wire droppers. The results of the contact wire dropper defects will be output in the form of graphics and statistical reports.

6. A method for dynamic detection of the bow wire and catenary suspension wire applied to the device described in claims 1-5, characterized in that, include: The command to acquire images is issued when the high-speed inspection vehicle begins to move; According to the image acquisition command, equally spaced pulse signals are obtained from the photoelectric encoder driven by the wheel; After obtaining the equally spaced pulse signals, images of the overhead contact line are captured and photographed. Deep learning methods are used to identify the acquired images, resulting in identified image data, which includes dropper data and support data. The image data is then analyzed, and feature data for the dropper and support data is generated based on the analysis results. This feature data is then preprocessed to obtain processed feature data. The processed feature data, dropper data, and support data are then fused to obtain fused catenary dropper and support data. Finally, the fused catenary dropper and support data are output in the form of graphics, text, and statistical reports.

7. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of claim 6.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method of claim 6.

9. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the method of claim 6.

Images