Track fault detection method and system
By image processing and clustering of tracks, and combining the preset model of the self-attention module for track fault detection, the problems of incomplete, inaccurate and unintelligent fault monitoring in the existing technology are solved, and more efficient track faults are achieved Test.
Patent Information
- Application Number
- CN202411461695.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-18
- Publication Date
- 2025-02-07
AI Technical Summary
The existing track fault detection methods have problems such as incomplete, inaccurate and unintelligent fault monitoring, and rely on manual testing and sensor reliability and accuracy requirements.
A track fault detection method is adopted. By dividing the continuous track into tracks to be detected, the track image to be detected is obtained, and clustering is performed according to the preset clustering rules, the target track image to be detected is selected for preprocessing, and the track edge is extracted Images, using preset models for fault detection, and the model's backbone network includes a self-attention module.
This method can reduce the detection frequency of tracks to be detected with the same characteristics, improve the track fault detection efficiency, and achieve more comprehensive, accurate and intelligent fault monitoring.
Smart Images
Figure CN119399720A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of rail transit detection, and in particular relates to a rail fault detection method and system. Background Art
[0002] At present, the track fault detection method is mainly carried out through testing by on-site staff and computer monitoring. For example, on-site staff use tools to knock, naked eye visual inspection, or use sensors such as laser, ultrasonic, infrared and eddy current to detect track faults. The inventors found that the above methods must rely on manual testing or computer monitoring by on-site staff. The subjective influence of on-site staff has a greater impact on the monitoring results. In addition, the scheme of track fault detection with sensors also has high requirements on the reliability and accuracy of the sensors themselves.
[0003] Therefore, the above solution has problems such as incomplete, inaccurate and unintelligent fault monitoring. Summary of the invention
[0004] The present invention provides a track fault detection method and system, which are used to solve the technical problems of incomplete, inaccurate and unintelligent fault monitoring.
[0005] In a first aspect, the present invention provides a track fault detection method, comprising:
[0006] Dividing the continuous track into at least one track to be detected, and acquiring a track image to be detected of the at least one track to be detected;
[0007] Clustering each track image to be detected according to a preset clustering rule to obtain at least one track image set to be detected, wherein a track image set to be detected includes at least one track image to be detected of the same type;
[0008] Selecting a target track image to be detected from a set of track images to be detected using a preset selection rule, and preprocessing the target track image to be detected to obtain a target image;
[0009] Performing edge extraction on the target image to obtain a track edge image;
[0010] Fault detection is performed on the track edge image through a preset model to obtain a fault detection result corresponding to the track image to be detected, wherein the backbone network of the preset model includes a self-attention module.
[0011] In a second aspect, the present invention provides a track fault detection system, comprising:
[0012] an acquisition module, configured to divide the continuous track into at least one track to be detected, and acquire a track image to be detected of the at least one track to be detected;
[0013] A clustering module is configured to cluster the various track images to be detected according to a preset clustering rule to obtain at least one set of track images to be detected, wherein a set of track images to be detected includes at least one track image to be detected of the same type;
[0014] A selection module is configured to select a target track image to be detected from a set of track images to be detected using a preset selection rule, and preprocess the target track image to be detected to obtain a target image;
[0015] An extraction module, configured to perform edge extraction on the target image to obtain a track edge image;
[0016] The detection module is configured to perform fault detection on the track edge image through a preset model to obtain a fault detection result corresponding to the track image to be detected, wherein the backbone network of the preset model includes a self-attention module.
[0017] In a third aspect, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the steps of the track fault detection method of any embodiment of the present invention.
[0018] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor executes the steps of the track fault detection method of any embodiment of the present invention.
[0019] The track fault detection method and system of the present application adopt a preset selection rule to select a target track image to be detected from a set of track images to be detected, and pre-process the target track image to be detected to obtain a target image. When a fault exists in the target image, other track images to be detected in the set of track images to be detected are subjected to fault detection, which can reduce the detection frequency of tracks to be detected with the same characteristics, thereby effectively improving the track fault detection efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0021] Figure 1A flow chart of a track fault detection method provided by one embodiment of the present invention;
[0022] Figure 2 A structural block diagram of a track fault detection system provided by an embodiment of the present invention;
[0023] Figure 3 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0024] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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.
[0025] See also Figure 1 , which shows a flow chart of a track fault detection method of the present application.
[0026] like Figure 1 As shown, the track fault detection method specifically includes the following steps:
[0027] Step S101 : dividing a continuous track into at least one track to be detected, and acquiring a track image to be detected of the at least one track to be detected.
[0028] Step S102 : clustering the track images to be detected according to a preset clustering rule to obtain at least one set of track images to be detected, wherein a set of track images to be detected includes at least one track image to be detected of the same type.
[0029] In this step, any two track images to be detected are input into a pre-built image recognition model, and the image recognition model outputs a first feature result and a second feature result corresponding to the any two track images to be detected; it is determined whether the difference between the first feature result and the second feature result is greater than a preset threshold; if it is greater than the preset threshold, the any two track images to be detected are divided into the same set of track images to be detected; if it is not greater than the preset threshold, the any two track images to be detected are divided into two different sets of track images to be detected.
[0030] It should be noted that the image recognition model may be a VGGNet network model, which is obtained through machine learning using multiple sets of training images and detection images.
[0031] Step S103 , selecting a target track image to be detected from a set of track images to be detected using a preset selection rule, and preprocessing the target track image to be detected to obtain a target image.
[0032] In this step, the feature results of each track image to be detected in a set of track images to be detected are obtained according to the image recognition model; and a track image to be detected corresponding to the maximum feature result among the feature results of each track image to be detected is selected as the target track image to be detected.
[0033] It should be noted that the target track image to be detected is grayed; the target track image to be detected after the graying process is image enhanced; the target track image to be detected after the image enhancement process is filtered to obtain the image to be segmented; the image to be segmented is segmented according to the preset segmentation rules to obtain the target image.
[0034] Step S104: extracting edges of the target image to obtain a track edge image.
[0035] In this step, the gradient strength and direction information of each pixel in the target image are obtained through the 8-directional template of the Sobel operator, and the target image is edge enhanced according to the gradient strength and direction information to obtain an edge enhanced image; the edge enhanced image is non-maximum suppressed to obtain a simplified edge image; the strong edge threshold and weak edge threshold of the refined edge image are determined by the Otsu method; the strong edge pixel points and the possible edge pixel points are determined according to the strong edge threshold and the weak edge threshold, and after connecting the strong edge pixel points and the adjacent possible edge pixel points, a track edge image is generated.
[0036] Step S105, performing fault detection on the track edge image through a preset model to obtain a fault detection result corresponding to the track image to be detected, wherein the backbone network of the preset model includes a self-attention module.
[0037] In this step, the track edge image is input into the preset model, and the self-attention module of the backbone network is used to generate an attention weight map according to the track edge image. After the track edge image is weighted according to the attention weight map by the backbone network, the fault detection result corresponding to the track edge image output by the preset model is obtained; the fault detection result is used to characterize whether there are fault features in the track edge image, and the fault features include the track fault type and the track fault position corresponding to the track fault type.
[0038] Specifically, after fault detection is performed on the track edge image through a preset model to obtain a fault detection result corresponding to the track image to be detected, when the fault detection result indicates that there are fault features in the track edge image, the track fault type in the fault features is obtained; the fault handling party is determined according to the track fault type, and a maintenance notification containing the fault features is sent to the fault handling party.
[0039] In summary, the method of the present application adopts a preset selection rule to select a target track image to be detected from a set of track images to be detected, and preprocesses the target track image to be detected to obtain a target image. When a fault exists in the target image, other track images to be detected in a set of track images to be detected are subjected to fault detection, which can reduce the detection frequency of tracks to be detected with the same characteristics, thereby effectively improving the efficiency of track fault detection.
[0040] See also Figure 2 , which shows a structural block diagram of a track fault detection system of the present application.
[0041] like Figure 2 As shown, the track fault detection system 200 includes an acquisition module 210 , a clustering module 220 , a selection module 230 , an extraction module 240 and a detection module 250 .
[0042] The acquisition module 210 is configured to divide the continuous track into at least one track to be detected, and acquire a track image to be detected of the at least one track to be detected;
[0043] The clustering module 220 is configured to perform clustering processing on each track image to be detected according to a preset clustering rule to obtain at least one set of track images to be detected, wherein a set of track images to be detected includes at least one track image to be detected of the same type;
[0044] The selection module 230 is configured to select a target track image to be detected from a set of track images to be detected using a preset selection rule, and preprocess the target track image to be detected to obtain a target image;
[0045] An extraction module 240 is configured to perform edge extraction on the target image to obtain a track edge image;
[0046] The detection module 250 is configured to perform fault detection on the track edge image through a preset model to obtain a fault detection result corresponding to the track image to be detected, wherein the backbone network of the preset model includes a self-attention module.
[0047] It should be understood that Figure 2 Modules and references documented in Figure 1Therefore, the operations and features described above for the method and the corresponding technical effects are also applicable to Figure 2 The modules in it will not be described in detail here.
[0048] In some other embodiments, the embodiments of the present invention further provide a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor is caused to execute the rail fault detection method in any of the above method embodiments;
[0049] As an implementation mode, the computer-readable storage medium of the present invention stores computer-executable instructions, and the computer-executable instructions are configured as follows:
[0050] Dividing the continuous track into at least one track to be detected, and acquiring a track image to be detected of the at least one track to be detected;
[0051] Clustering each track image to be detected according to a preset clustering rule to obtain at least one track image set to be detected, wherein a track image set to be detected includes at least one track image to be detected of the same type;
[0052] Selecting a target track image to be detected from a set of track images to be detected using a preset selection rule, and preprocessing the target track image to be detected to obtain a target image;
[0053] Performing edge extraction on the target image to obtain a track edge image;
[0054] Fault detection is performed on the track edge image through a preset model to obtain a fault detection result corresponding to the track image to be detected, wherein the backbone network of the preset model includes a self-attention module.
[0055] The computer-readable storage medium may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required by at least one function; the data storage area may store data created according to the use of the track fault detection system, etc. In addition, the computer-readable storage medium may include a high-speed random access memory, and may also include a memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the track fault detection system via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0056] Figure 3 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 3As shown, the device includes: a processor 310 and a memory 320. The electronic device may also include: an input device 330 and an output device 340. The processor 310, the memory 320, the input device 330 and the output device 340 may be connected via a bus or other means. Figure 3 The example of the bus connection is taken. The memory 320 is the computer-readable storage medium mentioned above. The processor 310 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions and modules stored in the memory 320, that is, the track fault detection method of the above method embodiment is realized. The input device 330 can receive input digital or character information, and generate key signal input related to user settings and function control of the track fault detection system. The output device 340 may include display devices such as display screens.
[0057] The electronic device can execute the method provided by the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not described in detail in this embodiment, please refer to the method provided by the embodiment of the present invention.
[0058] As an implementation mode, the electronic device is applied to a track fault detection system and is used for a client, and includes: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can:
[0059] Dividing the continuous track into at least one track to be detected, and acquiring a track image to be detected of the at least one track to be detected;
[0060] Clustering each track image to be detected according to a preset clustering rule to obtain at least one track image set to be detected, wherein a track image set to be detected includes at least one track image to be detected of the same type;
[0061] Selecting a target track image to be detected from a set of track images to be detected using a preset selection rule, and preprocessing the target track image to be detected to obtain a target image;
[0062] Performing edge extraction on the target image to obtain a track edge image;
[0063] Fault detection is performed on the track edge image through a preset model to obtain a fault detection result corresponding to the track image to be detected, wherein the backbone network of the preset model includes a self-attention module.
[0064] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiment.
[0065] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A track fault detection method, characterized in that: include: Dividing the continuous track into at least one track to be detected, and acquiring a track image to be detected of the at least one track to be detected; Clustering each track image to be detected according to a preset clustering rule to obtain at least one track image set to be detected, wherein a track image set to be detected includes at least one track image to be detected of the same type; Selecting a target track image to be detected from a set of track images to be detected using a preset selection rule, and preprocessing the target track image to be detected to obtain a target image; Performing edge extraction on the target image to obtain a track edge image; Fault detection is performed on the track edge image through a preset model to obtain a fault detection result corresponding to the track image to be detected, wherein the backbone network of the preset model includes a self-attention module.
2. A track fault detection method according to claim 1, characterized in that: The clustering of the to-be-detected track images according to a preset clustering rule to obtain at least one to-be-detected track image set comprises: Input any two to-be-detected track images into a pre-built image recognition model, and the image recognition model outputs a first feature result and a second feature result corresponding to the any two to-be-detected track images; Determine whether the difference between the first feature result and the second feature result is greater than a preset threshold; If it is greater than a preset threshold, the two track images to be detected are divided into the same track image set to be detected; If it is not greater than a preset threshold, the arbitrary two to-be-detected track images are divided into two different to-be-detected track image sets.
3. A track fault detection method according to claim 2, characterized in that: The method of selecting a target track image to be detected from a set of track images to be detected by using a preset selection rule includes: Acquire, according to the image recognition model, feature results of each track image to be detected in the set of track images to be detected; A track image to be detected corresponding to the maximum feature result among the feature results of the track images to be detected is selected as the target track image to be detected.
4. A track fault detection method according to claim 1, characterized in that: The preprocessing of the target to-be-detected track image to obtain the target image comprises: grayscale the target track image to be detected; Performing image enhancement processing on the target track image to be detected after the grayscale processing is completed; Performing filtering processing on the target track image to be detected after image enhancement processing to obtain an image to be segmented; The image to be segmented is subjected to image segmentation processing according to a preset segmentation rule to obtain the target image.
5. A track fault detection method according to claim 1, characterized in that: The performing fault detection on the track edge image by using a preset model to obtain a fault detection result corresponding to the track image to be detected includes: The track edge image is input into the preset model, and an attention weight map is generated according to the track edge image using the self-attention module of the backbone network. After the track edge image is weightedly processed by the backbone network according to the attention weight map, a fault detection result corresponding to the track edge image output by the preset model is obtained; the fault detection result is used to characterize whether there are fault features in the track edge image, and the fault features include a track fault type and a track fault position corresponding to the track fault type.
6. A track fault detection method according to claim 1, characterized in that: After performing fault detection on the track edge image by using a preset model to obtain a fault detection result corresponding to the track image to be detected, the method further includes: When the fault detection result indicates that there is a fault feature in the track edge image, obtaining the track fault type in the fault feature; A fault handler is determined according to the track fault type, and a maintenance notice containing the fault characteristics is sent to the fault handler.
7. A track fault detection system, characterized in that: include: an acquisition module, configured to divide the continuous track into at least one track to be detected, and acquire a track image to be detected of the at least one track to be detected; A clustering module is configured to cluster the various track images to be detected according to a preset clustering rule to obtain at least one set of track images to be detected, wherein a set of track images to be detected includes at least one track image to be detected of the same type; A selection module is configured to select a target track image to be detected from a set of track images to be detected using a preset selection rule, and preprocess the target track image to be detected to obtain a target image; An extraction module, configured to perform edge extraction on the target image to obtain a track edge image; The detection module is configured to perform fault detection on the track edge image through a preset model to obtain a fault detection result corresponding to the track image to be detected, wherein the backbone network of the preset model includes a self-attention module.
8. An electronic device, characterized in that: include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method described in any one of claims 1 to 6.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.