Railway Guardrail Defect Detection Method and Equipment

Abstract

The present invention provides a method and device for detecting defects in railway guardrails, including: collecting a first image of railway guardrail defects and a second image of railway guardrail defects, and obtaining an effective first detection model and an effective second detection model based on this; inputting the real-time collected railway guardrail image into the first detection model, if the detection result is that the railway guardrail has a first defect, the detection is completed, if the detection result is that the railway guardrail is normal, then input the real-time collected railway guardrail image into the second detection model for detection, and the second defect of the railway guardrail is detected. The present invention can detect various railway guardrail defects and improve the detection accuracy of railway guardrail defects.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of railway maintenance, and in particular, to a method and device for detecting defects in railway guardrails. Background Art

[0002] The integrity detection of railway guardrails is of great significance for maintaining the integrity of railway guardrails and ensuring the safety of railway operations. Current maintenance methods all require manual detection and troubleshooting of guardrails, which are difficult to apply on-site. At the same time, they are greatly affected by factors such as light and shadow, resulting in low integrity of manual judgment and a high false alarm rate. Therefore, developing a method and device for detecting defects in railway guardrails to effectively overcome the deficiencies in the above-mentioned related technologies has become an urgent technical problem in the industry. Summary of the Invention

[0003] In view of the above problems existing in the prior art, the embodiments of the present invention provide a method and device for detecting defects in railway guardrails.

[0004] In a first aspect, an embodiment of the present invention provides a method for detecting defects in railway guardrails, including: collecting a first railway guardrail defect image and a second railway guardrail defect image; dividing the first railway guardrail defect image into a first railway guardrail defect training data set and a first railway guardrail defect test data set, training a deep convolutional neural network using the first railway guardrail defect training data set to obtain a first detection model, testing the first detection model using the first railway guardrail defect test data set, and if the output result of the first detection model matches the expected result, determining that the first detection model is a valid model; dividing the second railway guardrail defect image into a second railway guardrail defect training data set and a second railway guardrail defect test data set, training a deep convolutional neural network using the second railway guardrail defect training data set to obtain a second detection model, testing the second detection model using the second railway guardrail defect test data set, and if the output result of the second detection model matches the expected result, determining that the second detection model is a valid model; inputting a real-time collected railway guardrail image into the first detection model, if the detection result is that the railway guardrail has a first defect, the detection is completed, and if the detection result is that the railway guardrail is normal, inputting the real-time collected railway guardrail image into the second detection model for detection to obtain a second defect of the railway guardrail.

[0005] Based on the above content of the method embodiment, in the method for detecting defects in railway guardrails provided by the embodiments of the present invention, the collecting of the first railway guardrail defect image includes: using an integrated industrial camera and a positioning sensor to collect a railway guardrail image and corresponding railway guardrail position information, and using the collapsed railway guardrail image as the first railway guardrail defect image.

[0006] Based on the content of the above method embodiments, in the railway guardrail defect detection method provided in the embodiments of the present invention, the acquisition of the second railway guardrail defect image includes: using the fracture image, damage image, and missing post cap image of the non-collapsed railway guardrail as the second railway guardrail defect image.

[0007] Based on the content of the above method embodiments, in the railway guardrail defect detection method provided in the embodiments of the present invention, for inputting the real-time acquired railway guardrail image into the first detection model, if the detection result is that the railway guardrail has a first defect, then the detection is completed, including: the on-vehicle inspection camera real-time acquires the railway guardrail image, inputs the acquired railway guardrail image into the first detection model, the first detection model detects whether the railway guardrail has collapsed, if the railway guardrail has collapsed, then outputs that the railway guardrail has a first defect, and the detection is completed.

[0008] Based on the content of the above method embodiments, in the railway guardrail defect detection method provided in the embodiments of the present invention, if the detection result is that the railway guardrail is normal, then input the real-time acquired railway guardrail image into the second detection model for detection to obtain the second defect of the railway guardrail, including: if the detection result is that the railway guardrail is normal, then crop the real-time acquired railway guardrail image, and input the cropped railway guardrail image into the second detection model for detection to obtain the second defect of the railway guardrail.

[0009] Based on the content of the above method embodiments, in the railway guardrail defect detection method provided in the embodiments of the present invention, the second defect includes: the railway guardrail is fractured or damaged.

[0010] Based on the content of the above method embodiments, in the railway guardrail defect detection method provided in the embodiments of the present invention, the second defect further includes: the post cap of the railway guardrail is missing.

[0011] Second aspect, an embodiment of the present invention provides a railway guardrail defect detection device, including: a first main module for collecting a first railway guardrail defect image and a second railway guardrail defect image; a second main module for dividing the first railway guardrail defect image into a first railway guardrail defect training data set and a first railway guardrail defect test data set, training a deep convolutional neural network using the first railway guardrail defect training data set to obtain a first detection model, testing the first detection model using the first railway guardrail defect test data set, and if the output result of the first detection model matches the expected result, determining that the first detection model is a valid model; a third main module for dividing the second railway guardrail defect image into a second railway guardrail defect training data set and a second railway guardrail defect test data set, training a deep convolutional neural network using the second railway guardrail defect training data set to obtain a second detection model, testing the second detection model using the second railway guardrail defect test data set, and if the output result of the second detection model matches the expected result, determining that the second detection model is a valid model; a fourth main module for inputting a real-time collected railway guardrail image into the first detection model, if the detection result is that the railway guardrail has a first defect, the detection is completed, if the detection result is that the railway guardrail is normal, inputting the real-time collected railway guardrail image into the second detection model for detection, and detecting a second defect of the railway guardrail.

[0012] Third aspect, an embodiment of the present invention provides an electronic device, including:

[0013] At least one processor; and

[0014] At least one memory communicatively connected to the processor, wherein:

[0015] The memory stores program instructions executable by the processor, and the processor can execute the railway guardrail defect detection method provided by any one of the various implementation manners of the first aspect by invoking the program instructions.

[0016] Fourth aspect, an embodiment of the present invention provides a non-transitory computer-readable storage medium, and the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions cause the computer to execute the railway guardrail defect detection method provided by any one of the various implementation manners of the first aspect.

[0017] The railway guardrail defect detection method and device provided by the embodiments of the present invention can detect various railway guardrail defects by using a multi-level deep convolutional neural network to identify different states of the railway guardrail, have good scene generalization for railway guardrail defect detection, strong anti-interference ability against external adverse environments, and improve the detection accuracy of railway guardrail defects. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0019] Figure 1 Flowchart of the railway guardrail defect detection method provided by the embodiment of the present invention;

[0020] Figure 2 Schematic structural diagram of the railway guardrail defect detection device provided by the embodiment of the present invention;

[0021] Figure 3 Schematic physical structure diagram of the electronic device provided by the embodiment of the present invention;

[0022] Figure 4 Schematic diagram of the cutting effect on a normal railway guardrail provided by the embodiment of the present invention;

[0023] Figure 5 Schematic diagram of the normal number of railway guardrail column caps provided by the embodiment of the present invention;

[0024] Figure 6 Schematic diagram of the missing number of railway guardrail column caps provided by the embodiment of the present invention. Detailed implementation manners

[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention. In addition, the technical features in each embodiment or a single embodiment provided by the present invention can be combined with each other arbitrarily to form a feasible technical solution. This combination is not restricted by the order of steps and / or the structural composition mode, but must be based on the fact that those of ordinary skill in the art can implement it. When the combination of technical solutions results in contradictions or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0026] The embodiment of the present invention provides a railway guardrail defect detection method. Refer to Figure 1, the method includes: collecting a first railway guardrail defect image and a second railway guardrail defect image; dividing the first railway guardrail defect image into a first railway guardrail defect training data set and a first railway guardrail defect test data set, training a deep convolutional neural network using the first railway guardrail defect training data set to obtain a first detection model, and testing the first detection model using the first railway guardrail defect test data set. If the output result of the first detection model matches the expected result, it is determined that the first detection model is a valid model; dividing the second railway guardrail defect image into a second railway guardrail defect training data set and a second railway guardrail defect test data set, training a deep convolutional neural network using the second railway guardrail defect training data set to obtain a second detection model, and testing the second detection model using the second railway guardrail defect test data set. If the output result of the second detection model matches the expected result, it is determined that the second detection model is a valid model; inputting the real-time collected railway guardrail image into the first detection model. If the detection result is that the railway guardrail has a first defect, the detection is completed. If the detection result is that the railway guardrail is normal, the real-time collected railway guardrail image is input into the second detection model for detection, and a second defect of the railway guardrail is detected.

[0027] Based on the content of the above method embodiment, as an optional embodiment, in the railway guardrail defect detection method provided in the embodiment of the present invention, the collecting the first railway guardrail defect image includes: using an integrated industrial camera and a positioning sensor to collect a railway guardrail image and corresponding railway guardrail position information, and using the collapsed railway guardrail image as the first railway guardrail defect image.

[0028] Based on the content of the above method embodiment, as an optional embodiment, in the railway guardrail defect detection method provided in the embodiment of the present invention, the collecting the second railway guardrail defect image includes: using the fracture image, damage image, and missing post cap image of the non-collapsed railway guardrail as the second railway guardrail defect image.

[0029] Based on the content of the above method embodiment, as an optional embodiment, in the railway guardrail defect detection method provided in the embodiment of the present invention, the inputting the real-time collected railway guardrail image into the first detection model, if the detection result is that the railway guardrail has a first defect, the detection is completed, includes: a vehicle-mounted inspection camera collects a railway guardrail image in real time, inputs the collected railway guardrail image into the first detection model, and the first detection model detects whether the railway guardrail has collapsed. If the railway guardrail has collapsed, it outputs that the railway guardrail has a first defect, and the detection is completed.

[0030] Based on the content of the above method embodiments, as an alternative embodiment, in the railway guardrail defect detection method provided in the embodiments of the present invention, if the detection result is that the railway guardrail is normal, the real-time acquired railway guardrail image is input into a second detection model for detection, and the second defect of the railway guardrail is detected, including: if the detection result is that the railway guardrail is normal, the real-time acquired railway guardrail image is cropped, and the cropped railway guardrail image is input into the second detection model for detection to obtain the second defect of the railway guardrail.

[0031] Specifically, refer to Figure 4 , the on-vehicle inspection camera captures the railway guardrail image, and then inputs the railway guardrail image into the first detection model. The deep convolutional neural network detects whether the guardrail has collapsed. If the guardrail has collapsed, it outputs that there is a defect in the guardrail; if the guardrail is normal, it crops the detected guardrail part as Figure 4 shown by the two solid boxes in.

[0032] Based on the content of the above method embodiments, as an alternative embodiment, in the railway guardrail defect detection method provided in the embodiments of the present invention, the second defect includes: the railway guardrail is broken or damaged.

[0033] Based on the content of the above method embodiments, as an alternative embodiment, in the railway guardrail defect detection method provided in the embodiments of the present invention, the second defect further includes: the column cap of the railway guardrail is missing.

[0034] Specifically, refer to Figure 5 and Figure 6 , the second detection model processes the cropped railway guardrail screenshot. If it detects that the guardrail is broken and damaged, it outputs that there is a defect in the guardrail; it detects the number of column caps in the guardrail. A normal guardrail has two column caps (such as the two caps in Figure 5 ). If there are only 0 or 1 column caps (such as one cap in Figure 6 ), it means that the column cap is missing and outputs that there is a defect in the guardrail; if there are 2 column caps, it means that the number of column caps is normal and outputs that the guardrail is normal.

[0035] The railway guardrail defect detection method provided in the embodiments of the present invention can detect various railway guardrail defects by using a multi-level deep convolutional neural network to identify different states of the railway guardrail. It has good scene generalization for railway guardrail defect detection, strong anti-interference ability against external adverse environments, and improves the detection accuracy of railway guardrail defects.

[0036] The implementation basis of each embodiment of the present invention is achieved through programmed processing by a device with a processor function. Therefore, in engineering practice, the technical solutions and functions of each embodiment of the present invention can be encapsulated into various modules. Based on this actual situation, on the basis of the above embodiments, an embodiment of the present invention provides a railway guardrail defect detection device, which is used to execute the railway guardrail defect detection method in the above method embodiment. Refer to Figure 2 , the device includes: a first main module for collecting a first railway guardrail defect image and a second railway guardrail defect image; a second main module for dividing the first railway guardrail defect image into a first railway guardrail defect training data set and a first railway guardrail defect test data set, training a deep convolutional neural network using the first railway guardrail defect training data set to obtain a first detection model, testing the first detection model using the first railway guardrail defect test data set, and if the output result of the first detection model matches the expected result, determining the first detection model as an effective model; a third main module for dividing the second railway guardrail defect image into a second railway guardrail defect training data set and a second railway guardrail defect test data set, training a deep convolutional neural network using the second railway guardrail defect training data set to obtain a second detection model, testing the second detection model using the second railway guardrail defect test data set, and if the output result of the second detection model matches the expected result, determining the second detection model as an effective model; a fourth main module for inputting the real-time collected railway guardrail image into the first detection model, if the detection result is that the railway guardrail has a first defect, the detection is completed, if the detection result is that the railway guardrail is normal, the real-time collected railway guardrail image is input into the second detection model for detection, and the second defect of the railway guardrail is detected.

[0037] The railway guardrail defect detection device provided by the embodiment of the present invention uses Figure 2 several modules of it. By using a multi-level deep convolutional neural network to identify different states of the railway guardrail, various railway guardrail defects can be detected, the scene generalization of railway guardrail defect detection is better, the anti-interference ability against external adverse environments is stronger, and the detection accuracy of railway guardrail defects is improved.

[0038] It should be noted that the device in the device embodiment provided by the present invention can be used not only to implement the method in the above method embodiment, but also to implement the methods in other method embodiments provided by the present invention. The difference lies only in setting corresponding functional modules. The principle is basically the same as that of the above device embodiment provided by the present invention. As long as those skilled in the art, on the basis of the above device embodiment, refer to the specific technical solutions in other method embodiments, obtain corresponding technical means by combining technical features, and the technical solutions composed of these technical means, and ensure the practicality of the technical solutions, they can improve the device in the above device embodiment to obtain corresponding device-type embodiments for implementing the methods in other method-type embodiments. For example:

[0039] Based on the content of the above device embodiment, as an optional embodiment, the railway guardrail defect detection device provided in the embodiment of the present invention further includes: a first sub-module for implementing the acquisition of the first railway guardrail defect image, including: using an integrated industrial camera and a positioning sensor to acquire a railway guardrail image and the corresponding railway guardrail position information, and taking the collapsed railway guardrail image as the first railway guardrail defect image.

[0040] Based on the content of the above device embodiment, as an optional embodiment, the railway guardrail defect detection device provided in the embodiment of the present invention further includes: a second sub-module for implementing the acquisition of the second railway guardrail defect image, including: taking the fracture image, damage image and column cap missing image of the non-collapsed railway guardrail as the second railway guardrail defect image.

[0041] Based on the content of the above device embodiment, as an optional embodiment, the railway guardrail defect detection device provided in the embodiment of the present invention further includes: a third sub-module for implementing inputting the real-time acquired railway guardrail image into a first detection model. If the detection result is that the railway guardrail has a first defect, the detection is completed, including: a vehicle-mounted inspection camera real-time acquires a railway guardrail image, inputs the acquired railway guardrail image into the first detection model, and the first detection model detects whether the railway guardrail has collapsed. If the railway guardrail has collapsed, it outputs that the railway guardrail has a first defect, and the detection is completed.

[0042] Based on the content of the above device embodiment, as an optional embodiment, the railway guardrail defect detection device provided in the embodiment of the present invention further includes: a fourth sub-module for implementing if the detection result is that the railway guardrail is normal, inputting the real-time acquired railway guardrail image into a second detection model for detection to obtain the second defect of the railway guardrail, including: if the detection result is that the railway guardrail is normal, cropping the real-time acquired railway guardrail image, and inputting the cropped railway guardrail image into the second detection model for detection to obtain the second defect of the railway guardrail.

[0043] Based on the content of the above device embodiments, as an alternative embodiment, the railway guardrail defect detection device provided in the embodiments of the present invention further includes: a fifth sub-module, configured to implement that the second defect includes: breakage or damage of the railway guardrail.

[0044] Based on the content of the above device embodiments, as an alternative embodiment, the railway guardrail defect detection device provided in the embodiments of the present invention further includes: a sixth sub-module, configured to implement that the second defect further includes: loss of the column cap of the railway guardrail.

[0045] The method of the embodiments of the present invention is implemented relying on an electronic device. Therefore, it is necessary to introduce the relevant electronic device. For this purpose, the embodiments of the present invention provide an electronic device, as Figure 3 shown, the electronic device includes: at least one processor, a communications interface, at least one memory, and a communication bus. Among them, the at least one processor, the communications interface, and the at least one memory complete mutual communication through the communication bus. The at least one processor can call the logical instructions in the at least one memory to execute all or part of the steps of the methods provided in the foregoing method embodiments.

[0046] In addition, when the logical instructions in the foregoing at least one memory are implemented in the form of a software functional unit and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, 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. The 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 the method embodiments of the present invention. The foregoing storage medium includes: various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc that can store program codes.

[0047] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0048] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment 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 such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0049] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to multiple embodiments of the present invention. Based on this understanding, 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 from that marked in the accompanying drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and sometimes may 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, and 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.

[0050] In this patent, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, the elements defined by the statement "comprising..." do not exclude the presence of additional identical elements in the process, method, article or device comprising the said elements.

[0051] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting defects in railway guardrails, characterized in that, Including: Collecting the first railway guardrail defect image and the second railway guardrail defect image; Dividing the first railway guardrail defect image into a first railway guardrail defect training data set and a first railway guardrail defect test data set, training a deep convolutional neural network with the first railway guardrail defect training data set to obtain a first detection model, and testing the first detection model with the first railway guardrail defect test data set. If the output result of the first detection model matches the expected result, determine that the first detection model is a valid model; divide the second railway guardrail defect image into a second railway guardrail defect training data set and a second railway guardrail defect test data set, train a deep convolutional neural network with the second railway guardrail defect training data set to obtain a second detection model, and test the second detection model with the second railway guardrail defect test data set. If the output result of the second detection model matches the expected result, determine that the second detection model is a valid model; input the real-time collected railway guardrail image into the first detection model. If the detection result is that the railway guardrail has a first defect, the detection is completed. If the detection result is that the railway guardrail is normal, input the real-time collected railway guardrail image into the second detection model for detection, and detect the second defect of the railway guardrail; The collecting of the first railway guardrail defect image includes: using an integrated industrial camera and a positioning sensor to collect a railway guardrail image and the corresponding railway guardrail position information, and taking the collapsed railway guardrail image as the first railway guardrail defect image; The collecting of the second railway guardrail defect image includes: taking the fracture image, damage image and missing column cap image of the non-collapsed railway guardrail as the second railway guardrail defect image; The inputting of the real-time collected railway guardrail image into the first detection model. If the detection result is that the railway guardrail has a first defect, the detection is completed, includes: a vehicle-mounted inspection camera collects the railway guardrail image in real time, inputs the collected railway guardrail image into the first detection model, and the first detection model detects whether the railway guardrail has collapsed. If the railway guardrail has collapsed, it outputs that the railway guardrail has a first defect, and the detection is completed; If the detection result is that the railway guardrail is normal, then input the real-time collected railway guardrail image into the second detection model for detection, and detect the second defect of the railway guardrail, includes: if the detection result is that the railway guardrail is normal, crop the real-time collected railway guardrail image, and input the cropped railway guardrail image into the second detection model for detection to obtain the second defect of the railway guardrail.

2. The railway guardrail defect detection method according to claim 1, characterized in that The second defect includes: the railway guardrail is fractured or damaged.

3. The railway guardrail defect detection method according to claim 2, characterized in that The second defect also includes: the column cap of the railway guardrail is missing.

4. A railway guardrail defect detection device, characterized in that, Including: A first main module for collecting the first railway guardrail defect image and the second railway guardrail defect image; The acquisition of the first railway guardrail defect images includes: using an integrated industrial camera and a positioning sensor to acquire railway guardrail images and corresponding railway guardrail position information, and taking the collapsed railway guardrail images as the first railway guardrail defect images; the acquisition of the second railway guardrail defect images includes: taking the fracture images, damage images, and missing column cap images of the non-collapsed railway guardrails as the second railway guardrail defect images; The second main module is used to divide the first railway guardrail defect images into a first railway guardrail defect training data set and a first railway guardrail defect test data set, train a deep convolutional neural network using the first railway guardrail defect training data set to obtain a first detection model, and test the first detection model using the first railway guardrail defect test data set. If the output result of the first detection model matches the expected result, it is determined that the first detection model is a valid model; The third main module is used to divide the second railway guardrail defect images into a second railway guardrail defect training data set and a second railway guardrail defect test data set, train a deep convolutional neural network using the second railway guardrail defect training data set to obtain a second detection model, and test the second detection model using the second railway guardrail defect test data set. If the output result of the second detection model matches the expected result, it is determined that the second detection model is a valid model; The fourth main module is used to input the real-time acquired railway guardrail images into the first detection model. If the detection result is that the railway guardrail has a first defect, the detection is completed, including: the on-vehicle inspection camera real-time acquires railway guardrail images, inputs the acquired railway guardrail images into the first detection model, and the first detection model detects whether the railway guardrail has collapsed. If the railway guardrail has collapsed, it outputs that the railway guardrail has a first defect, and the detection is completed; if the detection result is that the railway guardrail is normal, the real-time acquired railway guardrail images are input into the second detection model for detection to obtain the second defect of the railway guardrail, including: if the detection result is that the railway guardrail is normal, the real-time acquired railway guardrail images are cropped, and the cropped railway guardrail images are input into the second detection model for detection to obtain the second defect of the railway guardrail.

5. An electronic device, characterized in that, Including: At least one processor, at least one memory, and a communication interface; wherein, The processor, the memory, and the communication interface communicate with each other; The memory stores program instructions executable by the processor, and the processor calls the program instructions to execute the method according to any one of claims 1 to 3.

6. A non-transitory computer-readable storage medium, characterized in that, Stores computer instructions, and the computer instructions cause the computer to execute the method according to any one of claims 1 to 3.

Images