Steel box girder detection method, system, device and electronic equipment

Abstract

The present application is applicable to the technical field of railway steel box girder detection, and provides a steel box girder detection method, system, device and electronic equipment. The steel box girder detection system provided by the present application includes: an inspection robot placed inside the web of the steel box girder, the web of the girder including a plurality of connected boxes, the inspection robot being used to move from the first box of the plurality of boxes to the last box during the driving process inside the web of the girder in response to the received inspection instruction; an image acquisition device connected to the inspection robot, and used to move to each box driven by the inspection robot and to acquire an internal image of the web of the girder in each box; a communication module connected to the image acquisition device, and used to send all acquired internal images of the web of the girder to a defect recognition server; a defect recognition server connected to the communication module, and used to perform image feature recognition on all internal images of the web of the girder to obtain the web surface information of the girder.

Description

Technical Field

[0001] The present application belongs to the technical field of railway steel box girder detection, and more specifically, to a steel box girder detection method, system, device and electronic equipment. Background Art

[0002] Steel box girder is a commonly used structural form for long-span bridges. Due to its advantages such as light weight, high strength and good integrity, it is widely used as the main beam of long-span bridges such as cable-stayed bridges, suspension bridges and continuous bridges. Under the alternating effects of complex factors such as reciprocating loads (such as frequent train operation), environmental changes (such as alternating changes in temperature and humidity) and sudden disasters (such as earthquakes), high-speed rail box girders will develop tiny fatigue cracks, and even problems such as coating degradation, steel box girder corrosion, fatigue cracking, and component deformation. The development and accumulation of diseases will lead to the continuous deterioration of the service performance of the box girder, directly affecting the durability of the bridge structure. Fatigue fracture may even occur under extreme conditions, which is one of the main sources of danger to high-speed rail operation safety.

[0003] Compared with other forms of main beams, steel box girders have complex structural details and face more arduous maintenance and inspection tasks. If small defects are not discovered and repaired in time, they will greatly increase maintenance costs. If they continue to develop into major defects, they will affect the reliability of the entire bridge and even cause component failure.

[0004] The detection of defects or diseases in the web of steel box girders is still in the theoretical research stage. There is a lack of efficient and intelligent automated detection equipment. The daily maintenance and repair of the web of steel box girders are still mainly based on manual inspections, with a poor working environment and low efficiency. The detection results depend on the subjective judgment of the staff, making it difficult to conduct efficient periodic surveys and unable to conduct comparative analysis on the development trends of diseases. Summary of the invention

[0005] The purpose of the embodiments of the present application is to provide a steel box girder detection method, system, device and electronic equipment, aiming to solve the technical problems in the prior art that daily maintenance and repair of the web of the steel box girder are carried out by manual inspection, the detection results depend on the subjective judgment of the staff, and the detection efficiency is low.

[0006] To achieve the above object, according to a first aspect of the present application, a steel box girder detection system is provided, the steel box girder detection system comprising:

[0007] The inspection robot is placed inside the web of the steel box girder, the web of the girder including a plurality of connected boxes, and the inspection robot is used to move from the first box to the last box among the plurality of boxes in response to the received inspection instruction during the process of traveling inside the web of the girder;

[0008] An image acquisition device, connected to the inspection robot, is used to move to each of the boxes under the drive of the inspection robot and acquire an internal image of the beam web in each of the boxes;

[0009] A communication module, connected to the image acquisition device, for sending all the acquired images of the interior of the beam web to a defect recognition server;

[0010] The defect identification server is connected to the communication module, and is used for performing image feature recognition on all the internal images of the beam web to obtain the beam web surface information of the beam web, wherein the content of the image feature recognition includes at least one of the following: weld cracking, coating shedding, and steel plate rusting.

[0011] Optionally, in a possible implementation manner of the first aspect, the steel box girder detection system further includes:

[0012] The guide rail is pre-arranged inside the web of the steel box girder according to the structural parts to be inspected inside the web of the girder, and is used to drive the inspection robot to travel along the guide rail so that the inspection robot passes through the structural parts to be inspected during the driving process, and each of the boxes contains all or part of the structural parts to be inspected.

[0013] Optionally, in a possible implementation manner of the first aspect, the inspection robot includes:

[0014] A robot body, used for traveling to each of the box chambers along a guide rail inside the web of the steel box girder;

[0015] The inspection platform is connected to the robot body and is used to control the robot body to move along the guide rail in response to the received inspection instructions, and after the robot body moves to each of the boxes, control the mechanical arm connected to the robot body to rotate at multiple angles relative to the robot body, so that the mechanical arm drives the image acquisition device to capture multiple internal images of the beam web in each of the boxes at multiple angles.

[0016] Optionally, in a possible implementation manner of the first aspect, the inspection robot further includes:

[0017] The mechanical arms are arranged on both sides of the robot body, and each of the mechanical arms is provided with a rotatable mechanical wrist at one end away from the robot body, and the image acquisition device is detachably connected to the mechanical wrist.

[0018] Optionally, in a possible implementation manner of the first aspect, the image acquisition device includes:

[0019] A multi-eye visible light imaging module, used to obtain a high-definition image of the beam web surface in each of the chambers;

[0020] An infrared imaging module, used for acquiring a thermal infrared image in each of the chambers;

[0021] The data acquisition and storage module is used to acquire the high-definition image of the beam web surface and the thermal infrared image, and store the high-definition image of the beam web surface and the thermal infrared image as the internal image of the beam web.

[0022] Optionally, in a possible implementation of the first aspect, the beam web surface information includes: defect location information and defect attribute information, and the defect identification server is further used to perform image feature recognition on all internal images of the beam web through a steel box girder defect identification model to obtain the defect location information and defect attribute information inside the beam web.

[0023] Optionally, in a possible implementation of the first aspect, the defect identification server is further used to obtain a sample data set containing typical defects in a sample library of defects in the web of a steel box girder; and train a deep learning model based on the sample data set using a semi-supervised learning method to obtain the steel box girder defect identification model;

[0024] The disease identification server is further used to display the disease location information and the disease attribute information after identifying and obtaining the disease location information and the disease attribute information.

[0025] According to a second aspect of the present application, a steel box girder detection method is provided, the method comprising:

[0026] In response to the received inspection instruction, the inspection robot is controlled to travel inside the web of the steel box girder, wherein the web of the steel box girder includes a plurality of connected boxes, and the traveling process of the inspection robot is to move from the first box to the last box among the plurality of boxes;

[0027] When the inspection robot moves to each of the chambers, controlling an image acquisition device connected to the inspection robot to acquire an internal image of the beam web in each of the chambers;

[0028] All the collected internal images of the beam web are sent to a defect recognition server, so that the defect recognition server performs image feature recognition on all the internal images of the beam web to obtain the beam web surface information of the beam web, wherein the content of the image feature recognition includes at least one of the following: weld cracking, coating shedding, and steel plate rusting.

[0029] The second aspect and any implementation of the second aspect correspond to the first aspect and any implementation of the first aspect, respectively. The technical effects corresponding to the second aspect and any implementation of the second aspect can refer to the technical effects corresponding to the first aspect and any implementation of the first aspect, which will not be repeated here.

[0030] According to a third aspect of the present application, a steel box girder detection device is provided, the device comprising:

[0031] A driving control unit, for controlling the inspection robot to drive inside the web of the steel box girder in response to a received inspection instruction, wherein the web of the steel box girder includes a plurality of connected boxes, and the driving process of the inspection robot is to move from the first box to the last box among the plurality of boxes;

[0032] An acquisition control unit, used for controlling an image acquisition device connected to the inspection robot to acquire an internal image of the beam web in each of the chambers when the inspection robot moves to the chambers;

[0033] An image recognition unit is used to send all the collected images of the interior of the beam web to a defect recognition server, so that the defect recognition server can perform image feature recognition on all the images of the interior of the beam web to obtain the surface information of the beam web, wherein the content of the image feature recognition includes at least one of the following: weld cracking, coating shedding, and steel plate rusting.

[0034] The third aspect and any implementation of the third aspect correspond to the first aspect and any implementation of the first aspect, respectively. The technical effects corresponding to the third aspect and any implementation of the third aspect can refer to the technical effects corresponding to the first aspect and any implementation of the first aspect, which will not be repeated here.

[0035] In a fourth aspect, an embodiment of the present application provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any of the methods described above when executing the computer program.

[0036] In a fifth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and wherein the computer program implements any of the methods described above when executed by a processor.

[0037] In a sixth aspect, an embodiment of the present application provides a computer program product. When the computer program product is run on an electronic device, the electronic device executes any one of the methods described in the first aspect above.

[0038] It can be understood that the beneficial effects of the second to sixth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here.

[0039] The steel box girder inspection scheme provided in the example of this application is adopted, and the web of the steel box girder is divided into multiple boxes. The inspection robot moves inside the web of the steel box girder, that is, between multiple boxes, and drives the image acquisition device to acquire the internal image of the web in each box. In this way, the structural parts inside the web can be fully detected. The automated inspection process reduces manual intervention and improves the inspection efficiency of steel box girders. After that, all the internal images of the web are sent to the disease recognition server through the communication module for image feature recognition to obtain the web appearance information of the web corresponding to the internal image of the web. In this way, the potential disease problems inside the web can be accurately identified, avoiding errors in human subjective judgment, and providing decision support for operation and maintenance personnel, thereby ensuring the safety and reliability of the steel box girder. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0041] Figure 1 A schematic block diagram of a steel box girder detection system provided by the present application is shown;

[0042] Figure 2 A schematic block diagram of an optional steel box girder detection system provided by the present application is shown;

[0043] Figure 3 A schematic block diagram of an optional image acquisition device provided by the present application is shown;

[0044] Figure 4 A schematic block diagram of an optional steel box girder detection system provided by the present application is shown;

[0045] Figure 5 A flowchart for constructing an optional steel box girder defect identification model provided by the present application is shown;

[0046] Figure 6a A schematic diagram of an overall model of an optional inspection robot provided by the present application is shown;

[0047] Figure 6b A schematic diagram showing stress distribution of an optional overall model of an inspection robot provided by the present application is shown;

[0048] Figure 7 A schematic flow chart of an optional steel box girder detection method provided by the present application is shown;

[0049] Figure 8 It is a structural schematic diagram of a steel box girder detection device provided in an embodiment of the present application;

[0050] Fig. 9 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0051] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0052] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0053] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0054] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0055] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0056] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0057] First, some terms in the embodiments of the present application are explained to facilitate understanding by those skilled in the art.

[0058] Steel box girder, also called steel plate box girder, is a common structural form for long-span bridges. It is generally used on bridges with larger spans. It is called steel box girder because it looks like a box. It can be divided into single box girder bridge, double box girder bridge and multi-box girder bridge.

[0059] A graphics processing unit (GPU), also known as a display core, visual processor, or display chip, is a microprocessor that specializes in performing image and graphics-related calculations on personal computers, workstations, game consoles, and some mobile devices (such as tablets, smart phones, etc.).

[0060] The above is a brief introduction to the nouns involved in the embodiments of the present application, which will not be repeated below.

[0061] At present, the space inside the web of a box girder is usually small, the structure is complex, the light is poor, and the air is stagnant, and it is usually done manually. However, manual maintenance not only requires carrying a large amount of maintenance equipment, but also requires maintenance personnel to stay in a confined space for a long time. For the maintenance of large-span bridges, not only is the labor consumption huge, the detection accuracy is poor, the missed detection rate is high, and the detection efficiency is relatively low. It is difficult to achieve efficient periodic census and comparative analysis of the development trend of defects. In addition, some places with relatively high spaces are also difficult to repair.

[0062] With the growth of railway operation scale and the improvement of operation and maintenance requirements, there is a practical need for rapid detection and intelligent identification of the service status of the web of railway steel box girders during operation. It is necessary to develop steel box girder web disease inspection equipment to support periodic and extensive detection and evaluation of steel box girder webs to meet the increasingly urgent railway operation needs.

[0063] In order to solve the above-mentioned technical problems, this application example proposes a solution of intelligent inspection using inspection robots based on linear guide technology combined with machine vision inspection technology, in view of the structural characteristics of the web of steel box girders, inspection requirements, defect characteristics and the narrow space characteristics of the web of the girder, so as to realize the detection and accurate identification of steel box girder defects and replace manual inspection, so as to realize periodic inspection and change analysis of the apparent state of the internal structure and welds of the box girder, guide preventive maintenance, and improve the inspection efficiency and quality of the web of steel box girders.

[0064] According to an embodiment of the present application, a steel box girder detection system is provided. Figure 1 A schematic block diagram of a steel box girder detection system provided by the present application is shown, referring to Figure 1 As shown, by way of example and not limitation, a steel box girder detection system 100 includes:

[0065] The inspection robot 101 is placed inside the web of the steel box girder, which includes a plurality of connected boxes. The inspection robot is used to respond to received inspection instructions and move from the first box to the last box among the plurality of boxes while traveling inside the web of the girder.

[0066] The image acquisition device 102 is connected to the inspection robot 101, and is used to move to each box chamber driven by the inspection robot and acquire the internal image of the beam web in each box chamber.

[0067] The communication module 103 is connected to the image acquisition device 102 and is used to send all the acquired internal images of the beam web to the defect recognition server.

[0068] The defect identification server 104 is connected to the communication module 103 and is used to perform image feature recognition on all the internal images of the beam web to obtain the beam web appearance information of the beam web.

[0069] Optionally, the content of image feature recognition includes at least one of the following: weld cracking, coating shedding, and steel plate rusting.

[0070] The steel box girder detection system provided in the example of this application is particularly suitable for intelligent detection of the narrow space inside the web of the steel box girder. Based on the intelligent inspection and machine vision detection performed by the inspection robot inside the web of the steel box girder, the internal apparent state of the web of the steel box girder can be detected in real time. For example, according to the machine vision technology and the spatial characteristics of the detection object (i.e., the inside of the web of the steel box girder), the image acquisition device can be used to collect the internal apparent state of the web of the steel box girder and form a high-definition image. The standardized image acquisition process provided in the example of this application can meet the needs of high-frequency inspection of the service status of the web of the railway steel box girder during the operation period.

[0071] In one example, the inspection robot responds to the received inspection instructions and travels inside the web of the steel box girder. During the travel inside the web of the girder, the inspection robot moves from the first box chamber to the last box chamber among multiple boxes. Specifically, when the inspection robot moves to each box chamber, the internal image of the web of the girder in each box chamber is collected by an image acquisition device connected to the inspection robot.

[0072] Afterwards, all the internal images of the beam web collected by the image acquisition device are transmitted to the defect identification server set up on the ground through the communication module (for example, WiFi module, 4G module, 5G module, etc.). The defect identification server realizes intelligent identification of the apparent state of the beam web by performing image feature recognition on all the internal images of the beam web, and provides typical defect location information and defect attribute information.

[0073] By adopting the example of this application, the internal surface state of the web of the steel box girder can be quickly obtained through a fast and non-contact image acquisition method. At the same time, based on artificial intelligence technology, online identification of internal defects in the web of the steel box girder can be achieved. For example, it can be identified whether there are defects such as weld cracking, coating shedding, and steel plate rust inside the web of the steel box girder.

[0074] In one example, the image acquisition device 102 (such as an industrial camera) is used to move to each box room driven by the inspection robot, and after moving to each box room, obtain the internal image of the beam web of the box room, and the image includes various details inside the beam web. Before all the internal images of the beam web are transmitted to the disease identification server via the communication module, all the internal images of the beam web can be preprocessed, and the preprocessing steps can include at least one of the following: denoising, contrast enhancement, brightness adjustment, etc.

[0075] It should still be noted that due to the limited mechanical load capacity and power consumption load capacity of the inspection robot, the front-end beam web defect inspection device only has an image acquisition function, and uses a communication module to transmit the collected internal images of the beam web to the defect identification server 104 on the ground for further data analysis.

[0076] Optionally, the above communication module can be a 5G network communication module. It is understandable that 5G has the characteristics of high speed, low latency and large connection, which can meet the needs of large data volume such as high-definition video and images. Its transmission peak theoretical transmission speed can reach 10-20Gbps. Considering the actual signal strength, base station loss, transmission shielding and other factors, the maximum download rate of 5G can basically reach 1Gbps, and the uplink rate is about 100Mbps.

[0077] Assuming the imaging accuracy of the image acquisition device is 1mm, when the robot body horizontally unfolds the mechanical arm once, the image data volume of the image acquisition device is 2000MB, and the upload speed is 200 seconds. The 5G transmission rate meets the timeliness requirements of data transmission. At the same time, the longitudinal walking speed of the robot body and the telescopic efficiency of the mechanical arm can also ensure that the single acquisition image can be uploaded in the gap between the robot body movements.

[0078] After receiving the internal image of the beam web, the defect identification server 104 performs feature extraction on the internal image of the beam web to extract key information from the internal image of the beam web, such as image features of welds, coatings and steel plates, etc. These features may include texture, shape, color, etc.

[0079] Based on the extracted image features, the defect recognition server 104 uses a trained machine learning model or deep learning algorithm to classify and recognize the internal image of the beam web, for example, to determine whether the weld is cracked, whether the coating is peeling off, or whether the steel plate is rusted. Finally, the internal surface information of the beam web indicating the defects or diseases detected inside the beam web is obtained, which can be used for daily maintenance and repair decisions of the steel box girder to ensure the quality and reliability of the steel box girder.

[0080] It should be noted that when the disease identification server performs image recognition processing, the specific image feature recognition algorithm used can be selected according to the system design and actual application. This application example does not specifically limit the image feature recognition algorithm, which is subject to the implementation of this application example.

[0081] The steel box girder inspection scheme provided in the example of this application is adopted, and the web of the steel box girder is divided into multiple boxes. The inspection robot moves inside the web of the steel box girder, that is, between multiple boxes, and drives the image acquisition device to acquire the internal image of the web in each box. In this way, the structural parts inside the web can be fully detected. The automated inspection process reduces manual intervention and improves the inspection efficiency of steel box girders. After that, all the internal images of the web are sent to the disease recognition server through the communication module for image feature recognition to obtain the web appearance information of the web corresponding to the internal image of the web. In this way, the potential disease problems inside the web can be accurately identified, avoiding errors in human subjective judgment, and providing decision support for operation and maintenance personnel, thereby ensuring the safety and reliability of the steel box girder.

[0082] Figure 2 A schematic block diagram of an optional steel box girder detection system provided by the present application is shown, referring to Figure 2 As shown, optionally, in a possible implementation, the above steel box girder detection system further includes:

[0083] The guide rail 105 is pre-placed inside the web of the steel box beam according to the structural parts to be inspected inside the web of the beam, and is used to drive the inspection robot 101 to move along the guide rail so that the inspection robot passes through the structural parts to be inspected during the driving process. Each box contains all or part of the structural parts to be inspected.

[0084] Optionally, according to the structural parts to be inspected inside the web of the beam, a guide rail is pre-arranged inside the web of the steel box beam, and the guide rail is specifically laid between the exit holes and the entrance holes of the two boxes, allowing the inspection robot to move along a specific path inside the web of the beam, ensuring that the inspection robot can pass through each structural part to be inspected, and the inspection robot drives the camera and other image acquisition equipment to collect images of the structural part to be inspected. In this way, the steel box beam inspection system can accurately locate and detect defects or diseases in the web of the beam.

[0085] The inspection robot is guided by the guide rail to move between the boxes. The travel speed can be but is not limited to 5km / h. The inspection robot can quickly inspect the entire beam web according to the predetermined route. The inspection process is automated by setting up the guide rail, which reduces manual intervention, improves detection efficiency, and avoids the waste of time and resources of manual random inspections.

[0086] Still Figure 2 As shown, optionally, in a possible implementation, the inspection robot 101 includes:

[0087] The robot body 1010 is used to travel to each box chamber along the guide rail inside the web of the steel box girder;

[0088] The inspection platform 1011 is connected to the robot body 1010, and is used to control the robot body to move along the guide rail in response to the received inspection instructions, and after the robot body moves to each box chamber, control the mechanical arm connected to the robot body to rotate relative to the robot body at multiple angles, so that the mechanical arm drives the image acquisition device to capture multiple internal images of the beam web in each box chamber at multiple angles.

[0089] The inspection platform 1011 is the control center of the inspection robot 101. In response to the inspection command received, the robot body is controlled to move along the guide rail. After the robot body moves to each chamber, the inspection platform controls the connected mechanical arm to rotate relative to the robot body to change the image acquisition angle of the image acquisition device connected to the mechanical arm.

[0090] like Figure 2As shown, the robot body is located inside the web of the steel box girder, specifically placed above the guide rail, and can travel along the guide rail to each box chamber inside the web of the girder. For example, when traveling to the middle position of the first box chamber, the robot body is controlled to stop, and the robot arm is extended, and the robot wrist is rotated to drive the image acquisition device detachably connected on the robot wrist to perform multi-angle image acquisition. After obtaining the image acquisition result of the first box chamber, this acquisition is completed, and the robot body is controlled to stop when traveling to the middle position of the next box chamber, and the robot arm is extended again, and the robot wrist is rotated to drive the image acquisition device detachably connected on the robot wrist to perform multi-angle image acquisition, until the robot body travels to the last box chamber and completes the image acquisition in the last box chamber. During the inspection process, no direct intervention by staff is required, which improves the degree of automation of the steel box girder inspection system.

[0091] It can be understood that the robot body is the supporting basis and actuator of the steel box girder detection system. Depending on the different sensors it carries and the different application environments, the body structure, internal transmission system and its basic functions will vary.

[0092] In another possible implementation, Figure 2 As shown, the inspection robot 101 also includes:

[0093] The mechanical arms (such as Figure 2 The robot arms 1012 and 1013 are shown in the figure. Each robot arm is provided with a rotatable robot wrist (such as Figure 2 The robot wrist 1014 and the robot wrist 1015 are shown, and the image acquisition device 102 is detachably connected to each robot wrist.

[0094] In the above implementation, the design of the inspection robot is more complex to achieve more comprehensive and accurate internal image acquisition of the beam web. The inspection robot can be provided with two mechanical arms, which are arranged on both sides of the robot body. To improve the stability of the robot body, the mechanical arms can be installed on the center of gravity of the robot body. Each mechanical arm is connected to the robot body at one end close to the robot body, and each mechanical arm is provided with a rotatable mechanical wrist at the end away from the robot body (i.e., the tail end), and the image acquisition device can be detachably connected to the mechanical wrist.

[0095] Optionally, the manipulator wrist can form a detachable connection with the back of the industrial camera to realize the detachable design of the industrial camera. This detachable connection design makes the image acquisition equipment more convenient to replace or maintain, thereby improving the flexibility of the steel box girder detection system.

[0096] By placing the image acquisition device on the rotatable wrist of the robotic arm, the image acquisition device is driven to rotate when the wrist rotates, so that the image acquisition is more comprehensive, and the image acquisition device can capture the details inside the web of the beam at different angles. In this way, the robotic arm can move the image acquisition device at different angles to assist the image acquisition device in obtaining a more comprehensive image of the inside of the web of the beam.

[0097] In the example of this application, through precise control of the robotic arm, it is possible to assist the image acquisition equipment to accurately locate the structural part to be inspected in each box chamber, making the steel box girder inspection system more flexible and intelligent, and capable of acquiring images of the inside of the beam web at multiple angles, so as to facilitate the subsequent identification and evaluation of defects in the inside of the steel box girder web.

[0098] Figure 3 A schematic block diagram of an optional image acquisition device provided by the present application is shown, such as Figure 3 As shown, optionally, in a possible implementation, the image acquisition device 102 includes:

[0099] The multi-eye visible light imaging module 1020 is used to obtain a high-definition image of the beam web surface in each chamber.

[0100] The infrared imaging module 1021 is used to obtain the thermal infrared image of each chamber.

[0101] The data acquisition and storage module 1022 is used to acquire the high-definition image and thermal infrared image of the beam web surface, and store the high-definition image and thermal infrared image of the beam web surface as the internal image of the beam web.

[0102] Optionally, the image acquisition device 102 may be a camera, for example, an industrial camera.

[0103] In the above example of this application, the visible light imaging module uses light within the visible spectrum to capture the details of the beam web surface. A multi-eye visible light imaging module is used to obtain an apparent high-definition image of the beam web in each box chamber. The high-definition image provides clear visual information about the external condition of the beam web, which helps to detect fatigue cracks and other surface defects. It is understandable that infrared imaging can detect the thermal distribution of an object and reveal temperature changes inside the beam web. Therefore, the thermal infrared images collected by the infrared imaging module can help detect potential hot spots, abnormal temperatures and thermal cracks.

[0104] Afterwards, the data acquisition and storage module is used to first integrate different types of images together to form a complete internal image of the beam web, and then all the internal images of the beam web are reliably saved for subsequent analysis and processing. In summary, these modules in the image acquisition device work together to provide comprehensive visual information for fatigue crack detection of steel box girders, thereby realizing a mechanized and efficient detection process.

[0105] In this application example, the image acquisition device can set up multi-eye visible light imaging modules in the front, top, bottom, left and right positions, and then use the mechanical arm at a fixed point to perform multi-angle synchronous shooting in 5 positions. The image accuracy is designed to be 0.5-1mm, which meets the high-definition imaging of typical features such as steel structure cracks, surface coatings, and welds. Further, the apparent image of the three-dimensional state of the measured space can be obtained based on the multi-eye image.

[0106] In addition, refer to Figure 4 A schematic block diagram of an optional steel box girder detection system is shown. In the example of this application, the steel box girder detection system also includes: a power supply module. It is easy to understand that the ideal power supply method is external power supply, which can meet the working mode of unmanned, continuous online, and periodic detection of the steel box girder detection system. However, the internal structure of the beam web is complex and the wiring is inconvenient. In addition, considering factors such as structural safety, when it is difficult to obtain power from the outside, a battery power supply method can be used to achieve automatic charging. Or in other cases, manual charging measures can also be considered.

[0107] Optionally, in a possible implementation, the defect identification server is also used to obtain a sample data set of labeled typical defects contained in a steel box girder web defect sample library; a deep learning model is trained based on the sample data set using a semi-supervised learning method to obtain a steel box girder defect identification model; the defect identification server is also used to display the defect location information and the defect attribute information in a display interface after the defect location information and the defect attribute information are identified.

[0108] Optional, reference Figure 5 The flowchart of constructing an optional steel box girder defect recognition model shown in the figure first pre-processes a plurality of acquired steel box girder web defect images, such as weld cracking images, steel plate rusting images, coating peeling images, etc. The pre-processing step may include at least one of the following: denoising, contrast enhancement, brightness adjustment, etc. After pre-training the multi-modal deep learning model, the steel box girder web defect image is annotated using sample semi-automatic annotation software, and then the annotated steel box girder web defect image is used to generate an annotated typical defect sample data set, and the annotated typical defect sample data set is placed in the steel box girder web defect sample library.

[0109] In the example of this application, the deep learning method is generally used for the fine identification of steel box girder web defects. First, a labeled typical defect sample data set contained in the steel box girder web defect sample library is obtained. Then, based on this sample data set, a semi-supervised learning training method is used to train a multi-scale and multi-morphological deep learning model (for example, a steel box girder defect visual model) for the weld cracking, coating shedding, steel plate corrosion and other defect types of the steel box girder web structural components. Then, the final steel box girder defect recognition model is generated by combining the response recognition speed, recognition accuracy and other indicator requirements.

[0110] Optionally, in a possible implementation, the web surface information of the beam includes: defect location information and defect attribute information. The defect identification server also implements lightweight deployment and reasoning acceleration of the steel box girder defect identification model, and uses the steel box girder defect identification model to perform image feature recognition on all internal images of the web of the beam to obtain the defect location information and defect attribute information inside the web of the beam.

[0111] In the example of this application, the defect recognition server can specifically use a high-performance graphics processor GPU to perform image feature recognition on all images inside the web of the beam through a steel box beam defect recognition model, thereby obtaining the location information and attribute information of the defects inside the web of the beam. Optionally, the steel box beam defect recognition model is obtained by a deep learning model trained in a semi-supervised learning manner, and is used to identify potential fatigue cracks and other surface defects in the web of the steel box beam.

[0112] Based on the deep learning model trained on defect types such as weld cracking, coating shedding, and steel plate rust, image feature recognition is performed on all images inside the web of the beam to automatically extract defect attribute features, including length, width, area, shape, location, mileage, etc., and generate the final model output results, namely the defect location information and defect attribute information inside the web of the beam.

[0113] Intelligent identification and image acquisition are carried out simultaneously for the surface defects of the steel box beam web, such as weld cracking, coating shedding, and steel plate corrosion. The collected internal images of the beam web are transmitted to the defect identification server using the communication module / data transmission module. The robot body collects 2000MB of field image data through a single horizontal deployment of the robotic arm, and the upload speed is 200 seconds, meeting the timeliness requirements of data transmission. At the same time, the longitudinal walking speed of the robot body and the telescopic efficiency of the robotic arm can also ensure that the internal images of the beam web collected once can be uploaded in the gap between the robot body movements, improving the efficiency of on-site inspection.

[0114] To summarize, the example of this application utilizes image detection and intelligent recognition technology to realize rapid detection of the internal state of the railway steel beam bridge belly, meeting the full-range, periodic, rapid, and intelligent inspection of the railway steel box girder web, solving the shortcomings of low efficiency, many missed judgments, and harsh working environment of manual inspection, greatly improving the detection efficiency, and providing automated and intelligent detection equipment and technical means for the detection of steel box girder webs at railway engineering sites.

[0115] Compared to cables and conveyor belts, guide rails do not need to consider the risk of collision damage due to limited space compared to aircraft, and can ensure the stability of the inspection robot when moving. At the same time, the guide rails can carry a large mass, which makes it possible to meet the movement requirements of inspection robots with larger mass. In addition, on the inner plate of the steel box girder, multiple U-shaped ribs are arranged at intervals along the length of the steel box girder, which makes it possible for the track wheels to avoid the risk of being stuck by the U-shaped ribs when moving on the U-shaped ribs compared to rollers.

[0116] In order to accurately simulate the influence of the inspection robot on the detection equipment during its movement, the example of this application uses the finite element method to model the inspection robot. The inspection robot structure established in this application mainly includes: a robot body, an inspection platform, a retractable mechanical arm, a mechanical wrist, etc.

[0117] Refer to Figure 6a An optional overall model of the inspection robot shown in the figure simulates the physical behavior and structural response of the inspection robot in the actual operating environment to ensure that the various components of the inspection robot can work together and efficiently under various working conditions. By carefully simulating the robot body of the inspection robot, the fixing efficiency of the fixing clamp, the dynamic performance of the wheels, the flexibility of the mechanical arm, and the stability of the camera, the overall model helps predict and verify the performance of the inspection robot and ensure the reliability and practicality of its design. Specifically, the material properties, geometry, load conditions and their interactions of each component are reasonably simulated to achieve accurate prediction of the overall behavior of the inspection robot.

[0118] Refer to Figure 6b The schematic diagram of stress distribution of the overall model of an optional inspection robot is shown. Under the action of its own weight, the stress distribution of the inspection robot is relatively uniform. The place where the stress is more concentrated is the connection between the robot body and the mechanical arm. The maximum stress is 111MPa, which is much lower than the tensile strength of aluminum alloy. Therefore, the mechanical structure of the inspection robot provided in the example of this application has a large load-bearing capacity and can maintain structural integrity under extreme conditions. It should be understood that Figure 6b In the equation, S is the stress value in each direction, and Mi ses is the von Mises stress. Specifically, Mi ses is calculated through the stress in each direction. It must be a positive number and cannot indicate the direction. To some extent, it can be considered as the average value of the stress in each direction.

[0119] According to an embodiment of the present application, a steel box girder detection method is provided, which can be applied to the above-mentioned steel box girder detection system. Figure 7 A schematic flow chart of an optional steel box girder detection method provided by the present application is shown, such as Figure 7 As shown, the method includes:

[0120] S701, in response to the received inspection instruction, control the inspection robot to travel inside the web of the steel box girder, wherein the web of the girder includes a plurality of connected boxes, and the traveling process of the inspection robot is to move from the first box chamber to the last box chamber among the plurality of boxes.

[0121] S702, when the inspection robot moves to each chamber, control the image acquisition device connected to the inspection robot to acquire the internal image of the beam web in each chamber.

[0122] S703: Send all the collected internal images of the beam web to the defect recognition server, so that the defect recognition server performs image feature recognition on all the internal images of the beam web to obtain the beam web appearance information of the beam web.

[0123] The content of the above-mentioned image feature recognition includes at least one of the following: weld cracking, coating shedding, and steel plate rusting.

[0124] The steel box girder inspection scheme provided in the example of this application is adopted, and the web of the steel box girder is divided into multiple boxes. The inspection robot moves inside the web of the steel box girder, that is, between multiple boxes, and drives the image acquisition device to acquire the internal image of the web in each box. In this way, the structural parts inside the web can be fully detected. The automated inspection process reduces manual intervention and improves the inspection efficiency of steel box girders. After that, all the internal images of the web are sent to the disease recognition server through the communication module for image feature recognition to obtain the web appearance information of the web corresponding to the internal image of the web. In this way, the potential disease problems inside the web can be accurately identified, avoiding errors in human subjective judgment, and providing decision support for operation and maintenance personnel, thereby ensuring the safety and reliability of the steel box girder.

[0125] It should be noted that the steel box girder detection method and any implementation of the steel box girder detection method correspond to the steel box girder detection system and any implementation of the steel box girder detection system, respectively. The technical effects corresponding to the steel box girder detection method and any implementation of the steel box girder detection method can refer to the technical effects corresponding to the above-mentioned steel box girder detection system and any implementation of the steel box girder detection system, which will not be repeated here.

[0126] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0127] Corresponding to the steel box girder detection method described in the above embodiment, Figure 8is a schematic diagram of the structure of a steel box girder detection device provided in an embodiment of the present application. The device can be implemented as part or all of a computer device by software, hardware, or a combination of both. The computer device can be Fig. 9 Electronic equipment shown.

[0128] Reference Figure 8 , the steel box girder detection device comprises:

[0129] A driving control unit 801 is used to control the inspection robot to drive inside the web of the steel box girder in response to the received inspection instruction, wherein the web of the steel box girder includes a plurality of connected boxes, and the driving process of the inspection robot is to move from the first box to the last box among the plurality of boxes;

[0130] The acquisition control unit 802 is used to control the image acquisition device connected to the inspection robot to acquire the internal image of the beam web in each of the above-mentioned boxes when the inspection robot moves to each of the above-mentioned boxes;

[0131] The image recognition unit 803 is used to send all the collected internal images of the beam web to the defect recognition server, so that the defect recognition server can perform image feature recognition on all the internal images of the beam web to obtain the beam web surface information of the beam web, wherein the content of the image feature recognition includes at least one of the following: weld cracking, coating shedding, and steel plate rusting.

[0132] It should be noted that the steel box girder detection device provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0133] The functional units and modules in the above embodiments may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit, and the above integrated units may be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the protection scope of the embodiments of the present application.

[0134] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0135] An embodiment of the present application further provides an electronic device, the electronic device comprising one or more processors and a memory;

[0136] The memory is coupled to one or more processors, and the memory is used to store computer program codes, the computer program codes include computer instructions, and one or more processors call the computer instructions to enable the electronic device to execute the steel box girder detection method shown above.

[0137] The electronic device may be a mobile phone, a smart screen, a tablet computer, a wearable electronic device, an in-vehicle electronic device, an augmented reality (AR) device, a virtual reality (VR) device, a laptop computer, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), a projector, or a communication device such as a server, a storage device, a base station, or a smart car, etc. The embodiments of the present application do not impose any restrictions on the specific types of electronic devices.

[0138] An embodiment of the present application also provides a computer-readable storage medium, in which computer instructions are stored; when the computer-readable storage medium is run on an electronic device, the electronic device executes the steel box girder detection method shown above.

[0139] The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that includes one or more available media integrated therein. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape), an optical medium, or a semiconductor medium (e.g., a solid state drive (SSD)), etc.

[0140] The embodiment of the present application also provides a computer program product including computer instructions, which, when executed on an electronic device, enables the electronic device to execute the steel box girder detection method shown above.

[0141] The computer storage medium and computer program product provided in the above-mentioned embodiments of the present application are used to execute the method provided above. Therefore, the beneficial effects that can be achieved can refer to the corresponding beneficial effects of the method provided above, and will not be repeated here.

[0142] In the above embodiments, it can also be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transmitted from a website site, computer, server or data center by wired (such as: coaxial cable, optical fiber, data subscriber line (Digital Subscriber Line, DSL)) or wireless (such as: infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can access, or a data storage device such as a server or data center that contains one or more available media integrations. The available medium may be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a digital versatile disc (DVD)), or a semiconductor medium (such as a solid state drive (SSD)), etc.

[0143] Fig. 9 A schematic diagram of the structure of an electronic device provided for an embodiment of the present disclosure, the device may be a mobile phone, a computer, a digital broadcast terminal, a message transceiver, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0144] Device 700 may include one or more of the following components: a processing component 702 , a memory 704 , a power component 706 , a multimedia component 708 , an audio component 710 , an input / output (I / O) interface 712 , a sensor component 714 , and a communication component 716 .

[0145] The processing component 702 generally controls the overall operation of the device 700, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 702 may include one or more processors 720 to execute instructions to complete all or part of the steps of the above-mentioned method. In addition, the processing component 702 may include one or more modules to facilitate the interaction between the processing component 702 and other components. For example, the processing component 702 may include a multimedia module to facilitate the interaction between the multimedia component 708 and the processing component 702.

[0146] The memory 704 is configured to store various types of data to support operations on the device 700. Examples of such data include instructions for any application or method operating on the device 700, contact data, phone book data, messages, pictures, videos, etc. The memory 704 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0147] The power supply component 706 provides power to the various components of the device 700. The power supply component 706 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the device 700.

[0148] The multimedia component 708 includes a screen that provides an output interface between the device 700 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 708 includes a front camera and / or a rear camera. When the device 700 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera may receive external multimedia data. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

[0149] The audio component 710 is configured to output and / or input audio signals. For example, the audio component 710 includes a microphone (MIC), and when the device 700 is in an operating mode, such as a call mode, a recording mode, and a speech recognition mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in the memory 704 or sent via the communication component 716. In some embodiments, the audio component 710 also includes a speaker for outputting audio signals.

[0150] I / O interface 712 provides an interface between processing component 702 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include but are not limited to: a home button, a volume button, a start button, and a lock button.

[0151] The sensor assembly 714 includes one or more sensors for providing various aspects of status assessment for the device 700. For example, the sensor assembly 714 can detect the open / closed state of the device 700, the relative positioning of components, such as the display and keypad of the device 700, the sensor assembly 714 can also detect the position change of the device 700 or a component of the device 700, the presence or absence of user contact with the device 700, the orientation or acceleration / deceleration of the device 700, and the temperature change of the device 700. The sensor assembly 714 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 714 may also include an optical sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 714 may also include an accelerometer, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0152] The communication component 716 is configured to facilitate wired or wireless communication between the device 700 and other devices. The device 700 can access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 716 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 716 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (I rDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0153] In an exemplary embodiment, the apparatus 700 may be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above method.

[0154] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 704 including instructions, and the instructions can be executed by the processor 720 of the device 700 to perform the above method. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0155] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0156] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0157] In the embodiments provided in the present application, it should be understood that the disclosed devices / network equipment and methods can be implemented in other ways. For example, the device / network equipment embodiments described above are merely schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0158] 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 distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0159] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such 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 application, and should all be included in the protection scope of the present application.

Claims

1. A steel box girder detection system, characterized in that: The steel box girder detection system comprises: The inspection robot is placed inside the web of the steel box girder, the web of the girder including a plurality of connected boxes, and the inspection robot is used to move from the first box to the last box among the plurality of boxes in response to the received inspection instruction during the process of traveling inside the web of the girder; An image acquisition device, connected to the inspection robot, is used to move to each of the boxes under the drive of the inspection robot and acquire an internal image of the beam web in each of the boxes; A communication module, connected to the image acquisition device, for sending all the acquired images of the interior of the beam web to a defect recognition server; The disease identification server is connected to the communication module and is used to perform image feature recognition on all the internal images of the beam web to obtain the beam web surface information of the beam web, wherein the content of the image feature recognition includes at least one of the following: weld cracking, coating shedding, and steel plate rusting; The steel box girder inspection system further includes: a guide rail, which is pre-arranged inside the web of the steel box girder according to the structural parts to be inspected inside the web of the girder, and is used to drive the inspection robot to travel along the guide rail, so that the inspection robot passes through the structural parts to be inspected during the driving process, and each of the box chambers contains all or part of the structural parts to be inspected; Among them, the inspection robot includes: a robot body, used to travel along the guide rail inside the web of the steel box girder to each of the box chambers; an inspection platform, connected to the robot body, used to control the robot body to travel along the guide rail in response to the received inspection instructions, and after the robot body travels to each of the box chambers, control the mechanical arm connected to the robot body to rotate relative to the robot body at multiple angles, so that the mechanical arm drives the image acquisition device to capture multiple images of the internal web of each box chamber at multiple angles; the mechanical arms are arranged on both sides of the robot body, and each of the mechanical arms is provided with a rotatable mechanical wrist at the end away from the robot body, and the image acquisition device is detachably connected to the mechanical wrist.

2. The steel box girder detection system according to claim 1, characterized in that: The image acquisition device comprises: A multi-eye visible light imaging module, used to obtain a high-definition image of the beam web surface in each of the chambers; An infrared imaging module, used for acquiring a thermal infrared image in each of the chambers; The data acquisition and storage module is used to acquire the high-definition image of the beam web surface and the thermal infrared image, and store the high-definition image of the beam web surface and the thermal infrared image as the internal image of the beam web.

3. The steel box girder detection system according to claim 1, characterized in that: The beam web surface information includes: defect location information and defect attribute information. The defect identification server is also used to perform image feature recognition on all the internal images of the beam web through a steel box girder defect identification model to obtain the defect location information and defect attribute information inside the beam web.

4. The steel box girder detection system according to claim 3, characterized in that: The disease identification server is further used to obtain a sample data set containing typical diseases in a sample library of web diseases of steel box beams; and to train a deep learning model based on the sample data set using a semi-supervised learning method to obtain the steel box beam disease identification model. The disease identification server is further used to display the disease location information and the disease attribute information after identifying and obtaining the disease location information and the disease attribute information.

5. A steel box girder detection method, characterized in that: The method comprises: In response to the received inspection instruction, the inspection robot is controlled to travel inside the web of the steel box girder, wherein the web of the steel box girder includes a plurality of connected boxes, and the traveling process of the inspection robot is to move from the first box to the last box among the plurality of boxes; When the inspection robot moves to each of the chambers, controlling an image acquisition device connected to the inspection robot to acquire an internal image of the beam web in each of the chambers; Sending all the collected internal images of the beam web to a defect recognition server, so that the defect recognition server performs image feature recognition on all the internal images of the beam web to obtain the beam web surface information of the beam web, wherein the content of the image feature recognition includes at least one of the following: weld cracking, coating shedding, and steel plate rusting; The controlling the inspection robot to travel inside the web of the steel box girder comprises: Controlling the guide rails pre-arranged in the web of the steel box girder according to the structural parts to be inspected in the web, driving the inspection robot to travel along the guide rails, so that the inspection robot passes through the structural parts to be inspected during the travel, wherein each of the boxes contains all or part of the structural parts to be inspected; Among them, controlling the inspection robot to travel inside the web of the steel box girder includes: controlling the robot body of the inspection robot to travel along the guide rail inside the web of the steel box girder to each of the box chambers; controlling the image acquisition device connected to the inspection robot to collect the internal image of the web of the beam in each of the box chambers, including: in response to the received inspection instruction, controlling the robot body to travel along the guide rail, and after the robot body travels to each of the box chambers, controlling the mechanical arm connected to the robot body to rotate at multiple angles relative to the robot body, so that the mechanical arm drives the image acquisition device to collect multiple internal images of the web of the beam in each of the box chambers at multiple angles.

6. A steel box girder detection device, characterized in that: The device comprises: A driving control unit, for controlling the inspection robot to drive inside the web of the steel box girder in response to a received inspection instruction, wherein the web of the steel box girder includes a plurality of connected boxes, and the driving process of the inspection robot is to move from the first box to the last box among the plurality of boxes; An acquisition control unit, used for controlling an image acquisition device connected to the inspection robot to acquire an internal image of the beam web in each of the chambers when the inspection robot moves to the chambers; An image recognition unit is used to send all the collected images of the interior of the beam web to a defect recognition server, so that the defect recognition server performs image feature recognition on all the images of the interior of the beam web to obtain the beam web surface information of the beam web, wherein the content of the image feature recognition includes at least one of the following: weld cracking, coating shedding, and steel plate rusting; The driving control unit is further used to: control the guide rails pre-arranged inside the web of the steel box girder according to the structural parts to be detected inside the web, drive the inspection robot to travel along the guide rails, so that the inspection robot passes through the structural parts to be detected during the driving process, wherein each of the boxes contains all or part of the structural parts to be detected; Among them, the driving control unit is specifically used to: control the robot body of the inspection robot to travel along the guide rail inside the web of the steel box girder to each of the box chambers; the acquisition control unit is specifically used to: respond to the received inspection instructions, control the robot body to travel along the guide rail, and after the robot body travels to each of the box chambers, control the mechanical arm connected to the robot body to rotate relative to the robot body at multiple angles, so that the mechanical arm drives the image acquisition device to collect multiple images of the internal web of each of the box chambers at multiple angles.

7. An electronic 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, the method according to claim 5 is implemented.

Images