Automatic bridge defect detection system based on machine vision

Abstract

The invention relates to an automatic bridge defect detection system based on machine vision, an inspection guide rail is arranged at the bottom of a bridge of each inspection section, part of the inspection guide rail is erected and distributed along a joint of the bridge, each inspection vehicle is arranged on the inspection guide rail of each inspection section, and the inspection vehicles can move along the inspection guide rails. A camera is arranged on the inspection vehicle and is used for acquiring a seam image of the bridge; a section marking module, an identification module, a marking module and an inspection instruction module are configured on the inspection cloud server, and the inspection cloud server is used for being in communication connection with an inspection vehicle; the inspection instruction module is used for issuing a control instruction to the inspection vehicle, the section module is used for inputting the position of each inspection section, the identification module is used for carrying out image identification based on the obtained seam image to identify the crack position, and the marking module is used for marking the position of the crack position in the inspection section. The method has the effect of improving the detection efficiency.

Description

Technical Field

[0001] This application relates to the technical field of bridge construction, and in particular to an automated bridge defect detection system based on machine vision. Background Technology

[0002] Automated bridge defect detection is a key task in bridge maintenance and safety management. With the increasing number and aging of bridges, and their crucial role in transportation systems, accurate and efficient bridge defect detection has become paramount. Bridge defects may include cracks, corrosion, detachment, and structural displacement, all of which can threaten the safety and reliability of the bridge. Therefore, early detection and identification of these defects are essential for maintenance and repair efforts.

[0003] In bridge inspection, it is necessary to inspect the joints between bridges. However, the bottom of the bridge is difficult to reach. Traditional inspection methods often use truss or folding arm bridge inspection vehicles. A platform carrying inspection workers is sent to the bottom of the bridge by a hydraulically driven robotic arm. The workers then manually use handheld flaw detection equipment to inspect the joints between the bridges. This method has low inspection efficiency and therefore needs improvement. Summary of the Invention

[0004] To improve detection efficiency, this application provides an automated bridge defect detection system based on machine vision.

[0005] The automated bridge defect detection system based on machine vision provided in this application adopts the following technical solution: an automated bridge defect detection system based on machine vision, including an inspection cloud server, an inspection guide rail and an inspection vehicle; The inspection guide rail is installed at the bottom of the bridge in each inspection section. Sections of the inspection guide rail are installed and distributed along the joints of the bridge. Several consecutive joints of the bridge are defined as inspection sections. Multiple inspection vehicles are provided, and each inspection vehicle is installed on the inspection guide rail of each inspection section. The inspection vehicle can move along the inspection guide rail. The inspection vehicle is equipped with a camera, which is used to collect images of the bridge joints. The inspection cloud server is equipped with a section module, an identification module, a marking module, and an inspection instruction module. The inspection cloud server is used to communicate with the inspection vehicle. The inspection instruction module is used to issue control instructions to the inspection vehicle, the section module is used to record the location of each inspection section, the recognition module performs image recognition based on the acquired joint image to identify the crack location, and the marking module is used to mark the location of the crack in the inspection section.

[0006] Preferably, the inspection guide rail includes several joint guide rails, several connecting guide rails, and several connecting guide rails. The joint guide rails, the connecting guide rails, and the connecting guide rails are fixed to the bottom of the bridge by hangers. Each joint guide rail is erected along the joint of the bridge. The connecting guide rails are respectively set at both ends of the joint guide rails. The connecting guide rails respectively connect the connecting guide rail at the tail end of the joint guide rail and the connecting guide rail at the head end of the adjacent joint guide rail. The connecting guide rail can rotate to mate with the seam guide rail and the connecting guide rail respectively.

[0007] Preferably, a clearance space is formed in the middle of the joint guide rail, the connecting guide rail and the connecting guide rail, and the inspection vehicle is carried in the clearance space of the joint guide rail, the connecting guide rail and the connecting guide rail.

[0008] Preferably, a connecting shaft is fixed at the bottom of the bridge, a connecting plate is fixed at the upper end of the connecting guide rod, a rotating seat is provided at the center of the connecting plate, the lower end of the connecting shaft is rotatably mounted on the rotating seat, and a driving part for driving the connecting plate to rotate is provided on the connecting shaft, the driving part being controlled by the inspection command module.

[0009] Preferably, the drive unit includes a servo motor, which is fixed on the connecting shaft. A gear ring is fixed on the connecting plate and is coaxial with the connecting shaft. A drive gear that meshes with the gear ring is provided on the output shaft of the servo motor. The servo motor is powered on and connected to a wireless communication module that communicates with the inspection cloud server.

[0010] Preferably, the inspection vehicle includes a vehicle body, a controller is installed inside the vehicle body, a travel axle is rotatably mounted on both sides of the vehicle body, a travel wheel is installed on each travel axle, a travel drive assembly for driving the travel axle to rotate synchronously is installed inside the vehicle body, a power module for providing electrical energy is installed inside the vehicle body, the controller is connected to the travel drive assembly, the camera is connected to the controller, and a wireless communicator is connected to the controller; The vehicle body is equipped with a mounting rod, and the camera is mounted on the mounting rod so as to face the joint of the bridge.

[0011] Preferably, the mounting rod is slidably mounted on the vehicle body in a vertical direction, and the mounting rod has lifting racks distributed along its length. A lifting motor is mounted on the vehicle body, and a lifting gear that meshes with the rack is mounted on the output shaft of the lifting motor.

[0012] Preferably, an electric pan-tilt unit is mounted on the mounting rod, the camera is mounted on the electric pan-tilt unit, and an infrared sensor is provided on one side of the camera. The infrared sensor is used to detect the position of the bridge so as to control the rotation of the electric pan-tilt unit to keep the camera always facing the joint of the bridge.

[0013] Preferably, the travel drive assembly includes a travel drive motor, a travel drive shaft, a first travel drive bevel gear, and a second travel drive bevel gear. The second travel drive bevel gear is mounted on the travel shaft. The travel drive motor is fixed in the vehicle body. The travel drive shaft is connected to the output shaft of the travel drive motor via a coupling. The first travel drive bevel gear is coaxially mounted on the travel drive shaft and meshes with the second travel drive gear.

[0014] Preferably, a power supply bracket is arranged at the bottom of the bridge along the distribution direction of the inspection guide rail. A conductive metal sheet is provided on the bottom surface of the power supply bracket. A power receiving bracket is provided on the vehicle body. A power receiving metal sheet for contacting the conductive metal sheet is provided on the power receiving bracket. The power receiving metal sheet is connected to the power module through a wire.

[0015] In summary, this application includes at least one of the following beneficial technical effects: This application utilizes an inspection guide rail installed at the bottom of the bridge. An inspection vehicle moves along the guide rail to capture images of the bridge joints. These images are then used for image recognition to detect cracks in the joints. Once cracks are found, their locations are marked, facilitating subsequent maintenance by staff. This application achieves automated bridge defect detection through machine vision using cameras in the inspection vehicle, thereby improving inspection efficiency. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the installation of inspection guide rails on a bridge.

[0017] Figure 2 This is a schematic diagram of the layout of one embodiment of the inspection guide rail.

[0018] Figure 3 This is a schematic diagram of another embodiment of the inspection guide rail.

[0019] Figure 4 This is a system block diagram of the inspection cloud server and inspection vehicle.

[0020] Figure 5 This is a schematic diagram of the connecting guide rail structure.

[0021] Figure 6 This is a schematic diagram of the inspection vehicle's installation on the inspection guide rail.

[0022] Figure 7 This is a structural diagram of the inspection vehicle.

[0023] Explanation of reference numerals in the attached drawings: 1. Inspection guide rail; 101. Joint guide rail; 102. Connecting guide rail; 103. Connecting guide rail; 104. Clearance space; 105. Hanging rod; 106. Connecting shaft; 107. Connecting plate; 108. Rotating seat; 109. Servo motor; 110. Gear ring; 111. Drive gear; 2. Inspection vehicle; 21. Vehicle body; 22. Traveling shaft; 23. Traveling wheel; 24. Traveling drive assembly; 241. Traveling drive motor; 242. Traveling drive shaft; 243. First traveling drive bevel gear; 244. Second traveling drive bevel gear; 25. Mounting rod; 26. Electric pan-tilt head; 27. Infrared sensor; 28. Lifting rack; 29. ​​Lifting motor; 30. Lifting gear; 3. Camera; 4. Power supply bracket; 5. Conductive metal sheet; 6. Power receiving bracket; 7. Power receiving metal sheet. Detailed Implementation

[0024] The following is in conjunction with the appendix Figure 1-7 This application will be described in further detail.

[0025] An automated bridge defect detection system based on machine vision, referring to... Figure 1 As shown, it includes an inspection cloud server, inspection guide rail 1, and inspection vehicle 2.

[0026] A series of consecutive joints in a bridge are defined as inspection sections. A bridge is constructed by splicing together sections of concrete beams and slabs. The joints between adjacent beams and slabs are called bridge joints. A series of consecutive joints refers to the joints formed between adjacent beams and slabs, and these joints are arranged parallel to each other. In this embodiment, an inspection section is defined as 200 meters long. When the bridge is 1000 meters long, there will be 5 inspection sections.

[0027] Inspection guide rail 1 is set at the bottom of the bridge in each inspection section, that is, each inspection section is equipped with inspection guide rail 1, and each inspection guide rail 1 is equipped with an inspection vehicle 2. Sections of inspection guide rail 1 are erected and distributed along the joints of the bridge.

[0028] Specifically, the inspection guide rail 1 includes several joint guide rails 101, several connecting guide rails 102, and several connecting guide rails 103. The joint guide rails 101, connecting guide rails 102, and connecting guide rails 103 are fixed to the bottom of the bridge by hangers 105. Each joint guide rail 101 is erected along the joint of the bridge. The connecting guide rails 103 are respectively set at both ends of the joint guide rails 101. The connecting guide rails 103 are located near the two sides of the bridge. The connecting guide rails 102 are set at the edges of the bridge. The connecting guide rails 102 are respectively connected to the connecting guide rails 103 at the tail end of the joint guide rails 101 and the connecting guide rails 103 at the head end of the adjacent joint guide rails 101. The connecting guide rails 103 can rotate to connect to the joint guide rails 101 and the connecting guide rails 102 respectively.

[0029] In one embodiment, refer to Figure 2 As shown, the inspection guide rail 1 has a concave-convex orientation. The beginning and end of the adjacent joint guide rails 101 are interchanged. The wiring guide rails are distributed along the joints of the bridge. The connecting guide rail 102 is perpendicular to the wiring guide rail. One end of the connecting guide rail 102 is connected to the connecting guide rail 103 at the end of the joint guide rail 101, and the other end of the connecting guide rail 102 is connected to the connecting guide rail 103 at the beginning of the adjacent joint guide rail 101.

[0030] In another embodiment, refer to Figure 3 As shown, the inspection guide rail 1 is zigzag in shape, and the beginning and end of the adjacent joint guide rail 101 are on the same straight line. The wiring guide rail is distributed along the joint of the bridge. The connecting guide rail 102 is perpendicular to the wiring guide rail. One end of the connecting guide rail 102 is connected to the connecting guide rail 103 at the end of the joint guide rail 101, and the other end of the connecting guide rail 102 is connected to the connecting guide rail 103 at the beginning of the adjacent joint guide rail 101.

[0031] The joint guide rail 101, connecting guide rail 102 and connecting guide rail 103 have the same structure. A clearance space 104 is formed in the middle of the joint guide rail 101, connecting guide rail 102 and connecting guide rail 103. The inspection vehicle 2 is carried in the clearance space 104 of the joint guide rail 101, connecting guide rail 102 and connecting guide rail 103 to move along the joint guide rail 101, connecting guide rail 102 and connecting guide rail 103.

[0032] The structure of the connecting guide rail 103 is described below.

[0033] Reference Figure 4 , Figure 5 and Figure 6As shown, a connecting shaft 106 is fixed at the bottom of the bridge, and a connecting plate 107 is fixed at the upper end of the suspension rod 105 connecting the guide rail 103. A rotating seat 108 is set at the center of the connecting plate 107, and the lower end of the connecting shaft 106 is rotatably mounted on the rotating seat 108. A drive unit for driving the connecting plate 107 to rotate is set on the connecting shaft 106, and the drive unit is controlled by the inspection cloud server. The drive unit includes a servo motor 109, which is fixed on the connecting shaft 106. A gear ring 110 is fixed on the connecting plate 107, and the gear ring 110 is coaxially arranged with the connecting shaft 106. A drive gear 111 that meshes with the gear ring 110 is set on the output shaft of the servo motor 109. A wireless communication module that communicates with the inspection cloud server is electrically connected to the servo motor 109.

[0034] The connecting shaft 106 is located at the intersection of the extended connecting lines of the joint guide rail 101 and the connecting guide rail 102. By designing the lengths of the joint guide rail 101, the connecting guide rail 102 and the connecting guide rail 103, initially, the connecting guide rail 103 and the joint guide rail 101 are on the same straight line. One end of the connecting guide rail 103 is connected to the tail end of the joint guide rail 101. The inspection vehicle 2 can move along the joint guide rail 101 and move onto the connecting guide rail 103.

[0035] When the inspection vehicle 2 needs to move from the connecting guide rail 103 to the connecting guide rail 102, the connecting guide rail 103 is rotated so that it and the connecting guide rail 102 are on the same straight line, with the other end of the connecting guide rail 103 connecting to one end of the connecting guide rail 102. This allows the inspection vehicle 2 to move from the connecting guide rail 103 to the connecting guide rail 102. Similarly, the inspection vehicle 2 can subsequently move from the connecting guide rail 102 to the adjacent joint guide rail 101, thus enabling the inspection vehicle 2 to move along the inspection guide rail 1 and complete the inspection of one section of the bridge.

[0036] Multiple inspection vehicles 2 are set up, and each inspection vehicle 2 is set up on the inspection guide rail 1 of each inspection section. The inspection vehicle 2 can move along the inspection guide rail 1. The inspection vehicle 2 is equipped with a camera 3, which is used to collect images of the bridge joints.

[0037] The inspection cloud server is equipped with a section module, an identification module, a marking module, and an inspection instruction module. The inspection cloud server is used to communicate with the inspection vehicle 2.

[0038] The inspection command module is used to issue control commands to the inspection vehicle 2. The inspection command module is also used to issue control commands to the drive unit of the connecting guide rail 103.

[0039] The section module is used to input the location and number each inspection section. One number represents the corresponding inspection section. The latitude and longitude of each inspection section are also input, that is, the position of inspection guide rail 1 is recorded. The latitude and longitude of joint guide rail 101, connecting guide rail 103 and connecting guide rail 102 in inspection guide rail 1 are all input, so that the position of each point in each inspection guide rail 1 can be obtained.

[0040] The recognition module is used to acquire seam images captured by camera 3. Based on the acquired seam images, the recognition module performs image recognition to identify the location of the crack.

[0041] After the identification module identifies the location of the crack in the joint image, the location where the inspection vehicle 2 collected the joint image can be located based on the acquired joint image. In this way, the specific location of the inspection guide rail 1 in the inspection section can be obtained. The marking module is used to mark the location of the crack in the inspection section.

[0042] The inspection cloud server can control the inspection vehicle 2 to move or stop, and can control the connecting guide rail 103 to rotate. The inspection vehicle 2 is equipped with a positioning module, which can obtain the position of the inspection vehicle 2 and upload it to the inspection cloud server.

[0043] When the inspection vehicle 2 is inspecting a section, it first moves along the joint guide rail 101 of the inspection guide rail 1. The camera 3 of the inspection vehicle 2 captures images of the bridge joints in real time and uploads them to the inspection cloud server for image recognition.

[0044] When the inspection vehicle 2 moves to the connecting guide rail 103, the inspection cloud server detects the position of the inspection vehicle 2 and sends a control command to the drive unit of the connecting guide rail 103. The connecting guide rail 103 rotates to dock with the connecting guide rail 102. Since the drive unit of the connecting guide rail 103 is controlled by a servo motor 109, the rotation position of the connecting guide rail 103 can be confirmed, and the connecting guide rail 103 can be accurately rotated and docked.

[0045] At this time, the inspection vehicle 2 can move from the connecting guide rail 103 to the connecting guide rail 102, and then, based on the above method, it can enter the connecting guide rail 103 from the connecting guide rail 102 and enter the adjacent joint guide rail 101 to collect the joint image of the next bridge.

[0046] The inspection vehicle 2 continues until it completes the acquisition of joint images of all bridge joints in an inspection section, as well as the identification and location of cracks in the joint images.

[0047] The structure of inspection vehicle 2 is described below.

[0048] Reference Figure 6 and Figure 7 As shown, the inspection vehicle 2 includes a vehicle body 21, and a controller is installed inside the vehicle body 21. The controller is connected to a wireless communicator and a positioning module. The positioning module adopts a Beidou + GPS dual-mode antenna positioning module, which can improve the positioning accuracy to the centimeter level (1-2cm). The wireless communicator is used to communicate with the inspection cloud server for data transmission.

[0049] The bottom sides of the vehicle body 21 are respectively rotatably mounted with a travel axle 22, and each travel axle 22 is mounted with a travel wheel 23. The travel wheel 23 is horizontally mounted at the lower end of the travel axle 22, and the two travel wheels 23 are respectively engaged with the joint guide rail 101, the connecting guide rail 103 and the connecting guide rail 102 located on both sides.

[0050] The vehicle body 21 is equipped with a travel drive assembly 24 for driving the travel shaft 22 to rotate synchronously. The travel drive assembly 24 includes a travel drive motor 241, a travel drive shaft 242, a first travel drive bevel gear 243, and a second travel drive bevel gear 244. The second travel drive bevel gear 244 is mounted on the travel shaft 22. The travel drive motor 241 is fixed inside the vehicle body 21. The travel drive shaft 242 is connected to the output shaft of the travel drive motor 241 through a coupling. The first travel drive bevel gear 243 is coaxially mounted on the travel drive shaft 242 and meshes with the second travel drive bevel gear 241. A controller is connected to the travel drive assembly 24, specifically electrically connected to the travel drive motor 241. The controller is used to control the operation of the travel drive motor 241.

[0051] The vehicle body 21 is equipped with a power module for providing electrical energy. The power module uses a lithium battery and is used to provide power to all electrical components of the inspection vehicle 2, such as the controller, the walking drive component 24, and the camera 3.

[0052] A mounting rod 25 is installed on the vehicle body 21. The mounting rod 25 is set in a vertical direction. The camera 3 is installed on the mounting rod 25 so as to face the joint of the bridge. The camera 3 is electrically connected to the controller. The joint image captured by the camera 3 is transmitted to the inspection cloud platform after passing through the controller.

[0053] An electric pan-tilt head 26 is mounted on the mounting rod 25, and a camera 3 is mounted on the electric pan-tilt head 26. An infrared sensor 27 is set on one side of the camera 3. The infrared sensor 27 is used to detect the position of the bridge to control the rotation of the electric pan-tilt head 26 so that the camera 3 is always facing the joint of the bridge. The electric pan-tilt head 26 and the infrared sensor 27 are electrically connected to the controller. The infrared sensor 27 includes an infrared transmitter and an infrared receiver. The infrared transmitter emits infrared light towards the bridge, and the infrared receiver receives the infrared light reflected by the bridge. In this way, the direction of the bridge can be located, and then the pan-tilt head is used to control the rotation of the camera 3 so that the camera 3 can always face the bridge to capture images of the bridge joint.

[0054] It is worth noting that the mounting rod 25 is slidably mounted on the vehicle body 21 in a vertical direction. Since the joint guide rail 101, the connecting guide rail 102 and the connecting guide rail 103 form a clearance space 104 in the middle, the inspection guide rail 1 will not interfere with the up and down sliding of the mounting rod 25.

[0055] The mounting rod 25 has a lifting rack 28 distributed along its length. A lifting motor 29 is mounted on the vehicle body 21. A lifting gear 30 that meshes with the rack is mounted on the output shaft of the lifting motor 29. A controller is electrically connected to the lifting motor 29 and is used to control the operation of the lifting motor 29.

[0056] In order to power the power module, in this embodiment, a power supply bracket 4 is arranged at the bottom of the bridge along the distribution direction of the inspection guide rail 1. A conductive metal sheet 5 is provided on the bottom surface of the power supply bracket 4. A power receiving bracket 6 is provided on the vehicle body 21. A power receiving metal sheet 7 is provided on the power receiving bracket 6 for contacting the conductive metal sheet 5. The power receiving metal sheet 7 is connected to the power module through a wire. By connecting the conductive metal sheet 5 to the mains power, and then through the contact between the conductive metal sheet 5 and the power receiving metal sheet 7, the power module is charged, ensuring that the power module has sufficient power.

[0057] The following describes the workflow of the inspection vehicle 2, the inspection guide rail 1, and the inspection cloud server.

[0058] The section module of the inspection cloud server is also equipped with inspection points for inspecting the joints on both sides of the bridge. These inspection points are located on the joint guide rail 101 at the initial position and the connecting guide rail 103 at the subsequent position.

[0059] When the inspection vehicle 2 stops at the initial position of the inspection guide rail 1, the initial position is defined as the detection point. The inspection cloud server sends control commands to the inspection vehicle 2, and the inspection vehicle 2 controls the lifting motor 29 to move, so that the installation rod 25 is raised. The camera 3 can collect the joint image on the side of the bridge and perform image recognition on the joint image to see if there is a crack. If there is a crack, the crack location will be marked. After the inspection of that side of the bridge is completed, the mounting rod 25 is lowered, and the camera 3 is lowered to below the bridge. During the descent of the mounting rod 25, the camera 3, under the action of the electric pan-tilt head 26, always faces the bridge joint. Subsequently, the inspection cloud server sends control commands to the inspection vehicle 2, which moves along the joint guide rail 101. The camera 3 captures images of the joint at the bottom of the bridge, and the inspection cloud server performs image recognition on the joint image to identify whether there are cracks. If cracks are found, the location of the cracks is marked. When the inspection vehicle 2 moves to the detection point on the connecting guide rail 103 at the end of the joint guide rail 101, the connecting guide rail 103 is located on the other side of the bridge. Before the connecting guide rail 103 is switched, the inspection cloud server sends control commands to the inspection vehicle. 2. The inspection vehicle 2 controls the lifting motor 29 to raise the mounting rod 25. The camera 3 can then capture images of the joints on the other side of the bridge and perform image recognition to check for cracks. If cracks are found, their locations will be marked. After the images of the joints on the other side of the bridge are captured, the connecting guide rail 103 is switched. The inspection vehicle 2 moves from the connecting guide rail 103 to the connecting guide rail 102, and then from the connecting guide rail 102 to the detection point on the adjacent connecting guide rail 103 to inspect the adjacent side joints of the bridge. After the side joint inspection is completed, the inspection vehicle 2 moves to the joint guide rail 101 to inspect the adjacent joints of the bridge. This process is repeated until the joint inspection of one inspection section is completed.

[0060] In this application, the camera 3 of the inspection vehicle 2 can not only collect images of the joints at the bottom of the bridge, but also images of the joints on both sides of the bridge. Thus, when there is a difference in elevation between the top and bottom and the left and right sides of the bridge, defect detection can be performed through the joint images at the bottom and the joint images on the sides. The detection is highly automated and efficient.

[0061] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. An automated bridge defect detection system based on machine vision, characterized in that, It includes a patrol cloud server, a patrol guide rail (1), and a patrol vehicle (2); The inspection guide rail (1) is set at the bottom of the bridge in each inspection section. A portion of the inspection guide rail (1) is erected and distributed along the joints of the bridge. Several consecutive joints of the bridge are defined as inspection sections. Multiple inspection vehicles (2) are provided. Each inspection vehicle (2) is set on the inspection guide rail (1) of each inspection section. The inspection vehicle (2) can move along the inspection guide rail (1). The inspection vehicle (2) is equipped with a camera (3) for collecting images of the bridge joints. The inspection cloud server is equipped with a section module, an identification module, a marking module and an inspection instruction module. The inspection cloud server is used to communicate with the inspection vehicle (2). The inspection instruction module is used to issue control instructions to the inspection vehicle (2), the section module is used to record the location of each inspection section, the identification module performs image recognition based on the acquired joint image to identify the crack location, and the marking module is used to mark the location of the crack in the inspection section.

2. The automated bridge defect detection system based on machine vision according to claim 1, characterized in that, The inspection guide rail (1) includes several joint guide rails (101), several connecting guide rails (102), and several connecting guide rails (103). The joint guide rails (101), the connecting guide rails (102), and the connecting guide rails (103) are fixed to the bottom of the bridge by a hanger (105). Each joint guide rail (101) is erected along the joint of the bridge. The connecting guide rails (103) are respectively set at both ends of the joint guide rails (101). The connecting guide rails (102) are respectively connected to the connecting guide rails (103) at the tail end of the joint guide rails (101) and the connecting guide rails (103) at the head end of the adjacent joint guide rails (101). The connecting guide rail (103) can rotate to engage with the joint guide rail (101) and the connecting guide rail (102) respectively.

3. The automated bridge defect detection system based on machine vision according to claim 2, characterized in that, A clearance space (104) is formed in the middle of the joint guide rail (101), the connecting guide rail (102) and the connecting guide rail (103), and the inspection vehicle (2) is carried in the clearance space (104) of the joint guide rail (101), the connecting guide rail (102) and the connecting guide rail (103).

4. The automated bridge defect detection system based on machine vision according to claim 2, characterized in that, A connecting shaft (106) is fixed at the bottom of the bridge. A connecting plate (107) is fixed at the upper end of the rod (105) of the connecting guide rail (103). A rotating seat (108) is provided at the center of the connecting plate (107). The lower end of the connecting shaft (106) is rotatably mounted on the rotating seat (108). A driving part for driving the connecting plate (107) to rotate is provided on the connecting shaft (106). The driving part is controlled by the inspection command module.

5. The automated bridge defect detection system based on machine vision according to claim 4, characterized in that, The drive unit includes a servo motor (109), which is fixed on the connecting shaft (106). A gear ring (110) is fixed on the connecting plate (107). The gear ring (110) is coaxially arranged with the connecting shaft (106). A drive gear (111) that meshes with the gear ring (110) is provided on the output shaft of the servo motor (109). A wireless communication module that communicates with the inspection cloud server is electrically connected to the servo motor (109).

6. The automated bridge defect detection system based on machine vision according to claim 1, characterized in that, The inspection vehicle (2) includes a vehicle body (21), a controller is installed inside the vehicle body (21), a walking shaft (22) is rotatably installed on both sides of the vehicle body (21), a walking wheel (23) is installed on each of the walking shafts (22), a walking drive assembly (24) is installed inside the vehicle body (21) to drive the walking shafts (22) to rotate synchronously, a power module for providing electrical energy is installed inside the vehicle body (21), the controller is connected to the walking drive assembly (24), the camera (3) is connected to the controller, and a wireless communicator is connected to the controller; A mounting rod (25) is provided on the vehicle body (21), and the camera (3) is mounted on the mounting rod (25) facing the joint of the bridge.

7. The automated bridge defect detection system based on machine vision according to claim 6, characterized in that, The mounting rod (25) is slidably mounted on the vehicle body (21) in the vertical direction. The mounting rod (25) has lifting racks (28) distributed along its length. The vehicle body (21) is equipped with a lifting motor (29). The output shaft of the lifting motor (29) is equipped with a lifting gear (30) that meshes with the rack.

8. The automated bridge defect detection system based on machine vision according to claim 7, characterized in that, An electric gimbal (26) is mounted on the mounting rod (25), and the camera (3) is mounted on the electric gimbal (26). An infrared sensor (27) is provided on one side of the camera (3). The infrared sensor (27) is used to detect the position of the bridge so as to control the rotation of the electric gimbal (26) to keep the camera (3) facing the joint of the bridge.

9. The automated bridge defect detection system based on machine vision according to claim 6, characterized in that, The walking drive assembly (24) includes a walking drive motor (241), a walking drive shaft (242), a first walking drive bevel gear (243), and a second walking drive bevel gear (244). The second walking drive bevel gear (244) is mounted on the walking shaft (22). The walking drive motor (241) is fixed inside the vehicle body (21). The walking drive shaft (242) is connected to the output shaft of the walking drive motor (241) via a coupling. The first walking drive bevel gear (243) is coaxially mounted on the walking drive shaft (242). The first walking drive bevel gear (243) meshes with the second walking drive gear (111).

10. The automated bridge defect detection system based on machine vision according to claim 6, characterized in that, A power supply bracket (4) is arranged at the bottom of the bridge along the distribution direction of the inspection guide rail (1). A conductive metal sheet (5) is provided on the bottom surface of the power supply bracket (4). A power receiving bracket (6) is provided on the vehicle body (21). A power receiving metal sheet (7) is provided on the power receiving bracket (6) for contacting the conductive metal sheet (5). The power receiving metal sheet (7) is connected to the power module through a wire.

Images