On-line surface defect detection device for cast pipe
The automated inspection of cast pipes by online inspection device has solved the problem of low efficiency of manual inspection, realized rapid and comprehensive automatic inspection of surface defects of cast pipes, reduced the labor intensity of workers and improved the inspection efficiency of the production line.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- SAINT GOBAIN PIPELINES CO LTD
- Filing Date
- 2025-05-14
- Publication Date
- 2026-05-08
AI Technical Summary
In the current technology, the detection of defects on the outer surface of cast pipes relies on manual inspection, which is inefficient, labor-intensive for workers, and prone to missed detections, making it difficult to meet the high-efficiency inspection requirements of the production line.
An online surface defect detection device for cast iron pipes is adopted, which includes a pipe rotation unit, an image acquisition unit, and an intelligent recognition unit. The device automatically analyzes the surface images of the pipe fittings using a detection camera and an intelligent recognition unit, and achieves automatic detection by combining a deep learning model.
It has enabled automated and rapid detection of surface defects in cast pipes, reduced the labor intensity of workers, improved detection efficiency, provided timely feedback for production process adjustments, reduced defective products, and achieved full inspection.
Smart Images

Figure CN224216585U_ABST
Abstract
Description
Technical Field
[0001] This utility model relates to the field of pipe fitting inspection technology, specifically to an online surface defect detection device for cast pipes. Background Technology
[0002] After the pipes are centrifugally cast, their outer surface undergoes several visual inspections. The pipe number is manually recorded after each inspection, and a mark is printed upon acceptance. Currently, defect detection on the outer surface of ductile iron pipes is done manually, requiring a large number of quality control workers and an assembly line approach to the process.
[0003] This method is inefficient, labor-intensive, and personnel cannot maintain full concentration at all times to identify defects on the pipe surface, which may lead to missed detections. Furthermore, some defects are difficult to detect due to different observation positions, which is not conducive to coordinating with the production line to reduce the number of defective cast pipes.
[0004] In view of the above-mentioned defects, the creator of this utility model has finally obtained this utility model after a long period of research and practice. Utility Model Content
[0005] To address the aforementioned technical deficiencies, this utility model provides an online surface defect detection device for cast iron pipes, comprising a pipe rotation unit, an image acquisition unit, and an intelligent recognition unit. The pipe fitting is horizontally mounted on the pipe rotation unit, which drives the pipe fitting to rotate freely around its own axis. The image acquisition unit includes a support frame, a first adjustment component, a second adjustment component, and a detection camera. The first adjustment component is mounted on the support frame, the second adjustment component is mounted on the first adjustment component, and the detection camera is mounted on the second adjustment component, positioned directly above the pipe rotation unit. The second adjustment component drives the second adjustment component to move horizontally, and the second adjustment component drives the first adjustment component to move vertically. The intelligent recognition unit is simultaneously connected to both the pipe rotation unit and the image acquisition unit. The intelligent recognition unit controls the pipe rotation unit to rotate the pipe fitting and receives image information acquired by the image acquisition unit for analysis and detection.
[0006] Preferably, the rotating part of the pipe includes a plurality of rotating wheel sets and the fixed frame. The rotating wheel sets are arranged horizontally in a straight line on the fixed frame. Each rotating wheel set includes two horizontally symmetrically arranged rotating wheels. The axes of the two rotating wheels in the same rotating wheel set are parallel, and at least one of the rotating wheels is connected to a rotating motor for transmission. The pipe fitting is arranged above the two rotating wheels in the same rotating wheel set.
[0007] Preferably, the intelligent recognition unit includes a server, a positioning sensor, and a rotation sensor. The positioning sensor and the rotation sensor are both disposed on the rotating part of the pipe, and the positioning sensor, the rotation sensor, the rotating motor, and the detection camera are all connected to the server.
[0008] Preferably, the online surface defect detection device for the cast pipe further includes a pipe transport robot, which is used to transport the pipe fitting and is connected to the server.
[0009] Preferably, the identification unit further includes indicator lights, which are connected to the server. The indicator lights include a first warning light, a second warning light, and a third warning light, and the first warning light, the second warning light, and the third warning light all have different display colors.
[0010] Preferably, the first adjusting component includes a horizontal adjusting screw and a horizontal mounting block. The horizontal adjusting screw is rotatably disposed on the top of the support frame, and a plurality of the horizontal mounting blocks are threadedly connected to the horizontal adjusting screw. Each of the horizontal mounting blocks is provided with a second adjusting component, and the horizontal mounting blocks are equidistantly arranged along the extension direction of the horizontal adjusting screw.
[0011] Preferably, the first adjustment component further includes a horizontal guide rail and a horizontal slider. The horizontal guide rail is fixedly mounted on the support frame, and the extension direction of the horizontal guide rail is consistent with the extension direction of the horizontal adjustment screw. The horizontal slider is slidably connected on the horizontal guide rail, and the horizontal mounting block is fixedly connected to the horizontal slider in a one-to-one correspondence.
[0012] Preferably, the end of the horizontal adjusting screw is provided with an adjusting wheel.
[0013] Preferably, the second adjustment component includes an adjustment frame, the top end of which is fixedly connected to the horizontal mounting block. The adjustment frame contains a vertical adjustment screw, a mounting platform, and an adjustment motor. The adjustment motor is fixed to the adjustment frame, and its output shaft is coaxially and fixedly connected to the vertical adjustment screw. The vertical adjustment screw is vertically rotatable on the adjustment frame. The mounting platform is threadedly connected to the vertical adjustment screw. The detection camera is mounted on the mounting platform, and the adjustment motor is connected to the server.
[0014] Preferably, the installation platform is provided with a protective box, the bottom of the protective box is provided with an observation port, the detection camera is disposed inside the protective box, and the shooting port of the detection camera is provided corresponding to the observation port.
[0015] Compared with the prior art, the beneficial effects of this utility model are as follows: This utility model changes the traditional surface defect detection method from manual inspection to computer inspection, thus making automatic detection possible. At the same time, it can quickly detect pipeline surface defects and feed them back to the casting production workshop, so that the production process can be adjusted in time, reducing defective pipes and reducing the labor intensity of workers, thus realizing full inspection of pipelines. Attached Figure Description
[0016] Figure 1 This is a schematic diagram of the online surface defect detection device for the cast pipe.
[0017] Figure 2 This is a schematic diagram of the connection of the online surface defect detection device for the cast pipe;
[0018] Figure 3 This is a schematic diagram of the structure of the rotating part of the pipe;
[0019] Figure 4 This is a schematic diagram of the image acquisition unit.
[0020] Figure 5 This is a schematic diagram of the structure of the second adjustment component.
[0021] The numbers in the diagram represent:
[0022] 1-Pipe fitting; 2-Pipe rotating part; 3-Support frame; 4-First adjustment component; 5-Second adjustment component; 6-Detection camera; 21-Fixed frame; 22-Rotating wheel; 41-Horizontal adjustment screw; 42-Horizontal mounting block; 43-Adjusting wheel; 51-Adjusting frame; 52-Vertical adjustment screw; 53-Mounting platform; 54-Adjusting motor; 55-Protective housing. Detailed Implementation
[0023] The above-mentioned and other technical features and advantages of this utility model will be described in more detail below with reference to the accompanying drawings.
[0024] Example 1
[0025] like Figure 1 and Figure 2 As shown, Figure 1 This is a schematic diagram of the online surface defect detection device for the cast pipe. Figure 2 This is a schematic diagram of the connection of the online surface defect detection device for the cast pipe.
[0026] The online surface defect detection device for cast pipes of this utility model includes a pipe rotation unit 2, an image acquisition unit, and an intelligent recognition unit. A pipe fitting 1 is horizontally mounted on the pipe rotation unit 2, and the pipe rotation unit 2 drives the pipe fitting 1 to rotate freely around its own axis. The image acquisition unit includes a support frame 3, a first adjustment component 4, a second adjustment component 5, and a detection camera 6. The first adjustment component 4 is mounted on the support frame 3, the second adjustment component 5 is mounted on the first adjustment component 4, and the detection camera 6 is mounted on the second adjustment component 5, directly above the pipe rotation unit 2. The second adjustment component 5 drives the second adjustment component 5 to move horizontally, and the second adjustment component 5 drives the first adjustment component 4 to move vertically. The intelligent recognition unit is simultaneously connected to the pipe rotation unit 2 and the image acquisition unit. The intelligent recognition unit controls the pipe rotation unit 2 to rotate the pipe fitting 1 and receives the image information of the outer wall of the pipe fitting 1 acquired by the image acquisition unit for analysis and detection.
[0027] like Figure 3 As shown, Figure 3 This is a schematic diagram of the structure of the pipe rotating part; specifically, the pipe rotating part 2 includes several rotating wheel groups and a fixed frame 21. The rotating wheel groups are arranged horizontally in a straight line on the fixed frame 21. Each rotating wheel group includes two horizontally symmetrically arranged rotating wheels 22. The axes of the two rotating wheels 22 in the same rotating wheel group are parallel, and at least one of the rotating wheels 22 is connected to a rotating motor for transmission. The pipe fitting 1 is arranged above the two rotating wheels 22 in the same rotating wheel group, so that the pipe fitting 1 can rotate around its own axis after the rotating motor drives one of the rotating wheels 22 to rotate.
[0028] The intelligent identification unit includes a server, a positioning sensor, and a rotation sensor. Both the positioning sensor and the rotation sensor are mounted on the rotating pipe section 2. The positioning sensor, the rotation sensor, the rotating motor, and the detection camera 6 are all connected to the server. After the positioning sensor detects that the pipe fitting 1 is placed on the rotating pipe section 2, the server receives a pipe presence signal and controls the rotating motor to rotate the pipe fitting 1. The rotation sensor detects the number of rotations of the pipe fitting 1 and feeds this information back to the server, enabling the server to control the rotation state of the pipe fitting 1 in real time. The detection camera 6 can capture an image of the entire circumference of the outer wall of a corresponding segment of the pipe fitting 1 and transmit the image to the server for image analysis to determine defects.
[0029] The online surface defect detection device for the cast pipe also includes a pipe-carrying robot, which is used to load and unload the pipe fitting 1 onto the pipe rotating part 2. The pipe-carrying robot is connected to the server. Specifically, before detection, the pipe-carrying robot picks up the pipe fitting 1 to be inspected and places it on the pipe rotating part 2, and transmits a pipe-connection completion signal to the server. The server detects whether the fitting is in place using the positioning sensor. After the server completes the defect determination, it sends a pipe-feeding signal to the pipe-carrying robot, which then transports the inspected pipe fitting 1 from the pipe rotating part 2 to the next process.
[0030] Preferably, the identification unit further includes indicator lights connected to the server. The indicator lights include a first warning light, a second warning light, and a third warning light, each with a different display color. After the server obtains the defect determination result, it illuminates one of the first, second, or third warning lights to display different detection results to the operator.
[0031] The detection camera 6, positioned above the rotating part of the pipe fitting 1, is precisely adjusted in height via the second adjustment component 5 to ensure clear imaging. The detection camera 6 captures single-line pixels, constructing the final 2D image through these individual pixel lines. Software on the vision processor or image acquisition card stores each pixel line and then reconstructs the final 2D image from the pixel data. This unique image acquisition process excels at capturing images of rapidly moving discrete components on a conveyor belt, detecting all sides of cylindrical objects, and constructing images of very large objects. Thus, as the pipe fitting 1 rotates, the line scan camera captures an image of the complete outer surface of the pipe fitting 1, forming an image of the outer surface of the pipe fitting 1, used to detect defects on the complete outer surface of the cast pipe.
[0032] The server collects big data on defective pipe fittings 1, constructs and uses a deep learning model, and compares the images captured by the detection camera 6 with the deep learning model after the server performs noise filtering, contrast enhancement, image scaling, and other processing, thereby automatically judging the appearance quality of pipe fittings 1.
[0033] This invention replaces manual inspection with computer-based testing, making automatic inspection possible. It can also quickly detect surface defects in pipe fitting 1 and provide feedback to the casting production workshop, allowing for timely adjustments to the production process, reducing defective pipes, and lowering the labor intensity of workers, thus achieving full inspection of pipe fitting 1.
[0034] Example 2
[0035] like Figure 4 and Figure 5 As shown, Figure 4 This is a schematic diagram of the image acquisition unit. Figure 5 This is a schematic diagram of the structure of the second adjustment component.
[0036] The first adjustment component 4 includes a horizontal adjustment screw 41 and horizontal mounting blocks 42. The horizontal adjustment screw 41 is rotatably mounted on the top of the support frame 3, and several horizontal mounting blocks 42 are threadedly connected to the horizontal adjustment screw 41. Each horizontal mounting block 42 is provided with a second adjustment component 5, and the horizontal mounting blocks 42 are equidistantly arranged along the extension direction of the horizontal adjustment screw 41. By rotating the horizontal adjustment screw 41, the horizontal position of each horizontal mounting block 42 can be adjusted, while ensuring that the distance between adjacent horizontal mounting blocks 42 remains unchanged. Therefore, the detection camera 6 on each of the second adjustment components 5 can still ensure that the detection interval length remains unchanged before and after adjustment, ensuring full coverage detection of the pipe fitting 1.
[0037] The first adjustment component 4 further includes a horizontal guide rail and a horizontal slider. The horizontal guide rail is fixedly mounted on the support frame 3, and the extension direction of the horizontal guide rail is consistent with the extension direction of the horizontal adjusting screw 41. The horizontal slider is slidably connected on the horizontal guide rail, and the horizontal mounting block 42 is fixedly connected to the horizontal slider in a one-to-one correspondence. Through the structural arrangement of the horizontal guide rail and the horizontal slider, it can be ensured that the horizontal mounting block 42 moves stably along a straight line during the rotation of the horizontal adjusting screw 41.
[0038] The end of the horizontal adjusting screw 41 is provided with an adjusting wheel 43. The operator rotates the horizontal adjusting screw 41 through the adjusting wheel 43 to adjust the relative position of each detection camera 6 and the pipe rotating part 2. Generally, the alignment position is adjusted at the beginning of the installation.
[0039] The second adjustment component 5 includes an adjustment frame 51, the top of which is fixedly connected to the horizontal mounting block 42. The adjustment frame 51 contains a vertical adjustment screw 52, a mounting platform 53, and an adjustment motor 54. The adjustment motor 54 is fixed to the adjustment frame 51, and its output shaft is coaxially and fixedly connected to the vertical adjustment screw 52. The vertical adjustment screw 52 is vertically rotatable on the adjustment frame 51. The mounting platform 53 is threadedly connected to the vertical adjustment screw 52. The detection camera 6 is mounted on the mounting platform 53. The adjustment motor 54 rotates, causing the mounting platform 53 to move vertically. The adjustment motor 54 is connected to the server, allowing the server to adjust the viewing angle of the detection camera 6 to obtain better image data.
[0040] Preferably, the installation platform 53 is provided with a protective box 55, the bottom of the protective box 55 is provided with an observation port, the detection camera 6 is provided inside the protective box 55, and the shooting port of the detection camera 6 is provided corresponding to the observation port to shoot the pipe 1 directly below. The protective box 55 is used to protect the detection camera 6 and prevent the pipe 1 from being damaged by collision with the detection camera 6 during transportation.
[0041] The above description is merely a preferred embodiment of the present utility model and is illustrative rather than restrictive. Those skilled in the art will understand that many changes, modifications, and even equivalents can be made within the spirit and scope defined by the claims of the present utility model, all of which will fall within the protection scope of the present utility model.
Claims
1. An online surface defect detection device for cast pipes, characterized in that, The system includes a pipe rotation unit, an image acquisition unit, and an intelligent recognition unit. The pipe fitting is horizontally mounted on the pipe rotation unit, which drives the pipe fitting to rotate freely around its own axis. The image acquisition unit includes a support frame, a first adjustment component, a second adjustment component, and a detection camera. The first adjustment component is mounted on the support frame, the second adjustment component is mounted on the first adjustment component, and the detection camera is mounted on the second adjustment component, positioned directly above the pipe rotation unit. The second adjustment component drives the second adjustment component to move horizontally, and the second adjustment component drives the first adjustment component to move vertically. The intelligent recognition unit is simultaneously connected to both the pipe rotation unit and the image acquisition unit. The intelligent recognition unit controls the pipe rotation unit to rotate the pipe fitting and receives image information acquired by the image acquisition unit for analysis and detection.
2. The online surface defect detection device for cast pipes as described in claim 1, characterized in that, The rotating part of the pipe includes several sets of rotating wheels and a fixed frame. The sets of rotating wheels are arranged horizontally in a straight line on the fixed frame. Each set of rotating wheels includes two horizontally symmetrically arranged rotating wheels. The axes of the two rotating wheels in the same set are parallel, and at least one of the rotating wheels is connected to a rotating motor. The pipe fitting is located above the two rotating wheels in the same set.
3. The online surface defect detection device for cast pipes as described in claim 2, characterized in that, The intelligent recognition unit includes a server, a positioning sensor, and a rotation sensor. The positioning sensor and the rotation sensor are both installed on the rotating part of the pipe. The positioning sensor, the rotation sensor, the rotating motor, and the detection camera are all connected to the server.
4. The online surface defect detection device for cast pipes as described in claim 3, characterized in that, The online surface defect detection device for the cast pipe also includes a pipe transport robot, which is used to transport the pipe fittings and is connected to the server.
5. The online surface defect detection device for cast pipes as described in claim 4, characterized in that, The identification unit also includes indicator lights, which are connected to the server. The indicator lights include a first warning light, a second warning light, and a third warning light, and the first warning light, the second warning light, and the third warning light all have different display colors.
6. The online surface defect detection device for cast pipes as described in claim 3, characterized in that, The first adjustment component includes a horizontal adjustment screw and a horizontal mounting block. The horizontal adjustment screw is rotatably mounted on the top of the support frame, and a plurality of the horizontal mounting blocks are threadedly connected to the horizontal adjustment screw. Each of the horizontal mounting blocks is provided with a second adjustment component, and the horizontal mounting blocks are equidistantly arranged along the extension direction of the horizontal adjustment screw.
7. The online surface defect detection device for cast pipes as described in claim 6, characterized in that, The first adjustment component further includes a horizontal guide rail and a horizontal slider. The horizontal guide rail is fixedly mounted on the support frame, and the extension direction of the horizontal guide rail is consistent with the extension direction of the horizontal adjustment screw. The horizontal slider is slidably connected on the horizontal guide rail, and the horizontal mounting block is fixedly connected to the horizontal slider in a one-to-one correspondence.
8. The online surface defect detection device for cast pipes as described in claim 7, characterized in that, An adjusting wheel is provided at the end of the horizontal adjusting screw.
9. The online surface defect detection device for cast pipes as described in claim 6, characterized in that, The second adjustment component includes an adjustment frame, the top of which is fixedly connected to the horizontal mounting block. The adjustment frame contains a vertical adjustment screw, a mounting platform, and an adjustment motor. The adjustment motor is fixed to the adjustment frame, and its output shaft is coaxially and fixedly connected to the vertical adjustment screw. The vertical adjustment screw is vertically rotatable on the adjustment frame. The mounting platform is threadedly connected to the vertical adjustment screw. The detection camera is mounted on the mounting platform, and the adjustment motor is connected to the server.
10. The online surface defect detection device for cast pipes as described in claim 9, characterized in that, The installation platform is provided with a protective box, and the bottom of the protective box is provided with an observation port. The detection camera is installed inside the protective box, and the shooting port of the detection camera is set to correspond to the observation port.