Image recognition auxiliary device for seamless steel tube drawing detection

Abstract

The application discloses an image recognition auxiliary device for seamless steel pipe sampling and detection, which comprises a dark box, a cutting unit and a detection unit arranged in the dark box, and through holes for the seamless steel pipe to pass through are arranged on the front and back sides of the dark box; a plurality of support guide wheels for supporting the seamless steel pipe are horizontally arranged in the dark box; the cutting unit comprises a laser head and a controllable roller, the output end of the laser head acts on the radial outer surface of the seamless steel pipe, the controllable roller is located on one side of the seamless steel pipe, and the radial outer wall of the seamless steel pipe is crimped on the controllable roller; the detection unit comprises a sample bearing disc, a light supplementing lamp and a camera module, and the sample bearing disc is located at the bottom of the dark box. The device is integrated in a detection station of a seamless steel pipe production line, is used for sampling and cutting the steel pipe under the normal operation state of the production line, and is used for high-precision image acquisition and size analysis on the ring samples cut down, so that the wall thickness uniformity and the roundness of the whole steel pipe are evaluated.

Description

Technical Field

[0001] This invention relates to the field of metal material processing and inspection technology, specifically an image recognition auxiliary device for sampling inspection of seamless steel pipes. Background Technology

[0002] Seamless steel pipes are core pressure-bearing components in energy transmission, machinery manufacturing, and major infrastructure sectors, and are widely used in oil and gas extraction, chemical equipment, power plant boilers, and high-pressure hydraulic systems. These applications place extremely stringent requirements on the structural integrity and pressure-bearing safety of steel pipes. The uniformity of wall thickness and the roundness of geometric regularity are key quality indicators that directly determine the product's safety level, service life, and system reliability.

[0003] In the hot-rolling or cold-rolling process of seamless steel pipe manufacturing, real-time and accurate detection of the wall thickness uniformity and roundness of the finished pipe is a crucial step in ensuring product quality. Currently, the industry mainly uses the following two methods for detection: Manual inspection: At the end of the production line, workers use measuring tools such as calipers and micrometers to perform manual sampling measurements. However, this method has drawbacks such as low efficiency, high labor intensity, and susceptibility to subjective errors. Furthermore, it cannot achieve 100% inspection, posing a risk of missed quality checks and making it difficult to meet the needs of modern large-scale production.

[0004] Machine vision-based online inspection involves setting up industrial cameras on the production line to capture and identify moving steel pipes. However, the actual production environment is complex, with interference factors such as oil stains, moisture, vibration, and uneven lighting. Furthermore, as the steel pipe travels on the roller conveyor, the relative position and distance between it and the fixed camera fluctuate slightly, causing changes in imaging scale and blurred edges. This severely affects the accuracy of image recognition algorithms in locating wall thickness and contour edges, resulting in measurement deviations.

[0005] In summary, existing technologies cannot simultaneously achieve high precision, high efficiency, and full automation in harsh production line environments. Therefore, there is an urgent need for an automated testing device that can overcome environmental interference, achieve stable sample transmission and precise positioning, and form a closed-loop control with the production process. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art and propose an image recognition auxiliary device for seamless steel pipe sampling inspection. It can still achieve high-precision and high-stability automatic inspection even in special scenarios where online visual inspection is affected by oil, water vapor, vibration, uneven lighting, and the movement state of the steel pipe in the production line environment.

[0007] To address the aforementioned technical problems, this invention provides an image recognition auxiliary device for seamless steel pipe sampling inspection, comprising a darkroom and a cutting unit and an inspection unit disposed within the darkroom. The darkroom has through holes on its front and rear sides for the seamless steel pipe to pass through, and several horizontally arranged support guide wheels inside the darkroom to support the seamless steel pipe.

[0008] The cutting unit includes a laser head and a controllable roller. The output end of the laser head acts on the radial outer surface of the seamless steel tube to cut a ring-shaped sample from the seamless steel tube. The controllable roller is located on one side of the seamless steel tube, and the radial outer wall of the seamless steel tube is pressed against the controllable roller. The rotation of the controllable roller drives the seamless steel tube to move, achieving precise control of the cutting position.

[0009] The detection unit includes a sample carrier plate, a supplementary light, and a camera module. The sample carrier plate, located at the bottom of the dark chamber, is used to receive and temporarily hold the cut sample to be tested. Above the sample carrier plate is a mounting bracket installed on the inner wall of the dark chamber. The supplementary light and camera module are mounted on the mounting bracket, positioned directly above the sample carrier plate. The supplementary light provides a uniform and stable illumination environment for the sample, while the camera module captures high-resolution still images of the sample when it is stationary on the sample carrier plate for subsequent image analysis.

[0010] Preferably, an auxiliary roller is installed on the other side of the seamless steel pipe, with its lower surface abutting against the seamless steel pipe to enhance the stability of the steel pipe transmission. An arc-shaped buffer plate is installed below the laser head, with one end integrally formed with the body of the dark chamber and the other end extending to the front end of the sample carrier plate to receive the cut sample and guide it into the sample carrier plate. A flexible buffer pad is installed on the surface of the arc-shaped buffer plate to absorb the impact energy when the sample falls. Movable guide plates are symmetrically installed on both sides of the end of the arc-shaped buffer plate to guide and correct the sample falling to the end, ensuring that it falls accurately into the center area of ​​the sample carrier plate. The guide plate has a teardrop-shaped structure, with its enlarged end mounted on the arc-shaped buffer plate via a rotating shaft, and its tip facing the sample carrier plate, using its geometry to achieve a stable guiding effect.

[0011] Preferably, the sample carrier includes a rectangular carrier frame, a carrier plate, and a rotating shaft. A through hole is provided in the center of the carrier plate, through which the rotating shaft is fixedly mounted. Both ends of the rotating shaft are rotatably mounted on the carrier frame, and one end of the rotating shaft is connected to the output end of a drive assembly. After image acquisition and analysis are completed, the drive assembly drives the rotating shaft to rotate, causing the carrier plate to flip, allowing the sample to slide out under gravity. The drive assembly then drives the carrier plate back to a horizontal position to receive the next sample. The surface of the carrier plate is provided with a soft buffer layer, and the surface of the soft buffer layer has a grid scale. The soft buffer layer prevents the sample from bouncing or shifting on the carrier plate, and the grid scale provides a precise physical dimension reference for subsequent image processing.

[0012] Preferably, the supplementary light is a coaxial light source or a ring light source, and the optical axis of the camera module is perpendicular to the bearing surface of the sample carrier plate to ensure that the acquired image is free of perspective distortion.

[0013] This invention has the following characteristics and beneficial effects: High detection accuracy: The sample to be tested is stationary and in a fixed position on the sample carrier plate, and the imaging environment light source and distance are kept constant inside the dark box, which completely eliminates the imaging error caused by vibration and distance changes during online dynamic shooting, thus greatly improving the edge recognition accuracy.

[0014] Precise and stable transmission and positioning: The arc-shaped buffer plate, flexible buffer pad and movable guide plate work together to effectively absorb the impact of sample falling and correct the landing point deviation, ensuring that the sample to be tested falls accurately and stably into the center area of ​​the sample carrier plate.

[0015] Strong anti-interference capability: The core processes such as cutting, transmission, and detection are all integrated inside a closed dark box, which effectively avoids external interference such as oil, water vapor, dust and stray light in the production line environment, ensuring the stability and consistency of image acquisition.

[0016] High degree of automation: The sample carrier plate is connected to the drive component through a rotating shaft, which can automatically flip over to unload and reset after the test is completed, realizing a continuous and automated testing process without manual intervention, which significantly improves the testing efficiency. Attached Figure Description

[0017] Figure 1 This is a cross-sectional view of an image recognition auxiliary device for seamless steel pipe sampling inspection according to an embodiment of the present invention.

[0018] Figure 2 for Figure 1 Another sectional view.

[0019] Figure 3 This is a schematic diagram of another embodiment.

[0020] Figure label: 1-Dark box, 2-Laser head, 3-Controllable roller, 4-Supporting guide wheel, 5-Arc-shaped buffer plate, 6-Auxiliary roller, 7-Roller shaft, 8-Seamless steel pipe, 9-Mounting frame, 10-Fill light, 11-Camera module, 12-Sample carrier plate, 13-Rotating shaft, 14-Guide plate, 15-Sample to be tested. Detailed Implementation

[0021] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0022] Example 1 This embodiment provides an image recognition auxiliary device for the sampling inspection of seamless steel pipes, such as... Figure 1 and Figure 2 As shown, the device is integrated into the inspection station of the seamless steel pipe production line. It is used to sample and cut steel pipes during normal operation of the production line, and to perform high-precision image acquisition and dimensional analysis on the cut circular samples to evaluate the wall thickness uniformity and roundness of the entire steel pipe.

[0023] Specifically, the device includes a closed dark box 1. The dark box 1 has through holes on both the front and rear sides for seamless steel pipes 8 to pass through, allowing the seamless steel pipes on the production line to pass horizontally through the dark box. Several horizontally arranged support guide wheels 4 are located below the seamless steel pipes 8 inside the dark box 1, used to support and guide the transport of the seamless steel pipes and restrict their vertical movement. An auxiliary roller 6 is also provided on one side of the seamless steel pipe 8. The auxiliary roller 6 is rotatably mounted on the side wall of the dark box 1 via a roller shaft 7, with its lower surface abutting against the surface of the seamless steel pipe 8. Together with the support guide wheels 4, it maintains the stable posture of the steel pipe during transport, preventing longitudinal jumping or lateral swaying.

[0024] Furthermore, a cutting unit is installed inside the dark box 1, which includes a laser head 2 and a controllable roller 3. In this embodiment, the laser head 2 uses a high-power fiber laser with an output power of 1kW-3kW. Its laser output end is located directly above the seamless steel pipe 8, enabling it to generate a high-energy-density laser beam to cut the radial outer surface of the seamless steel pipe 8. During the cutting process of the laser head 2, the controllable roller 3 presses tightly against one side surface of the seamless steel pipe 8, driving the steel pipe to rotate and precisely controlling the cutting position. After cutting, a ring-shaped test sample 15 is separated from the steel pipe. This test sample 15 has the same thermal history and process parameters as the parent material, and can truly reflect the forming quality of this section of the steel pipe.

[0025] Understandably, the controllable roller 3 includes a roller and a drive mechanism. The drive mechanism drives the roller to rotate, which in turn drives the seamless steel pipe to rotate.

[0026] Furthermore, an arc-shaped buffer plate 5 is disposed directly below the laser head 2. One end of the arc-shaped buffer plate 5 is integrally formed with the body of the dark chamber 1, and the other end extends downward at an angle to the front end of the sample carrier plate 12. The surface of the arc-shaped buffer plate 5 is covered with a flexible buffer pad, which in this embodiment is a polyurethane rubber pad, to absorb the impact energy of the sample 15 falling and prevent the sample from deforming due to collision.

[0027] Furthermore, movable guide plates 14 are symmetrically arranged on both sides of the end of the arc-shaped buffer plate 5. The guide plates 14 have a teardrop-shaped structure, and their enlarged ends are mounted on the arc-shaped buffer plate 5 via a pivot, with their tips facing the sample support plate 12. When the sample 15 to be tested slides down the arc-shaped buffer plate 5 to the end, it will impact the guide plates 14 on both sides. After being impacted, the guide plates 14 will swing slightly around the pivot, which will buffer and decelerate the sample. At the same time, under the action of an elastic element such as a torsion spring, the guide plates 14 will reset and generate a pushing force towards the center, guiding the sample 15 to fall accurately into the central area of ​​the sample support plate 12 directly below.

[0028] Furthermore, a detection unit is also provided inside the dark chamber 1, which includes a sample carrier plate 12, a supplementary light 10, and a camera module 11. The sample carrier plate 12 is located at the bottom of the dark chamber 1, and its structure includes a rectangular carrier frame, a carrier plate, and a rotating shaft 13. A through hole is provided in the middle of the carrier plate, through which the rotating shaft 13 is fixedly mounted. Both ends of the rotating shaft 13 are rotatably mounted on the carrier frame via bearings, and one end is connected to the output end of the drive component. In this embodiment, the drive component is a servo motor. A soft buffer layer is provided on the upper surface of the carrier plate. In this embodiment, a silicone pad is used. The surface of the soft buffer layer is etched or printed with a high-precision grid scale, wherein the grid scale is a square grid with a side length of 1mm. When the sample 15 to be tested falls into the carrier plate, the soft buffer layer can effectively absorb residual impact and prevent the sample from bouncing or shifting on the carrier plate; at the same time, the grid scale provides a precise physical dimension reference for subsequent image processing.

[0029] Throughout the process, the sample 15 is in contact with the flexible material, which protects the sample 15 and ensures the accuracy of the test. This prevents the sample from being damaged or deformed during the movement, which would affect the accuracy of the test.

[0030] Specifically, a mounting bracket 9 is installed directly above the sample carrier tray 12, mounted on the inner wall of the darkroom 1. A supplementary light 10 and a camera module 11 are fixedly mounted on the mounting bracket 9. The supplementary light 10 uses a high-brightness ring-shaped LED light source, which emits light that evenly illuminates the carrier surface of the sample carrier tray 12, eliminating shadows and improving contour contrast. The camera module 11 uses an industrial area array camera, such as a 5-megapixel CMOS camera, with its optical axis perpendicular to the carrier surface of the sample carrier tray 12, ensuring that the acquired image is free of perspective distortion. A filter can be added in front of the lens of the camera module 11 to filter out ambient stray light, further improving image clarity.

[0031] Understandably, once the sample 15 to be tested is stationary on the sample support plate 12, the control module triggers the camera module 11 to capture a high-resolution still image. This image is then transmitted to the industrial control computer, i.e., the data analysis unit.

[0032] In this embodiment, the image analysis algorithm is executed using the following steps: (1) Scale calibration and coordinate system establishment: First, based on the grid scale in the image, use calibration algorithms such as linear regression or perspective transformation to establish the mapping relationship between pixel coordinates and actual physical coordinates, correct the possible distortion of the lens, and obtain the physical size corresponding to each pixel, such as 0.01mm / pixel.

[0033] (2) Contour recognition and edge extraction: Subpixel edge detection algorithms such as Canny edge detection combined with Zernike moment fitting are used to accurately extract the inner and outer contours of the sample 15 to be tested from the image.

[0034] (3) Wall thickness calculation: Select at least 64 equally divided points on the inner and outer contours, calculate the distance between each pair of corresponding points, and obtain the wall thickness value at that point. Statistically analyze all wall thickness values ​​to obtain the maximum, minimum, average, and range of the wall thickness.

[0035] (4) Calculation of roundness error: Perform least squares circle fitting on the inner and outer contour point sets respectively, calculate the radial distance from each contour point to the fitted circle, and take the maximum radial deviation as the roundness error.

[0036] (5) Quality judgment and feedback: The calculated wall thickness range and roundness error are compared with the preset process standards. If the threshold is exceeded, the control module sends process adjustment instructions to the production line main control system through industrial communication protocols such as OPC UA, such as adjusting the mill pressure and guide position, and marks the batch of steel pipes and triggers an alarm.

[0037] Finally, after image acquisition and data analysis are completed, the control module sends a command to the servo motor of the drive component. The servo motor drives the carrier plate to rotate around its axis by a certain angle, for example, 110°, via the rotating shaft 13. At this time, the sample 15 to be tested slides off the carrier plate under the action of gravity and falls into the waste collection tank on the side. Subsequently, the servo motor drives the carrier plate to rotate in the opposite direction, returning it to the initial horizontal working position, ready to receive the next sample 15 to be tested. The entire unloading and resetting process takes no more than 2 seconds, realizing continuous automated testing synchronized with the production line's production rhythm.

[0038] Example 2 This embodiment is an optimization based on Embodiment 1. To further improve the stability of the sample 15 during transport, several guide wheels (not shown) are provided on the guide plate 14. Because the sample needs to be protected, it comes into contact with flexible materials, resulting in high friction. The guide wheels reduce this friction, allowing the sample to smoothly enter the sample carrier tray 12 and ensuring it falls horizontally into the tray.

[0039] Example 3 The difference between this embodiment and Embodiment 1 is that, as Figure 3 As shown, to further improve the cushioning and service life of the curved buffer plate 5, a wavy support buffer unit is set below the curved buffer plate 5. It is understandable that the sample under test has a relatively large mass, and during the fall, it impacts the buffer pad and the curved buffer plate, which can easily be damaged over time. Therefore, by setting up the support buffer unit, the service life of the buffer pad and the curved buffer plate is increased, and the shock absorption effect is further enhanced.

[0040] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. An image recognition auxiliary device for sampling inspection of seamless steel pipes, characterized in that, The device includes a dark box (1) and a cutting unit and a detection unit set inside the dark box (1). The dark box (1) has through holes on its front and rear sides for the seamless steel pipe (8) to pass through. Several support guide wheels (4) for supporting the seamless steel pipe (8) are horizontally arranged inside the dark box (1). The cutting unit includes a laser head (2) and a controllable roller (3). The output end of the laser head (2) acts on the radial outer surface of the seamless steel pipe (8). The controllable roller (3) is located on one side of the seamless steel pipe (8). The radial outer wall of the seamless steel pipe (8) is pressed onto the controllable roller (3). The detection unit includes a sample carrier plate (12), a supplementary light (10), and a camera module (11). The sample carrier plate (12) is located at the bottom of the dark box (1). A mounting frame (9) installed on the inner wall of the dark box (1) is provided above the sample carrier plate (12). The supplementary light (10) and the camera module (11) are installed on the mounting frame (9) and are located directly above the sample carrier plate (12).

2. The image recognition auxiliary device for seamless steel pipe sampling inspection according to claim 1, characterized in that, An auxiliary roller (6) is provided on the other side of the seamless steel pipe (8), and the lower surface of the auxiliary roller (6) abuts against the seamless steel pipe (8).

3. The image recognition auxiliary device for seamless steel pipe sampling inspection according to claim 1, characterized in that, An arc-shaped buffer plate (5) is provided below the laser head (2). One end of the arc-shaped buffer plate (5) is integrally formed with the body of the dark box (1), and the other side of the arc-shaped buffer plate (5) extends to the front end of the sample carrier plate (12).

4. The image recognition auxiliary device for seamless steel pipe sampling inspection according to claim 3, characterized in that, The surface of the arc-shaped buffer plate (5) is provided with a flexible buffer pad.

5. The image recognition auxiliary device for seamless steel pipe sampling inspection according to claim 3, characterized in that, The arc-shaped buffer plate (5) has movable guide plates (14) symmetrically arranged on both sides of its end.

6. The image recognition auxiliary device for seamless steel pipe sampling inspection according to claim 5, characterized in that, The guide plate (14) has a teardrop-shaped structure. The enlarged end of the guide plate (14) is mounted on the arc-shaped buffer plate (5) through a rotating shaft. The tip of the guide plate (14) faces the sample carrier plate (12).

7. The image recognition auxiliary device for seamless steel pipe sampling inspection according to claim 1, characterized in that, The sample carrier plate (12) includes a rectangular carrier frame, a carrier plate and a rotating shaft (13). The carrier plate has a through hole in the middle. The carrier plate is fixedly installed on the rotating shaft through the through hole. The two ends of the rotating shaft are rotatably installed on the carrier frame. One end of the rotating shaft is connected to the output end of the drive component.

8. The image recognition auxiliary device for seamless steel pipe sampling inspection according to claim 7, characterized in that, The surface of the bearing plate is provided with a soft buffer layer, and the surface of the soft buffer layer is provided with a grid scale.

9. The image recognition auxiliary device for seamless steel pipe sampling inspection according to claim 1, characterized in that, The supplementary light (10) is a coaxial light source or a ring light source, and the optical axis of the camera module (11) is perpendicular to the bearing surface of the sample carrier plate (12).

10. An image recognition auxiliary device for seamless steel pipe sampling inspection according to claim 1, characterized in that, The test sample (15) formed after the laser head (2) cuts the seamless steel pipe (8) is located on the arc-shaped buffer plate (5).

11. An image recognition auxiliary device for seamless steel pipe sampling inspection according to claim 2, characterized in that, The auxiliary roller (6) is installed in the dark box (1) via a roller shaft (7), and the auxiliary roller (6) is rotatably mounted on the roller shaft (7).

12. The image recognition auxiliary device for seamless steel pipe sampling inspection according to claim 1, characterized in that, The dark box (1) is a closed shell, and the sample carrier plate (12), the fill light (10) and the camera module (11) are all housed inside the dark box (1).

Images