A vision-based welding defect detection apparatus

Abstract

This invention relates to a vision-based welding defect detection device, which mainly consists of a conveyor belt, a label reader, an L-shaped support, x-y-z direction displacement adjustment components, an industrial camera, and a computer terminal. The conveyor belt transports the welded metal sheet, the label reader reads the QR code label on the sheet and transmits the product information to the computer terminal, the industrial camera adjusts its position in the x, y, and z directions via the displacement adjustment components, the computer terminal determines the optimal shooting position based on the product information and controls the camera, simultaneously acquiring the captured images for defect detection analysis, and storing the detection results in association with the product information. Compared with existing technologies, this invention can achieve automated and precise detection of welding defects in different metal sheets, effectively improving detection efficiency and accuracy, meeting the needs of strict welding quality control in industrial production, and has broad application prospects.

Description

Technical Field

[0001] This invention relates to a welding defect detection device, and more particularly to a vision-based welding defect detection device. Background Technology

[0002] In modern industrial production, welding, as a crucial joining process, is widely used in the processing and manufacturing of sheet metal, such as in automobile manufacturing, shipbuilding, and steel structure engineering. However, due to various factors during the welding process, such as fluctuations in welding process parameters, differences in the quality of welding materials, varying skill levels of operators, and changes in the welding environment, various welding defects inevitably occur, such as porosity, cracks, lack of fusion, and slag inclusions. These welding defects seriously affect the quality and performance of the welded joint, thereby threatening the safety and reliability of the entire product.

[0003] To ensure welding quality, welding defect detection has become an indispensable part of the production process. Traditional welding defect detection methods mainly include manual visual inspection, ultrasonic testing, and radiographic testing. Manual visual inspection relies on the experience and skills of the inspectors, is highly subjective, has low efficiency, and can easily lead to visual fatigue after prolonged work, thus affecting the accuracy and reliability of the inspection. Although ultrasonic testing and radiographic testing can detect internal defects to a certain extent, they are complex to operate, require professional technicians, use expensive equipment, and may pose certain health risks during the process. Furthermore, they are not suitable for rapid, real-time detection of welding defects.

[0004] With the rapid development of computer and machine vision technologies, vision-based welding defect detection methods have gradually gained attention and application. This method utilizes industrial cameras to capture images of the welding area, which are then analyzed and processed by a computer to identify welding defects. However, existing vision-based welding defect detection equipment has some limitations. In the image capture stage, most devices use fixed-position industrial cameras or can only perform simple manual adjustments, failing to flexibly adjust the shooting position and angle according to the characteristics of different products and the differences in welding locations. This makes it difficult to obtain clear and accurate images of the welding area, thus affecting the accuracy and reliability of defect detection. Furthermore, during the inspection process, for a large number of different batches and specifications of welded products, there is a lack of effective means to accurately correlate and manage product information with inspection data, which is detrimental to subsequent quality traceability and production process improvement.

[0005] Therefore, there is an urgent need for a new vision-based welding defect detection device that can solve the above problems, so as to improve the accuracy and efficiency of welding defect detection and meet the needs of modern industrial production for strict control of welding quality. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the existing technology and provide a vision-based welding defect detection device that can realize automated and precise detection of welding defects in different metal plates, effectively improve detection efficiency and accuracy, meet the needs of strict control of welding quality in industrial production, and has broad application prospects.

[0007] The objective of this invention can be achieved through the following technical solutions:

[0008] This invention provides a vision-based welding defect detection device, comprising a conveyor belt, a label reader, an L-shaped bracket, an xyz direction displacement adjustment component, an industrial camera, and a computer terminal, wherein specifically:

[0009] Conveyor belts are used to transport welded metal sheets to be inspected;

[0010] A label reader is located at the beginning of the conveyor belt and is used to read the QR code labels on the metal sheet.

[0011] The xyz direction displacement adjustment component is located on the crossbar of the L-shaped bracket;

[0012] The industrial camera is mounted on the xyz direction displacement adjustment component. The position of the industrial camera relative to the conveyor belt surface in the three orthogonal directions of x, y, and z is adjusted by the xyz direction displacement adjustment component to determine the optimal shooting position.

[0013] The computer terminal is communicatively connected to the tag reader and the industrial camera. The computer terminal is used to acquire product information of the metal sheet read by the tag reader, and at the same time, the computer terminal is used to acquire surface photos of the metal sheet taken by the industrial camera, and to perform defect detection based on the surface photos of the metal sheet.

[0014] Furthermore, support rods for supporting the conveyor belt are provided on both sides of the conveyor belt.

[0015] Furthermore, the L-shaped bracket includes a vertical rod connected to the ground and a horizontal rod connected to the top of the vertical rod, the vertical rod being perpendicular to the horizontal rod.

[0016] Furthermore, the xyz direction displacement adjustment component includes a horizontal lead screw and slider assembly disposed on the crossbar, a vertical carrier plate disposed on the horizontal lead screw and slider assembly, a vertical lead screw and slider assembly disposed on the vertical carrier plate, and a connecting rod disposed on the vertical lead screw and slider assembly, with the industrial camera disposed on the connecting rod.

[0017] Furthermore, the crossbar is provided with a transverse groove along its length.

[0018] Furthermore, the transverse lead screw and slider assembly includes a first servo motor mounted on the crossbar, a transverse lead screw connected to the output end of the first servo motor and disposed in the transverse groove, and a first slider threadedly matched with the transverse lead screw. The first slider can be displaced along the transverse groove under the drive of the first servo motor.

[0019] Furthermore, the vertical carrier plate is connected to the first slider;

[0020] The vertical carrier plate is provided with a vertical sliding groove.

[0021] Furthermore, the vertical screw and slider assembly includes a second servo motor mounted on the vertical carrier plate, a vertical screw connected to the output end of the second servo motor and disposed in the vertical groove, and a second slider threadedly matched with the vertical screw. The second slider can be displaced along the vertical groove under the drive of the second servo motor.

[0022] Furthermore, the connecting rod is connected to the second slider, and the industrial camera is located at the lower end of the connecting rod.

[0023] Furthermore, both the first servo motor and the second servo motor are communicatively connected to the computer terminal.

[0024] Compared with the prior art, the present invention has the following beneficial effects:

[0025] 1) Traditional welding defect detection equipment often uses fixed positions or universal shooting settings, making it difficult to achieve personalized optimal shooting angles and positions for different products. This invention solves the problem of not being able to perform targeted and precise shooting based on the characteristics of different products by setting a label reader at the beginning of the conveyor belt to read the QR code labels on the metal sheets, and a computer terminal to obtain the corresponding product information. Each product information corresponds to an optimal shooting position, and the computer terminal uses xyz direction displacement adjustment components to precisely adjust the shooting position of the industrial camera. This solves the problem of not being able to perform targeted and precise shooting based on the characteristics of different products, thus better capturing accurate image information of each product surface for subsequent defect detection, improving the effectiveness and accuracy of the detection.

[0026] 2) With a large number of welded products, accurately knowing the inspection status of each product and facilitating subsequent quality traceability is a major challenge. This invention uses a label reader to read QR code labels to obtain product information. A computer terminal then matches this information with the defect detection results obtained from subsequent shooting and analysis based on the optimal shooting position. This avoids mismatches and confusion between inspection results and specific products, effectively solving the problems of inaccurate information and difficulty in matching specific products during product quality traceability. This allows enterprises to accurately grasp the quality status of each welded product and make corresponding production improvement decisions.

[0027] 3) In the past, adjusting the position of industrial cameras relied on human experience and manual operation, which was not only inefficient but also difficult to guarantee that the optimal shooting position was achieved every time. This invention utilizes the collaborative communication between a computer terminal, a tag reader, and xyz displacement adjustment components to automatically determine and instruct the adjustment of the industrial camera position based on product information. This achieves automation and intelligence in shooting position adjustment, overcoming the problems of insufficient precision, cumbersome operation, and inability to quickly respond to the shooting needs of different products in manual adjustment methods. It improves the automation level and detection efficiency of the entire welding defect detection process. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of the main structure of a vision-based welding defect detection device.

[0029] Figure 2 This is a side view schematic diagram of a vision-based welding defect detection device.

[0030] In the diagram: 1. Conveyor belt, 2. Tag reader, 3. Metal sheet, 4. L-shaped bracket, 5. Industrial camera, 6. Computer terminal, 7. Support rod, 8. Vertical carrier plate, 9. First servo motor, 10. Second servo motor, 11. Connecting rod. Detailed Implementation

[0031] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. Component models, material names, connection structures, control methods, algorithms, and other features not explicitly described in this technical solution are considered common technical features disclosed in the prior art.

[0032] Example 1

[0033] This embodiment provides a vision-based welding defect detection device, including a conveyor belt 1, a label reader 2, an L-shaped bracket 4, an xyz direction displacement adjustment assembly, an industrial camera 5, and a computer terminal 6, as detailed below. Figure 1 and Figure 2 .

[0034] Conveyor belt 1 is used to transport the welded metal sheets 3 to be inspected. Conveyor belt 1 is a conventional industrial electric conveyor belt, made of rubber or metal, with a certain degree of friction on its surface. It is used to carry and transport the welded metal sheets 3 to be inspected. Driven by a motor, it achieves uniform speed movement, ensuring that the metal sheets pass through the inspection area sequentially. Its running speed can be adjusted according to actual inspection needs, playing a fundamental role in material transport within the entire inspection equipment. Support rods 7 are provided on both sides of conveyor belt 1 to support it.

[0035] The label reader 2 is located at the beginning of the conveyor belt 1 and is used to read the QR code labels on the metal plate 3. It is a photoelectric or laser scanning device, installed above the beginning of the conveyor belt 1 at a suitable height from the surface of the conveyor belt. It can quickly and accurately read the QR code label information affixed to the metal plate 3 and transmit the read product information to the computer terminal 6 in the form of an electrical signal. This provides key identification data for subsequent targeted testing operations based on product characteristics, and realizes the accurate correlation between product information and the testing process.

[0036] The L-shaped bracket 4 is typically constructed from welded or assembled metal materials. It includes a vertical rod connected perpendicularly to the ground and a horizontal rod connected perpendicularly to the top of the vertical rod. The vertical rod provides stable support for the entire bracket, while the horizontal rod is used to mount the xyz displacement adjustment components. Its structural design ensures that the industrial camera 5 can be stably positioned above the appropriate detection location during the inspection process, and it possesses sufficient strength and rigidity. The L-shaped bracket 4 includes a vertical rod connected to the ground and a horizontal rod connected to the top of the vertical rod, with the vertical rod perpendicular to the horizontal rod.

[0037] The xyz direction displacement adjustment assembly is mounted on the crossbar of the L-shaped bracket 4. The xyz direction displacement adjustment assembly includes a horizontal lead screw and slider assembly mounted on the crossbar, a vertical carrier plate 8 mounted on the horizontal lead screw and slider assembly, a vertical lead screw and slider assembly mounted on the vertical carrier plate 8, and a connecting rod 11 mounted on the vertical lead screw and slider assembly. The industrial camera 5 is mounted on the connecting rod 11.

[0038] The industrial camera 5 is mounted on the xyz direction displacement adjustment component. The position of the industrial camera 5 relative to the surface of the conveyor belt 1 in the three orthogonal directions of x, y, and z is adjusted by the xyz direction displacement adjustment component to determine the optimal shooting position.

[0039] Computer terminal 6 is communicatively connected to the tag reader 2 and the industrial camera 5. Computer terminal 6 is used to acquire product information from the metal sheet 3 read by the tag reader 2, and simultaneously acquires surface photographs of the metal sheet 3 taken by the industrial camera 5, and performs defect detection based on these photographs. Computer terminal 6 can be a commercially purchased industrial control computer with built-in industry-standard image processing software and a database. The specific software is existing technology and not an innovation of this invention, so it will not be elaborated further. Computer terminal 6 receives product information transmitted by the tag reader 2 and surface photographs of the metal sheet 3 taken by the industrial camera 5. It analyzes and processes the images, uses existing image processing algorithms to identify welding defects, and associates and stores the detection results with the product information. Simultaneously, based on the preset correspondence between product information and the optimal shooting position, it sends control commands to the xyz-direction displacement adjustment component to achieve automatic and precise adjustment of the shooting position of the industrial camera 5. It is the control core and data processing center of the entire detection equipment, coordinating the work between various components, completing the welding defect detection task, and realizing the management and traceability of detection data.

[0040] A transverse groove is provided along the length of the crossbar. The transverse screw-slider assembly includes a first servo motor 9 mounted on the crossbar, a transverse screw connected to the output end of the first servo motor 9 and disposed in the transverse groove, and a first slider threadedly matched with the transverse screw. The first slider can be displaced along the transverse groove under the drive of the first servo motor 9.

[0041] The first servo motor 9 serves as the power source for the horizontal lead screw and slider assembly. It is connected to the computer terminal 6 via a control circuit, receives control signals from the computer terminal 6, and drives the horizontal lead screw to rotate, thereby causing the first slider to move within the horizontal groove of the horizontal rod. The second servo motor 10 is mounted on the vertical carrier plate 8 and drives the vertical lead screw in the vertical lead screw and slider assembly to rotate. It is communicatively connected to the computer terminal 6 and controls the second slider to move up and down within the vertical groove of the vertical carrier plate 8 according to the instructions from the computer terminal 6.

[0042] A vertical carrier plate 8 is connected to the first slider; the vertical carrier plate 8 is provided with a vertical groove. The vertical screw slider assembly includes a second servo motor 10 disposed on the vertical carrier plate 8, a vertical screw connected to the output end of the second servo motor 10 and disposed in the vertical groove, and a second slider threadedly matched with the vertical screw. The second slider can be displaced along the vertical groove under the drive of the second servo motor 10.

[0043] The connecting rod 11 is connected to the second slider, and the industrial camera 5 is located at the lower end of the connecting rod 11. Both the first servo motor 9 and the second servo motor 10 are communicatively connected to the computer terminal 6. The connecting rod 11 is typically a metal rod-shaped structure, with one end connected to the second slider of the vertical lead screw slider assembly, and the other end used to mount the industrial camera 5. Its length and shape are designed according to the installation requirements and detection range of the industrial camera 5. Its main function is to fix the industrial camera 5 in a suitable position and connect it to the vertical lead screw slider assembly, ensuring that the industrial camera 5 can adjust its position vertically as the second slider moves.

[0044] Traditional welding defect detection equipment often uses fixed positions or universal shooting settings, making it difficult to achieve personalized optimal shooting angles and positions for different products. This invention solves the problem of not being able to perform targeted and precise shooting based on the characteristics of different products by setting a label reader 2 at the beginning of the conveyor belt 1 to read the QR code labels on the metal sheet 3, and a computer terminal 6 to obtain the corresponding product information. Each product information corresponds to an optimal shooting position, and the computer terminal 6 uses the xyz direction displacement adjustment component to precisely adjust the shooting position of the industrial camera 5. This allows for better capture of accurate image information of each product surface for subsequent defect detection, improving the effectiveness and accuracy of the detection.

[0045] The above description of the embodiments is provided to enable those skilled in the art to understand and use the invention. It will be apparent to those skilled in the art that various modifications can be made to these embodiments, and the general principles described herein can be applied to other embodiments without inventive effort. Therefore, the present invention is not limited to the above embodiments, and any improvements and modifications made by those skilled in the art based on the disclosure of the present invention without departing from the scope of the invention should be within the protection scope of the present invention.

Claims

1. A vision-based welding defect detection device, characterized in that, include: Conveyor belt (1) is used to transport the welded metal sheet (3) to be inspected; A label reader (2) is located at the beginning of the conveyor belt (1) and is used to read the QR code label on the metal plate (3); L-shaped bracket (4); The xyz direction displacement adjustment component is located on the crossbar of the L-shaped bracket (4); An industrial camera (5) is mounted on the xyz direction displacement adjustment component. The position of the industrial camera (5) relative to the surface of the conveyor belt (1) in the three orthogonal directions x, y, and z is adjusted by the xyz direction displacement adjustment component to determine the optimal shooting position. The computer terminal (6) is communicatively connected to the tag reader (2) and the industrial camera (5). The computer terminal (6) is used to obtain product information of the metal sheet (3) read by the tag reader (2). At the same time, the computer terminal (6) is used to obtain surface photos of the metal sheet (3) taken by the industrial camera (5) and perform defect detection based on the surface photos of the metal sheet (3).

2. The vision-based welding defect detection device according to claim 1, characterized in that, The conveyor belt (1) is provided with support rods (7) on both sides for supporting the conveyor belt (1).

3. The vision-based welding defect detection device according to claim 1, characterized in that, The L-shaped bracket (4) includes a vertical rod connected to the ground and a horizontal rod connected to the top of the vertical rod, wherein the vertical rod is perpendicular to the horizontal rod.

4. The vision-based welding defect detection device according to claim 3, characterized in that, The xyz displacement adjustment assembly includes a horizontal lead screw and slider assembly on the crossbar, a vertical carrier plate (8) on the horizontal lead screw and slider assembly, a vertical lead screw and slider assembly on the vertical carrier plate (8), and a connecting rod (11) on the vertical lead screw and slider assembly. The industrial camera (5) is mounted on the connecting rod (11).

5. The vision-based welding defect detection device according to claim 4, characterized in that, The crossbar is provided with a transverse groove along its length.

6. The vision-based welding defect detection device according to claim 5, characterized in that, The transverse lead screw and slider assembly includes a first servo motor (9) mounted on a crossbar, a transverse lead screw connected to the output end of the first servo motor (9) and mounted in the transverse groove, and a first slider threadedly matched with the transverse lead screw. The first slider can be displaced along the transverse groove under the drive of the first servo motor (9).

7. The vision-based welding defect detection device according to claim 6, characterized in that, The vertical carrier plate (8) is connected to the first slider; The vertical carrier plate (8) is provided with a vertical sliding groove.

8. The vision-based welding defect detection device according to claim 7, characterized in that, The vertical screw and slider assembly includes a second servo motor (10) mounted on the vertical carrier plate (8), a vertical screw connected to the output end of the second servo motor (10) and mounted in the vertical groove, and a second slider threadedly matched with the vertical screw. The second slider can be displaced along the vertical groove under the drive of the second servo motor (10).

9. A vision-based welding defect detection device according to claim 8, characterized in that, The connecting rod (11) is connected to the second slider, and the industrial camera (5) is located at the lower end of the connecting rod (11).

10. A vision-based welding defect detection device according to claim 8, characterized in that, The first servo motor (9) and the second servo motor (10) are both communicatively connected to the computer terminal (6).

Images