Control method, system and equipment for strip welding alignment based on machine vision

Abstract

This invention provides a machine vision-based control method, system, and device for strip welding alignment. The method comprises: when the welding machine is ready to weld, the production unit sends a signal to a control unit; the control unit triggers an imaging unit to capture an image; an image recognition unit identifies the captured image and transmits the coordinate position information of the strip head to the control unit; and the control unit determines whether the strip head is within a specified area and notifies the production unit accordingly. This invention specifically addresses the problem of strip alignment in the fore-aft direction and utilizes machine vision as a sensor, resulting in fast recognition speed and low installation requirements.

Description

Technical Field

[0001] The present invention relates to the field of strip steel welding centering, and in particular to a control method, system and equipment for strip steel welding centering based on machine vision. Background Art

[0002] When welding with a dual-channel narrow-lap welder in a continuous annealing unit, if the leading and trailing strip heads extend beyond the welding range, a triple-stack weld will form, resulting in insufficient weld strength. This can easily cause the strip to break at the weld during coiling. This necessitates manual alignment of the strip heads with the welder's centerline before welding. However, manual alignment also presents challenges such as low efficiency, low reliability, and increased fatigue.

[0003] The key technology in automatic control is sensing technology. Due to the non-invasiveness, high reliability, high ease of use and high scalability of visual technology, the application of visual sensors in the industrial world is becoming more and more common.

[0004] Chinese invention patent CN105618596B discloses a control method and apparatus for the stitching and centering of thin strip in a continuous annealing furnace. The technical solution involves using an edge-finding probe to detect the strip edge and then calculate the strip center. A rotating hydraulic cylinder is then used to align the plane centers of the forward and backward strips and the center of the mill unit. This patent document uses an edge-finding probe to detect the strip edge, which improves both efficiency and accuracy. However, this patent document only addresses the horizontal centering of the strip during transport and does not address the front-to-back alignment of the strip.

[0005] Chinese utility model patent CN201157957Y discloses a centering device for welding and conveying stainless steel strips in a rolling mill. The device consists of a base, a work platform, a cylinder, two columns, a column crossbeam, and a roller. The work platform is mounted on the base, the cylinder is positioned between the two, the two columns are fixed to the base and positioned above the work platform, the column crossbeam is positioned between the two columns, and the roller is positioned at the stainless steel strip inlet and outlet on the work platform. The device features a pad on each inner side of the two columns, secured to the base via bearings and springs. This patent primarily focuses on the problem of strip head and tail deviation before entering the welding equipment, and similarly does not address how to properly center the strip in the front-to-back direction. Summary of the Invention

[0006] In view of the defects in the prior art, the purpose of the present invention is to provide a control method, system and equipment for strip welding alignment based on machine vision.

[0007] According to the present invention, a strip welding alignment system based on machine vision includes: an imaging unit, an edge device 6;

[0008] The edge device 6 includes a control unit and an image recognition unit; the imaging unit includes an industrial camera 5 and a light source 4;

[0009] The control unit is communicatively connected to the imaging unit, the control unit is communicatively connected to the image recognition unit, the imaging unit is communicatively connected to the image recognition unit, and the control unit is communicatively connected to the production unit;

[0010] The control unit 6 receives the positioning signal of the front strip 1 and the rear strip 2 during welding from the production unit, and triggers the imaging unit to collect images;

[0011] The imaging unit, triggered by the control unit 6, provides illumination through the light source 4 and uses the industrial camera 5 to capture position images of the preceding and following strips 1 and 2 before welding begins from the front side of the welding machine of the production unit, and transmits the images to the image recognition unit.

[0012] The image recognition unit receives the image from the imaging unit, pre-processes the image, extracts the position of the side section area of ​​the strip, and returns the strip head position information to the control unit, wherein the strip head position information includes the position information of the leading strip head and the position information of the trailing strip head;

[0013] The control unit obtains the strip head position information from the image recognition unit, determines whether the front strip head and the rear strip head are in their respective designated areas, calculates the displacement of the front strip head and the rear strip head to the center position, and sends the correction control signal to the production unit.

[0014] Preferably, after the control unit receives the position coordinates of the strip head in the image pixel coordinate system, it converts the calibrated information into the world coordinate system to obtain the position coordinates of the strip head in the world coordinate system, and determines whether the strip head is in the specified area; if it is in the specified area, the control unit sends a normal signal to the production unit, the automatic centering is completed, and the welding process starts; if it is not in the specified area, the control unit calculates the displacement of the strip head to the center position, and finally sends a correction control signal to the production unit.

[0015] Preferably, the preprocessing of the collected image performed by the image recognition unit includes extracting the effective area, graying and noise reduction, and then the image recognition unit performs image segmentation on the side cross-section area of ​​the strip to obtain the pixel coordinate position of the strip head in the image, and finally transmits the strip head coordinate position information to the control unit.

[0016] According to a strip steel welding centering method based on machine vision provided by the present invention, the strip steel welding centering system based on machine vision is used for centering control.

[0017] Preferably, it includes:

[0018] Step 1: The following strip steel stops after reaching the welding area, and the welding machine is ready for welding. At this time, the production unit sends a signal to the control unit of the strip steel welding alignment system based on machine vision;

[0019] Step 2: After receiving the signal, the control unit triggers the imaging unit of the machine vision-based strip welding alignment system to collect images, and transmits the collected images to the image recognition unit of the machine vision-based strip welding alignment system;

[0020] Step 3: After the image recognition unit pre-processes the collected image, it identifies and transmits the strip head coordinate position information to the control unit;

[0021] Step 4: After receiving the strip head coordinate position information, the control unit determines whether the strip head is within the specified area, notifies the production unit accordingly, and returns to step 1 to continue execution.

[0022] Preferably, in step 2, the industrial camera 5 and the light source 4 of the imaging unit work together to capture images of the front and side areas of the welding machine.

[0023] Preferably, in step 3, the preprocessing includes extracting the effective area, graying and denoising, and then performing image segmentation on the side section area of ​​the strip to obtain the pixel coordinate position of the strip head in the image.

[0024] Preferably, in step 4, after receiving the coordinate position information of the strip head in the image pixel coordinate system, the control unit converts the calibrated information into the world coordinate system to obtain the position coordinates of the strip head in the world coordinate system, and determines whether the strip head is within the specified area. If it is within the area, the control unit sends a normal signal to the unit, automatically completing the centering and starting the welding process. If it is not within the area, the control unit calculates the displacement of the strip head to the center position, and finally sends a correction control signal to the production unit, returning to step 1 to continue execution.

[0025] According to the present invention, a strip steel welding device is provided, which includes a production unit and the strip steel welding centering system based on machine vision, wherein the control unit of the strip steel welding centering system based on machine vision is communicatively connected to the production unit.

[0026] According to a strip steel welding device provided by the present invention, the strip steel welding centering method based on machine vision is adopted to perform centering control on the strip steel in the front-to-back direction.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] 1. By adopting intelligent visual perception technology and edge computing architecture, the present invention solves the problem of requiring manual judgment at the welding machine of the continuous annealing unit to determine whether the steel plate is aligned at the beginning of welding, thereby achieving fully automatic strip welding of the continuous annealing unit and significantly reducing the rate of weld breakage and re-welding.

[0029] 2. By adopting non-invasive machine vision intelligent perception technology, the present invention can carry out upgrades and renovations without affecting the production process and equipment operation, significantly lowering the technical threshold for introducing artificial intelligence solutions and reducing the economic losses caused by downtime due to intelligent manufacturing upgrades and renovations.

[0030] 3. The present invention specifically solves the problem of strip centering in the front-to-back direction, and uses machine vision technology in sensing technology, which has fast recognition speed and low installation threshold. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Other features, objects and advantages of the present invention will become more apparent upon reading the detailed description of non-limiting embodiments with reference to the following drawings:

[0032] Figure 1 It is a structural principle diagram of the present invention.

[0033] Figure 2 Flow chart of the method of the present invention.

[0034] The figure shows:

[0035] DETAILED DESCRIPTION

[0036] The present invention will be described in detail below with reference to specific embodiments. The following examples will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those skilled in the art, several changes and improvements can be made without departing from the scope of the present invention. These all fall within the scope of protection of the present invention.

[0037] The present invention uses machine vision to identify the position of the strip head, judges the centering condition of the strip during welding of the continuous annealing unit, and sends a control signal by communicating with the unit.

[0038] According to the present invention, a strip welding alignment system based on machine vision is provided, which includes: an imaging unit, an edge device 6, wherein the edge device 6 includes a control unit and an image recognition unit; the control unit and the imaging unit are communicatively connected, the control unit and the image recognition unit are communicatively connected, the imaging unit and the image recognition unit are communicatively connected, and the control unit and the production unit are communicatively connected.

[0039] The imaging unit is located on the side of the welding machine. The imaging unit includes an industrial camera 5 and a light source 4. It collects position images of the front strip 1 and the rear strip 2 before welding begins and transmits the images to the image recognition unit.

[0040] The control unit 6 receives the positioning signal of the front strip 1 and the rear strip 2 during welding from the production unit, triggers the imaging unit to perform image acquisition, obtains the strip head position information from the image recognition unit, including the front strip head position information and the rear strip head position information, determines whether the front strip head and the rear strip head are in the specified area, calculates the displacement of the front strip head and the rear strip head to the center position respectively, and sends the correction control signal to the production unit.

[0041] The image recognition unit receives the image from the imaging unit, pre-processes the image, extracts the position of the side section area of ​​the strip, and returns the strip head position information to the control unit.

[0042] According to the present invention, a strip welding alignment method based on machine vision is provided, comprising:

[0043] Step 1: The following strip steel stops after reaching the welding area, and the welding machine is ready for welding. At this time, the production unit sends a signal to the control unit of the strip steel welding alignment system based on machine vision.

[0044] Step 2: After receiving the signal, the control unit triggers the imaging unit of the strip welding centering system based on machine vision to perform image acquisition. The industrial camera 5 and light source 4 of the imaging unit work together to acquire images of the front and side areas of the welding machine, and transmit the acquired images to the image recognition unit of the strip welding centering system based on machine vision.

[0045] Step 3: The image recognition unit preprocesses the collected image, which includes extracting the effective area, grayscale and noise reduction, and then performing image segmentation on the side section area of ​​the strip to obtain the pixel coordinate position of the strip head in the image. Finally, the strip head coordinate position information is transmitted to the control unit.

[0046] Step 4: After receiving the position coordinates of the strip head in the image pixel coordinate system, the control unit converts the calibrated information into the world coordinate system to obtain the position coordinates of the strip head in the world coordinate system. It then determines whether the strip head is within the specified area. If so, the control unit sends a normal signal to the unit, completing automatic centering and starting the welding process. If not, the control unit calculates the displacement of the strip head to reach the center position and finally sends a correction control signal to the production unit, returning to step 1.

[0047] The above describes specific embodiments of the present invention. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art may make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. The embodiments of this application and the features in the embodiments may be combined with each other in any manner unless there is a conflict.

Claims

1. A strip welding centering system based on machine vision, characterized in that: include: Imaging units, edge devices; The edge device includes a control unit and an image recognition unit; the imaging unit includes an industrial camera and a light source; The control unit is communicatively connected to the imaging unit, the control unit is communicatively connected to the image recognition unit, the imaging unit is communicatively connected to the image recognition unit, and the control unit is communicatively connected to the production unit; The control unit receives the positioning signal of the front and rear strips during welding from the production unit and triggers the imaging unit to collect images; The imaging unit, triggered by the control unit, provides illumination through the light source and uses an industrial camera to capture position images of the preceding and following strips before welding from the front side of the welding machine of the production unit, and transmits the images to the image recognition unit. The image recognition unit receives the image from the imaging unit, pre-processes the image, extracts the position of the side section area of ​​the strip, and returns the strip head position information to the control unit, wherein the strip head position information includes the position information of the leading strip head and the position information of the trailing strip head; The control unit obtains the strip head position information from the image recognition unit, determines whether the leading and trailing strip heads are in their respective designated areas, calculates the displacement of the leading and trailing strip heads to the center position, and sends a deviation correction control signal to the production unit. After receiving the position coordinates of the strip head in the image pixel coordinate system, the control unit converts the calibrated information into the world coordinate system, obtains the position coordinates of the strip head in the world coordinate system, and determines whether the strip head is within the specified area. If it is within the specified area, the control unit sends a normal signal to the production unit, automatically completes the centering, and starts the welding process. If it is not within the specified area, the control unit calculates the displacement of the strip head to the center position and finally sends a correction control signal to the production unit. The image recognition unit performs preprocessing on the collected image, including extracting the effective area, graying and reducing noise. Then, the image recognition unit performs image segmentation on the side cross-section area of ​​the strip to obtain the pixel coordinate position of the strip head in the image, and finally transmits the strip head coordinate position information to the control unit.

2. A strip welding centering method based on machine vision, characterized in that: The strip welding centering system based on machine vision as described in claim 1 is used for centering control.

3. The strip welding alignment method based on machine vision according to claim 2, characterized in that: include: Step 1: The following strip steel stops after reaching the welding area, and the welding machine is ready for welding. At this time, the production unit sends a signal to the control unit of the machine vision-based strip steel welding alignment system according to claim 1; Step 2: After receiving the signal, the control unit triggers the imaging unit of the machine vision-based strip welding alignment system to collect images, and transmits the collected images to the image recognition unit of the machine vision-based strip welding alignment system; Step 3: After the image recognition unit pre-processes the collected image, it identifies and transmits the strip head coordinate position information to the control unit; Step 4: After receiving the strip head coordinate position information, the control unit determines whether the strip head is within the specified area, notifies the production unit accordingly, and returns to step 1 to continue execution.

4. The strip welding alignment method based on machine vision according to claim 3 is characterized in that: In step 2, the industrial camera and light source of the imaging unit work together to capture images of the front and side areas of the welding machine.

5. The strip welding alignment method based on machine vision according to claim 3 is characterized in that: In step 3, the preprocessing includes extracting the effective area, graying and denoising, and then performing image segmentation on the side section area of ​​the strip to obtain the pixel coordinate position of the strip head in the image.

6. The strip welding centering method based on machine vision according to claim 3 is characterized in that: In step 4, after the control unit receives the coordinate position information of the strip head in the image pixel coordinate system, it converts the calibrated information into the world coordinate system to obtain the position coordinates of the strip head in the world coordinate system, and determines whether the strip head is within the specified area. If it is within the area, the control unit sends a normal signal to the unit, the automatic centering is completed, and the welding process begins. If it is not within the area, the control unit calculates the displacement of the strip head to the center position, and finally sends a correction control signal to the production unit, and returns to step 1 to continue execution.

7. A strip steel welding device, characterized in that: It comprises a production unit and the machine vision-based strip welding centering system according to claim 1, wherein the control unit of the machine vision-based strip welding centering system is communicatively connected to the production unit.

8. A strip steel welding device, characterized in that: The strip welding centering method based on machine vision as described in claim 2 is used to control the centering of the strip in the front-to-back direction.

Images