Hot-state steel billet image dynamic tracking control method based on visual technology
By installing industrial cameras and image processing systems on the billet rolling line, dynamic tracking control of billet images is realized, solving the problems of mistriggering and signal disorder in high-speed rolling, and improving the accuracy and safety of the rolling process .
Patent Information
- Application Number
- CN202411233631.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-04
- Publication Date
- 2024-11-29
AI Technical Summary
The prior art relies on sensors in the temperature detection of steel billets, which leads to prone to false triggering and tracking signal disorders during high-speed rolling, which requires manual real-time intervention and correction, which poses safety hazards.
Using a dynamic tracking control method for hot-state billet images based on vision technology, by installing multiple industrial cameras on the rolling line, the image processing computer stitches and preprocesses the images, and uses machine vision algorithms to extract the tracking information of the billets and the information of the PLC. Perform comparisons, automatically alarms and corrections to avoid manual operations.
Improve the accuracy and safety of billet tracking, reduce the risk of manual intervention, and ensure the stability and safety of the rolling process.
Smart Images

Figure CN119035273A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the automatic control technology of medium and thick plates in the metallurgical field, and in particular to a control method for dynamic tracking of steel billet images in a temperature waiting area of a medium and thick plate finishing mill. Background Art
[0002] With the development of relevant theoretical technologies in the field of computer and control, the level of automation control of medium and thick plate rolling production lines is getting higher and higher, and high-level automation control has been achieved. However, high-level automation control relies heavily on detection instruments. For example, in the process of billet temperature detection, the position of the billet is mainly detected by hot metal detectors and thermometers installed on the rolling line, and the information is transmitted to the PLC of the automation system. The billet is tracked after automatic analysis and judgment of the information. Although this method can meet basic production needs, there are still some problems. The main problem is that the implementation of this method is to detect a certain position point through a sensor, and the information provided is limited. When the rolling rhythm is accelerated, false triggering is easy to occur due to the close distance between the billets, and there is a risk of tracking signal disorder, which requires manual intervention and correction in real time. Summary of the invention
[0003] The present invention provides a control method for dynamic tracking of hot billet images based on visual technology. The method can improve the tracking accuracy of the hot billet and automatically alarm and trigger an automatic interlocking function when the actual billet and the image are inconsistent, thereby avoiding safety hazards caused by manual operation during abnormal tracking.
[0004] To achieve the above-mentioned purpose, the present application adopts the following technical solution: a control method for dynamic tracking of hot billet images based on visual technology, characterized in that it comprises the following steps: S1: Install multiple industrial cameras above one side of the rolling line warm-up roller, so that the detection field of view formed by the multiple industrial cameras completely covers the roller range, the industrial cameras are connected to the image processing computer, and the image processing computer is connected to the PLC of the basic automation system; S2: The image processing computer seamlessly stitches the images collected by each industrial camera according to the actual size, and synthesizes the corresponding multiple images into a complete panoramic image of the roller conveyor; S3: In the production process of medium and thick plates, the grayscale histogram equalization and median filtering algorithm are used to pre-process the roller panoramic image to obtain the source image for billet edge detection; S4: Process and analyze the source image based on the machine vision algorithm, and the image processing computer extracts the tracking information of the steel billet on the roller, including the number of steel billets in the queue in the tracking area, the length of each steel billet in the queue, and the head and tail positions of each steel billet in the queue; S5: Compare the tracking information obtained by the image processing computer with the tracking information obtained by the PLC of the basic automation system: When the tracking information obtained by the two is consistent, the execution continues according to the original settings; When the tracking information obtained by the two is inconsistent, it is determined whether there is a problem of overlapping plates. If there is no overlapping plate problem and the deviation of the head or tail position of the slab obtained by the two is greater than 150mm, the image processing computer will send the obtained deviated head or tail position of the slab to the PLC, triggering the alarm flag, and the PLC will correct the corresponding tracking information according to the position sent by the image processing computer; if there is a problem of overlapping plates, the image processing computer will send a signal of overlapping plates to the PLC, and the PLC will make a chain control to prevent the overlapping plates from entering the rolling area to avoid rolling abnormalities; when the overlapping plate problem is resolved and returns to normal, the system resumes the automatic rolling function.
[0005] Furthermore, after the industrial camera is installed, the industrial camera is calibrated using the roller in the image or equipment parts of known size.
[0006] Furthermore, the PLC performs interlocking control operations to prevent the stacked plates from entering the rolling area as follows: first, the warm-up roller is controlled to suspend conveying, then a stacked plate alarm is issued to the operator, and finally the operator performs separation processing of the stacked plates.
[0007] Furthermore, the method of issuing the stacking alarm to the operator is: displaying the alarm on the operation interface and / or issuing the alarm through an audible and visual alarm.
[0008] Furthermore, the data communication between the image processing computer and the PLC of the basic automation system adopts the TCP / IP protocol, and a heartbeat detection function is added to the communication process, specifically: the image processing computer sends a heartbeat signal to the PLC at a regular interval, and the PLC needs to return confirmation data after receiving the heartbeat data; if the image processing computer times out from receiving the heartbeat confirmation data, an alarm signal is generated; when a network failure occurs, the image processing computer reinitializes the connection parameters, and after the network is normal, automatically attempts to reconnect until the network communication connection is normal.
[0009] The beneficial effect of the present invention is that the method of the present invention can be used as a verification of the tracking method of the existing automation system. The two work together to ensure that the tracking information is accurate and can take corresponding measures when an abnormality occurs, avoiding rolling abnormalities and further improving the automation level of the rolling mill area. BRIEF DESCRIPTION OF THE DRAWINGS
[0010] Figure 1 It is a system block diagram of the present invention.
[0011] In the figure, 100, roller table; 200, industrial camera; 300, image processing computer; 400, PLC. DETAILED DESCRIPTION
[0012] The technical solution in this application will be described below in conjunction with the accompanying drawings.
[0013] like Figure 1 As shown, the present invention provides a control method for dynamic tracking of hot billet images based on visual technology, and the steps are as follows.
[0014] S1: A plurality of industrial cameras 200 are installed above one side of the rolling line warm-up roller 100, so that the detection field of view jointly formed by the plurality of industrial cameras 200 completely covers the roller range, the industrial cameras 200 are connected to an image processing computer 300, and the image processing computer 300 is connected to a PLC (programmable logic controller) 400 of a basic automation system.
[0015] Specifically, the number of industrial cameras is determined according to the actual roller range that needs to be monitored, ensuring that the detection field of view can completely cover the roller range that needs to be monitored to achieve state monitoring, quantity identification and position detection of steel billets. After the industrial camera is installed, it needs to be calibrated first. The present invention can use the rollers in the image and equipment parts of known size to calibrate the industrial camera, so that the coordinates of the camera can be clarified. By calibrating the industrial camera, distortion can be eliminated and more realistic size information of the object in the image can be obtained.
[0016] S2: The image processing computer seamlessly stitches the images collected by each industrial camera according to the actual size. The stitching process uses key point correction technology to synthesize the corresponding multiple images into a complete panoramic image of the roller.
[0017] S3: In the production process of medium and thick plates, the grayscale histogram equalization and median filtering algorithm are used to preprocess the roller panoramic image to obtain an image with prominent features of the steel billet, which is used as the source image for steel billet edge detection.
[0018] S4: Process and analyze the source image based on the machine vision algorithm, and the image processing computer extracts the tracking information of the steel billet on the roller, including the number of steel billets in the queue in the tracking area, the length of each steel billet in the queue, and the head and tail positions of each steel billet in the queue; S5: Compare the tracking information obtained by the image processing computer with the tracking information obtained by the PLC of the basic automation system; when the tracking information obtained by the two is consistent, continue to execute according to the original setting.
[0019] When the tracking information obtained by the two is inconsistent, determine whether there is a problem with overlapping plates. 1. If there is no overlapping plate problem and the deviation of the head or tail position of the slab obtained by the two is greater than 150mm, the image processing computer will send the obtained position of the head or tail of the slab with deviation to the PLC, triggering the alarm flag, and the PLC will correct the corresponding tracking information according to the position sent by the image processing computer. 2. If there is a problem with overlapping plates, the image processing computer will send a signal of overlapping plates to the PLC, and the PLC will make a chain control to prevent the overlapping plates from entering the rolling area to avoid rolling abnormalities; when the overlapping plate problem is resolved and returns to normal, the system will resume the automatic rolling function.
[0020] The interlocking control operation of the PLC to prevent the stacked plates from entering the rolling area is specifically: first controlling the warming roller to stop conveying, then issuing a stacked plate alarm to the operator, and finally the operator performs the stacked plate separation process. Specifically, the stacked plate alarm is issued to the operator in the following ways: displaying an alarm on the operation interface and / or issuing an alarm through an audible and visual alarm.
[0021] Furthermore, the data communication between the image processing computer and the PLC of the basic automation system adopts the TCP / IP protocol, and a heartbeat detection function is added to the communication process, specifically: the image processing computer sends a heartbeat signal to the PLC at a regular interval, and the PLC needs to return confirmation data after receiving the heartbeat data; if the image processing computer times out from receiving the heartbeat confirmation data, an alarm signal is generated; when a network failure occurs, the image processing computer reinitializes the connection parameters, and after the network is normal, automatically attempts to reconnect until the network communication connection is normal.
[0022] The control method for dynamic tracking of hot billet images based on visual technology of the present invention can realize the functions of industrial camera for status monitoring, quantity identification and position detection of billets in the entire waiting area, and then through algorithm analysis, it can accurately provide billet position and quantity information without manual assistance throughout the process, that is, obtain stable and accurate queue tracking parameters. By comparing and verifying the billet tracking information obtained based on machine vision with the tracking information obtained by the PLC of the basic automation system, timely compensation correction or alarm can be given when billet tracking errors occur, which can improve the tracking accuracy of the slabs in the waiting area and improve the safety and reliability of production.
[0023] The above contents are only used to illustrate the technical solution of the present invention. Simple modifications or equivalent substitutions of the technical solution of the present invention made by ordinary technicians in this field do not deviate from the essence and scope of the technical solution of the present invention.
Claims
1. A control method for dynamic tracking of hot billet images based on visual technology, characterized in that: The steps include: S1: Install multiple industrial cameras above one side of the rolling line warm-up roller, so that the detection field of view formed by the multiple industrial cameras completely covers the roller range, the industrial cameras are connected to the image processing computer, and the image processing computer is connected to the PLC of the basic automation system; S2: The image processing computer seamlessly stitches the images collected by each industrial camera according to the actual size, and synthesizes the corresponding multiple images into a complete panoramic image of the roller conveyor; S3: In the production process of medium and thick plates, the grayscale histogram equalization and median filtering algorithm are used to pre-process the roller panoramic image to obtain the source image for billet edge detection; S4: Process and analyze the source image based on the machine vision algorithm, and the image processing computer extracts the tracking information of the steel billet on the roller, including the number of steel billets in the queue in the tracking area, the length of each steel billet in the queue, and the head and tail positions of each steel billet in the queue; S5: Compare the tracking information obtained by the image processing computer with the tracking information obtained by the PLC of the basic automation system: When the tracking information obtained by the two is consistent, the execution continues according to the original settings; When the tracking information obtained by the two is inconsistent, it is determined whether there is a problem of overlapping plates. If there is no overlapping plate problem and the deviation of the head or tail position of the slab obtained by the two is greater than 150mm, the image processing computer will send the obtained deviated head or tail position of the slab to the PLC, triggering the alarm flag, and the PLC will correct the corresponding tracking information according to the position sent by the image processing computer; if there is a problem of overlapping plates, the image processing computer will send a signal of overlapping plates to the PLC, and the PLC will make a chain control to prevent the overlapping plates from entering the rolling area to avoid rolling abnormalities; when the overlapping plate problem is resolved and returns to normal, the system resumes the automatic rolling function.
2. The control method for dynamic tracking of hot billet images based on visual technology according to claim 1 is characterized in that: After the industrial camera is installed, the roller in the image or equipment parts with known dimensions are used to calibrate the industrial camera.
3. The control method for dynamic tracking of hot billet images based on visual technology according to claim 1 is characterized in that: The PLC performs the interlocking control operation to prevent the stacked plates from entering the rolling area as follows: first, the warm-up roller is controlled to suspend the conveyance, then a stacked plate alarm is issued to the operator, and finally the operator performs the separation process of the stacked plates.
4. The control method for dynamic tracking of hot billet images based on visual technology according to claim 3 is characterized in that: The method of issuing the stacking alarm to the operator is: displaying the alarm on the operation interface and / or issuing the alarm through an audible and visual alarm.
5. The control method for dynamic tracking of hot billet images based on visual technology according to any one of claims 1 to 4, characterized in that: The data communication between the image processing computer and the PLC of the basic automation system adopts the TCP / IP protocol, and a heartbeat detection function is added to the communication process, specifically: the image processing computer sends a heartbeat signal to the PLC at a regular interval, and the PLC needs to return confirmation data after receiving the heartbeat data; if the image processing computer times out when receiving the heartbeat confirmation data, an alarm signal is generated; when a network failure occurs, the image processing computer reinitializes the connection parameters, and after the network is normal, automatically attempts to reconnect until the network communication connection is normal.
Citation Information
Patent Citations
Visual identification and positioning method and device for workpiece sorting and sorting system
CN114289332A
System for detecting steel blank in heating furnace based on machine vision
CN115719346A
Bar sawing area material tracking method and device
CN118351362A
Image processing device and image processing program
US20140015956A1
Cited By
Rolling motion trail monitoring method
CN121082701A
