Detection method for strip head positioning of cold rolling recoiling machine and related equipment

Abstract

The invention provides a cold rolling coiler strip head positioning detection method and related equipment, and relates to the technical field of target detection, the method comprises the following steps: obtaining a strip head positioning image of strip steel; the strip head positioning image is recognized based on a detection model, a recognition result is obtained, and the recognition result is used for judging whether strip head positioning of the strip steel is correct or not; and under the condition that the identification result shows that the strip head of the strip steel is correctly positioned, coiling operation of the strip steel is executed.

Description

Technical Field

[0001] The present application relates to the field of target detection technology, and in particular to a detection method for cold rolling coiler head positioning and related equipment. Background Art

[0002] On the cold rolling production line, the accurate positioning of the strip head during the strip coiling process is the key link to ensure the coiling quality. The accuracy of the strip head positioning directly affects the coiling effect. If the positioning is improper, "strip head marks" or other defects may be formed during the coiling process, which will have an adverse effect on product quality.

[0003] Although the existing cold rolling production line is equipped with a lead positioning device, the problem of inaccurate positioning still exists in actual production. At present, most positioning detection relies on manual observation, which is not only inefficient, but also consumes a lot of human resources, and is difficult to achieve continuous monitoring, and cannot meet the high precision and high efficiency requirements of modern industrial production. Therefore, a lead positioning detection method is urgently needed to solve the above problems. Summary of the invention

[0004] A series of simplified concepts are introduced in the Summary of the Invention section, which will be further described in detail in the Detailed Description of the Invention section. The Summary of the Invention section of this application does not mean to attempt to define the key features and essential technical features of the claimed technical solution, nor does it mean to attempt to determine the scope of protection of the claimed technical solution.

[0005] In a first aspect, the present application provides a method for detecting the head positioning of a cold rolling coiler, comprising:

[0006] Acquire the strip head positioning image of the strip;

[0007] Recognize the strip head positioning image based on the detection model to obtain a recognition result, wherein the recognition result is used to determine whether the strip head positioning of the strip is correct;

[0008] When the above identification result indicates that the strip head of the above strip is positioned correctly, the coiling operation of the above strip is performed.

[0009] In a feasible implementation, it also includes:

[0010] Obtaining a historical strip head positioning image of the strip steel;

[0011] Based on the above historical leading positioning images, the network model is trained to obtain the above detection model.

[0012] In a feasible implementation, it also includes:

[0013] The above network model is a YOLOv8 network model, wherein the loss function of the above network model is a CIOU loss function.

[0014] In a feasible implementation manner, the above-mentioned obtaining of the historical strip head positioning image of the above-mentioned strip includes:

[0015] Based on the speed information of the cold rolling production line, a smooth transition algorithm is used to dynamically adjust the camera's acquisition frame rate to obtain the historical strip head positioning image of the above-mentioned strip.

[0016] In a feasible implementation manner, the above-mentioned head positioning image is recognized based on the detection model to obtain a recognition result, including:

[0017] Based on the output result of the above detection model, determine the leading bounding box information and the positioning area bounding box information, wherein the leading bounding box information includes the center point coordinates of the leading bounding box and the height of the leading bounding box, the center point coordinates of the leading bounding box include the ordinate of the center point of the leading bounding box, the positioning area bounding box information includes the center point coordinates of the positioning area bounding box and the height of the positioning area bounding box, and the center point coordinates of the positioning area bounding box include the ordinate of the center point of the positioning area bounding box;

[0018] Based on the ordinate of the center point of the tape head boundary box and the height of the tape head boundary box, the ordinate of the upper surface of the tape head is calculated, wherein the ordinate of the upper surface of the tape head is the longitudinal coordinate between the first contact of the tape head with the winding area and the winding machine core shaft;

[0019] Calculate the upper surface ordinate of the positioning area and the lower surface ordinate of the positioning area based on the ordinate of the center point of the positioning area boundary box and the height of the positioning area boundary box;

[0020] Based on the ordinate of the upper surface of the tape head, the ordinate of the upper surface of the positioning area and the ordinate of the lower surface of the positioning area, a recognition result is obtained.

[0021] In a feasible implementation manner, the identification result obtained based on the upper surface ordinate of the tape head, the upper surface ordinate of the positioning area, and the lower surface ordinate of the positioning area includes:

[0022] When the longitudinal coordinate of the upper surface of the tape head is greater than or equal to the first preset value and less than or equal to the second preset value, the recognition result is normal;

[0023] When the longitudinal coordinate of the upper surface of the belt head is less than the first preset value or greater than the second preset value, the recognition result is abnormal and an alarm mechanism is triggered;

[0024] The first preset value is an interpolation value of the upper surface vertical coordinate of the positioning area and the deviation value, and the second preset value is a sum of the lower surface vertical coordinate of the positioning area and the deviation value.

[0025] In a feasible implementation, it also includes:

[0026] In the case where the above identification result is abnormal, the tape head positioning detection interface is controlled to display the tape head positioning image and alarm information, wherein the tape head positioning detection interface includes the start time, end time, cold rolling production line and uncoiling machine number.

[0027] In the second aspect, the present application proposes a detection device for the head positioning of a cold rolling coiler, comprising:

[0028] An image acquisition unit, used for acquiring a strip head positioning image of the strip;

[0029] A strip head recognition unit, used to recognize the strip head positioning image based on the detection model to obtain a recognition result, wherein the recognition result is used to determine whether the strip head positioning of the strip is correct;

[0030] The strip coiling unit is used to perform the coiling operation of the strip when the above identification result indicates that the strip head of the above strip is correctly positioned.

[0031] In a third aspect, an electronic device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor is used to implement the steps of a method for detecting the head positioning of a cold rolling coiler as described in any one of the first aspects above when executing the computer program stored in the memory.

[0032] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for detecting the head positioning of a cold rolling coiler as described in any one of the first aspects above is implemented.

[0033] In summary, the embodiment of the present invention provides a detection method for the lead positioning of a cold rolling coiler, and realizes the automation and high precision of the lead positioning detection through deep learning technology. The YOLOv8 neural network model combined with the CIOU loss function enables the system to accurately identify the position of the lead and the positioning area, realizes automatic positioning detection, greatly reduces the error and labor cost caused by traditional manual detection, and improves production efficiency. The system has real-time feedback and alarm functions, and can immediately issue an alarm when the lead positioning deviation exceeds the set threshold, display detailed alarm information, and facilitate operators to quickly adjust, thereby ensuring the continuous and stable operation of the cold rolling production line and ensuring product quality consistency. In addition, the system designs a dynamic frame rate adjustment strategy to adjust the camera acquisition frequency in real time according to the production line speed, ensure that high-quality images are captured at critical moments, and avoid data redundancy, thereby realizing efficient data acquisition and storage management. The detection system is based on a modular design of deep learning, has excellent maintainability and scalability, and can further optimize the detection effect by retraining the model, bringing significant economic benefits and reliable production guarantees to enterprises.

[0034] The detection method for the cold rolling coiler head positioning proposed by the present invention, other advantages, objectives and features of the present invention will be reflected in part through the following description, and in part will be understood by technical personnel in the field through research and practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present specification. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0036] Figure 1 A schematic flow chart of a method for detecting the head positioning of a cold rolling coiler provided in an embodiment of the present application;

[0037] Figure 2 A schematic diagram of the recognition result of the leading positioning detection model provided in the embodiment of the present application;

[0038] Figure 3 A schematic diagram of a lead positioning detection interface provided in an embodiment of the present application;

[0039] Figure 4 A schematic diagram of the structure of a detection device for head positioning of a cold rolling coiler provided in an embodiment of the present application;

[0040] Figure 5 A schematic diagram of the structure of an electronic device for detecting the head positioning of a cold rolling coiler provided in an embodiment of the present application. DETAILED DESCRIPTION

[0041] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices. The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments.

[0042] See also Figure 1 , which is a flow chart of a detection method for the head positioning of a cold rolling coiler provided in an embodiment of the present application, which may specifically include:

[0043] S110, obtaining a strip head positioning image of the strip;

[0044] For example, the acquisition of the strip head positioning image relies on a high-definition camera installed on the coiling line, which monitors the starting part of the strip (i.e. the strip head) and the positioning area in real time. The camera uses an image sensor with a high frame rate and a wide dynamic range, which can adapt to the high-speed operation of the cold-rolled strip and capture the detailed information of the strip head under different working conditions. The camera parameters are also intelligently adjusted according to the changes in light and temperature in the cold rolling environment, such as automatically adjusting the aperture and shutter speed, to ensure that clear images are captured in complex environments.

[0045] S120, recognizing the strip head positioning image based on the detection model to obtain a recognition result, wherein the recognition result is used to determine whether the strip head positioning of the strip is correct;

[0046] Exemplarily, the acquired lead positioning image is analyzed based on the trained target detection model to accurately identify the relative position of the lead and the positioning area. The detection model uses the YOLOv8 deep learning network structure to generate bounding box information of the lead and the positioning area, including the center point coordinates, width and height, through real-time analysis of the lead positioning image. The detection model extracts the specific position coordinates of the lead and the positioning area from the image, providing a reliable data basis for the subsequent judgment of positioning accuracy.

[0047] S130. When the identification result indicates that the strip head is positioned correctly, coiling the strip is performed.

[0048] For example, when the recognition result of the detection model indicates that the strip head positioning meets the preset accuracy requirements, the strip can smoothly enter the coiler and maintain a high-quality winding effect. Specifically, the recognition result analyzes the distance relationship between the strip head and the positioning area through a deviation comparison algorithm. If the deviation value is within the allowable threshold range (such as the upper surface of the strip head is within the preset deviation range of the upper and lower surfaces of the positioning area), it is confirmed that the strip head is positioned correctly.

[0049] When the strip head is positioned correctly, the coiler will coil the strip onto the coil core according to the established process and parameters. In the coiling operation, the precise initial positioning of the strip head is crucial to the stability of the subsequent winding, because the alignment between the strip head and the coiler directly affects the tightness and flatness of the winding.

[0050] In summary, the embodiment of the present invention provides a detection method for the lead positioning of a cold rolling coiler, which realizes the automation and high precision of the lead positioning detection through deep learning technology. The YOLOv8 neural network model is combined with the CIOU loss function to accurately identify the position of the lead and the positioning area, realize automatic positioning detection, greatly reduce the error and labor cost caused by traditional manual detection, and improve production efficiency.

[0051] In some examples, it also includes:

[0052] Obtaining a historical strip head positioning image of the strip steel;

[0053] Based on the above historical leading positioning images, the network model is trained to obtain the above detection model.

[0054] The above network model is a YOLOv8 network model, wherein the loss function of the above network model is a CIOU loss function.

[0055] For example, the high-definition camera installed on the strip production line collects the historical strip head positioning images and saves them in the image database. These historical data include the positioning images of the strip head under different production conditions, as well as its changes under different light, speed and angle conditions. This variety of data provides rich samples for subsequent model training and improves the adaptability of the model in actual production scenarios.

[0056] Through the automated video acquisition program, the strip head image is acquired in real time on the production line and saved in the database at set intervals. Specifically, the camera will dynamically adjust the acquisition frequency according to the speed of the production line to ensure high-precision acquisition at critical moments. At the same time, the image will be divided into different status labels, such as "normal positioning" or "positioning deviation", to provide a basis for subsequent annotation and training.

[0057] After the historical image data is collected, the image data will be processed, including resizing, contrast enhancement, etc. Labeling tools (such as LabelImg) are used to mark the leading and positioning areas on the image, and a label file for training is generated. After classification, these data are divided into training set, validation set and test set in a ratio of 8:1:1, laying the foundation for effective model training.

[0058] Through the historical leading positioning images obtained, these labeled data are used to train the network model to improve the adaptability and detection accuracy of the detection model in the actual production environment. In order to meet the real-time detection needs of the cold rolling production environment, the YOLOv8 network model was selected as the core detection model.

[0059] The YOLOv8 model structure mainly consists of three parts: Backbone (for feature extraction), Neck (feature fusion) and Head (bounding box prediction). In head positioning detection, Backbone extracts the features of the head and positioning area, Neck fuses information of different scales through the feature pyramid network to improve the model's detection ability for heads of different sizes, and Head finally outputs the bounding box, category confidence and positioning results.

[0060] During the model training process, the CIOU (Complete Intersection over Union) loss function is used to optimize the model parameters. Unlike the traditional IoU (Intersection over Union) loss function, the CIOU loss not only considers the overlapping area of ​​the bounding box of the leader and the positioning area when calculating the bounding box error, but also measures factors such as the center point distance and aspect ratio. This method can provide more accurate error feedback in leader positioning detection, which helps the model converge quickly and improves detection accuracy. This method optimizes the detection accuracy and adaptability of the model. In particular, when the leader position is slightly deviated, CIOU can detect the error more sensitively, improving the model's tolerance to deviations and detection accuracy.

[0061] Driven by historical data, the model is tuned in combination with the actual production situation of strip steel. During the training process, the model gradually becomes stable by adjusting hyperparameters such as learning rate and batch size, and dynamically adjusting the learning rate in combination with the cosine annealing strategy. The specific training process is as follows:

[0062] The pre-trained model of YOLOv8 is used as the basis to reduce the training time caused by cold start. Then the leading positioning dataset is loaded for fine-tuning training.

[0063] The initial learning rate is set to 0.001, the batch size is 16, the number of training times is 150, and the Adam optimizer is used to dynamically update each parameter. The model automatically adjusts the learning rate and optimization strategy in each round of training to achieve better accuracy and convergence.

[0064] After each round of training, the performance of the model is evaluated using the validation set, and the training effect of the model is judged by the reduction of CIOU loss and the improvement of recognition accuracy. If the accuracy is found to be insufficient, the model structure and training parameters are adjusted to continue optimization to ensure the high accuracy of the model in actual detection.

[0065] After training, the model will be deployed to the lead positioning detection system to detect the lead positioning in the production process in real time. At the same time, the system retains the online update mechanism, adding newly collected historical data to the model training set, and regularly retraining or fine-tuning the model, so as to achieve continuous learning and accuracy improvement of the detection system.

[0066] By acquiring historical lead positioning images and training and optimizing the detection model based on them, this implementation not only improves the detection accuracy and robustness of the YOLOv8 model, but also ensures that the detection system can adapt to the lead positioning requirements under a variety of production environments and different production parameters, providing an efficient and intelligent lead positioning solution for the cold rolling production line.

[0067] In some examples, the step of obtaining the historical strip head positioning image of the strip steel includes:

[0068] Based on the speed information of the cold rolling production line, a smooth transition algorithm is used to dynamically adjust the camera's acquisition frame rate to obtain the historical strip head positioning image of the above-mentioned strip.

[0069] For example, in the production process of cold-rolled steel strip, the running speed of the steel strip changes dynamically. In order to ensure that the camera can capture high-precision images at the critical moment of strip head positioning, and reduce the amount of data collection and optimize storage at non-critical moments, the camera acquisition frame rate is adjusted based on the information of the production line speed. This adjustment process uses a smooth transition algorithm to ensure the smoothness and stability of image acquisition, and mainly includes the following two modules:

[0070] The production line speed monitoring module is used to collect the speed data of the production line in real time as a reference for frame rate adjustment. Real-time production line speed information is obtained through the data interface of the sensor or control system. These data can come from the PLC system of the coiler, speed sensor or other measuring equipment, and accurately feedback the speed of the strip movement and the state of the strip head approaching the coiler.

[0071] The frame rate control module adjusts the frame rate of the camera according to the real-time speed to ensure the continuity and accuracy of image acquisition. According to the speed information transmitted by the production line speed monitoring module, the appropriate frame rate is dynamically calculated and the camera acquisition frequency is adjusted in real time.

[0072] The core of the frame rate adjustment strategy is based on the actual speed of the production line and the movement trajectory of the leader. The system divides the production line into different key areas and uses dynamic frame rate adjustment to ensure that images can be captured at a higher frame rate when the leader enters the key area, while maintaining a low frame rate acquisition at other times to reduce storage pressure.

[0073] The frame rate-speed mapping function is introduced to achieve frame rate optimization, and the frame rate of the camera is adjusted by the head movement speed. The speed of the production line is v (unit: m / s), and the frame rate is F (unit: frame / s). The mapping relationship can be designed into the following two parts according to the needs of different stages of the production line:

[0074] High-speed section (when the strip head is about to enter the coiler or is already in the critical operating area):

[0075]

[0076] Among them, v current is the current production line speed; v max F is the maximum speed during operation; max is the maximum frame rate of the camera; F min is the minimum frame rate of the camera; in this section, adjust the frame rate to the maximum value F max , usually set to 60 frames per second or even higher depending on the performance of the camera. At this time, the acquisition frequency is the maximum to ensure that no key frames are missed.

[0077] Low speed section (when the strip head is far away from the coiler or in a non-critical area):

[0078]

[0079] When the line speed is slow or the lead is away from the critical area, the frame rate is reduced to a minimum. min , for example 5 frames / second, to avoid the generation of excessive data.

[0080] The frame rate adjustment is more intelligent, and a mechanism based on regional triggering is designed. By real-time monitoring of the distance between the belt head and the coiler, the production line is divided into different collection areas, and each area triggers a different frame rate adjustment strategy: Area A is the early warning area (the belt head is far from the coiler). In this area, the belt head has not yet entered the key area of ​​coiling, and the frame rate can be adjusted to a lower value to save storage resources. At this time, the frame rate is F min Area B is a high frame rate area (the strip head is close to the coiler). When the strip head enters the predetermined key area (such as less than 1 meter from the coiler), the frame rate is adjusted to the maximum value F. max , ensuring that all details of the strip head entering the coiler are captured. Area C is the release area (the strip head leaves the coiler). When the strip head successfully enters the coiler and is wound, the frame rate gradually drops to the medium value F med , to avoid recording unnecessary redundant data.

[0081] To avoid abrupt visual effects or data loss when switching frame rates, a frame rate smoothing transition algorithm is introduced. The frame rate will not jump directly, but will be gradually adjusted through a smoothing function to ensure smooth camera acquisition. The specific smoothing adjustment process is as follows:

[0082] F(t)=F prev +(F new -F prev )·(1-e- α·t )

[0083] Among them, F prev is the current frame rate, F new is the target frame rate, α is the control parameter of the smooth adjustment speed, by adjusting α, the smoothness of the frame rate adjustment can be controlled, and t is the time step.

[0084] To further improve the accuracy of frame rate adjustment, a real-time feedback and self-correction mechanism is designed. After each frame rate adjustment, the camera will analyze the quality of the captured image in real time and compare it with the expected value. If the image is blurred or frames are lost, the frame rate will be increased and the corresponding camera parameters will be readjusted; if the image quality meets expectations, the current frame rate will be maintained. This self-correction mechanism can ensure the stability of image acquisition and avoid data quality degradation caused by frame rate mismatch.

[0085] Based on the frame rate adjustment strategy, a data compression and storage optimization mechanism is also introduced. In the high frame rate area, the image is stored with high quality, while in the low frame rate area, the image compression algorithm is used to reduce the storage demand and ensure that the data storage and retrieval are more efficient.

[0086] Through the above mechanism, the acquisition frame rate of the camera can be dynamically adjusted based on the change of the speed of the cold rolling production line and a smooth transition algorithm can be used to ensure that high-precision historical lead positioning image data is obtained in the key areas of the strip production line. This method not only improves the accuracy of image acquisition, but also reduces the data burden through data compression and storage optimization, making the collected data more efficient and more in line with production needs.

[0087] In some examples, the above-mentioned head positioning image is recognized based on the detection model to obtain a recognition result, including:

[0088] Based on the output result of the above detection model, determine the leading bounding box information and the positioning area bounding box information, wherein the leading bounding box information includes the center point coordinates of the leading bounding box and the height of the leading bounding box, the center point coordinates of the leading bounding box include the ordinate of the center point of the leading bounding box, the positioning area bounding box information includes the center point coordinates of the positioning area bounding box and the height of the positioning area bounding box, and the center point coordinates of the positioning area bounding box include the ordinate of the center point of the positioning area bounding box;

[0089] Based on the ordinate of the center point of the tape head boundary box and the height of the tape head boundary box, the ordinate of the upper surface of the tape head is calculated, wherein the ordinate of the upper surface of the tape head is the longitudinal coordinate between the first contact of the tape head with the winding area and the winding machine core shaft;

[0090] Calculate the upper surface ordinate of the positioning area and the lower surface ordinate of the positioning area based on the ordinate of the center point of the positioning area boundary box and the height of the positioning area boundary box;

[0091] Based on the ordinate of the upper surface of the tape head, the ordinate of the upper surface of the positioning area and the ordinate of the lower surface of the positioning area, a recognition result is obtained.

[0092] For example, in the tape head positioning detection system, based on the tape head information and positioning area information output by the deep learning detection model (such as YOLOv8), a series of coordinate calculations are performed to determine whether the tape head is accurately positioned in the winding area. This process is further extended to trigger an alarm mechanism based on the detection results to ensure positioning accuracy, and to enhance the fault tolerance of the system by setting preset upper and lower limits, such as Figure 2 As shown in FIG. 1 , it is the recognition result of the head positioning detection model, where head represents the head and line represents the positioning area.

[0093] The detection model analyzes the head positioning image and outputs the bounding box information of the head and positioning area. The bounding box contains the center point coordinates, width and height, providing accurate data support for subsequent coordinate calculation and positioning judgment. The center point coordinates are represented by the horizontal coordinate x and the vertical coordinate y, which are the geometric center positions of the head and positioning area respectively; the width w and the height h represent the horizontal and vertical size ranges of the head and positioning area respectively; the upper and lower boundary positions of the head and positioning area can be calculated through the center point position and the bounding box size.

[0094] The upper surface of the strip head is the first place where the strip head contacts the winding area, so it is very important to calculate this position accurately. Based on the center point and height of the strip head output by the model, the ordinate of the upper surface of the strip head is expressed as:

[0095]

[0096] Among them, y 1 h is the ordinate of the center point of the bounding box output by the YOLOv8 model; 1 y is the longitudinal dimension of the belt head; head-top It is the longitudinal position when the tape head first contacts the winding area during the winding process. This value will be used to determine the precise positioning status of the tape head in the subsequent comparison process.

[0097] The positioning area is the designated position where the leader must fall accurately. In order to determine the accuracy of positioning, the upper and lower surface vertical coordinates of the positioning area are calculated through the positioning area bounding box information output by the model, expressed as:

[0098]

[0099] Among them, y 2 h is the vertical coordinate of the center point of the positioning area bounding box; 2 The vertical dimension of the positioning area; region-top y is the vertical coordinate of the upper surface of the positioning area; region-bottom It is the ordinate of the lower surface of the positioning area; the upper surface of the positioning area is the boundary of the area closest to the winding machine core shaft in the longitudinal direction, and the ordinate of the upper surface is calculated by subtracting half of the height of the area from the ordinate of the center point of the positioning area; the lower surface of the positioning area is the boundary of the area farthest from the winding machine core shaft in the longitudinal direction, and the ordinate of the lower surface is calculated by adding half of the height of the area to the ordinate of the center point of the positioning area.

[0100] For example, the reason why the ordinate is used for calculation instead of the abscissa in this embodiment is that the key requirements for strip head positioning are mainly concentrated in the vertical direction (i.e., longitudinal direction) of the strip. This precise control in the vertical direction is to ensure that the strip head is accurately aligned with the mandrel when entering the coiler, thereby achieving stable winding. Therefore, the ordinate plays a key role in determining the positioning accuracy, including:

[0101] The movement of the strip in the cold rolling production line is continuously transmitted in the horizontal direction (i.e., transverse direction), and the core of strip head positioning is to ensure that the strip head is accurately aligned in the vertical direction (longitudinal direction). When the strip enters the coiler, the main risk of deviation is in the vertical direction, so focusing on the vertical axis can more accurately reflect the actual deviation between the strip head and the positioning area.

[0102] The coiling process requires that the strip head and the mandrel are accurately connected to ensure that the strip can be wound smoothly and evenly on the mandrel. If the strip head deviates too much in the longitudinal direction, the strip head may not contact the mandrel or will produce excessive tension, resulting in coiling defects (such as wrinkles, slippage, etc.). The transverse coordinate has limited influence on the initial position of the strip head in the coiling, so it has no decisive role in positioning judgment.

[0103] The longitudinal position of the winding area and the mandrel is relatively fixed. By detecting whether the tape head is in the correct upper and lower boundaries in the longitudinal direction, it can be effectively determined whether it is within the positioning area. This longitudinal upper and lower boundary judgment is to ensure the longitudinal accuracy of the tape head entering the winding area, so that the tape head can contact the mandrel at the first time to ensure smooth winding.

[0104] In some examples, the upper surface ordinate of the tape head, the upper surface ordinate of the positioning area, and the lower surface ordinate of the positioning area are used to obtain the identification result, including:

[0105] When the longitudinal coordinate of the upper surface of the tape head is greater than or equal to the first preset value and less than or equal to the second preset value, the recognition result is normal;

[0106] When the longitudinal coordinate of the upper surface of the belt head is less than the first preset value or greater than the second preset value, the recognition result is abnormal and an alarm mechanism is triggered;

[0107] The first preset value is an interpolation value of the upper surface vertical coordinate of the positioning area and the deviation value, and the second preset value is a sum of the lower surface vertical coordinate of the positioning area and the deviation value.

[0108] For example, in order to enhance fault tolerance, upper and lower limit preset values ​​are introduced to judge the normality and abnormality of the lead positioning. The deviation value is the allowable deviation value set by the system, taking into account the slight fluctuations in the production process, so that the system can provide a certain tolerance for normal positioning at the boundary of the positioning range. In this embodiment, the first preset value is y th-top =y region-top -δ, the second preset value is y th-bottom =y region-botton +δ, where the deviation value δ can be set to 5.

[0109] The upper surface ordinate of the belt head is positioned within the preset value range, expressed as:

[0110] When the ordinate y of the upper surface of the belt head head-top The following conditions are met,

[0111] y th-top ≤y head-top ≤y th-bottom

[0112] It is determined that the tape head positioning is normal and the winding operation can continue.

[0113] When head-top Outside the above range, that is, less than y th-top or greater than y th-bottom , it is determined that the lead positioning is abnormal and an alarm needs to be triggered in time to avoid quality defects during the winding process. At this time, the system sets the recognition result as abnormal and prompts the operator to check the positioning.

[0114] In some examples, it also includes:

[0115] When the above identification result is abnormal, the tape head positioning detection interface is controlled to display the tape head positioning image and alarm information, wherein the tape head positioning detection interface includes the start time, end time, cold rolling production line and uncoiling machine number.

[0116] For example, in belt head positioning detection, when the recognition result is judged to be abnormal, the system will display the belt head positioning image and detailed alarm information on the belt head positioning detection interface. This alarm mechanism is designed to provide operators with immediate abnormal feedback to facilitate rapid identification and processing of belt head positioning deviations. The system uses model calculations to determine whether the belt head positioning deviation exceeds the preset range (that is, the vertical coordinate of the upper surface of the belt head does not fall between the upper and lower surfaces of the positioning area). If it is judged to be "abnormal", the system will display the belt head positioning image and alarm information on the detection interface to help operators quickly understand the deviation. Figure 3 The figure shows the lead positioning detection interface.

[0117] The system displays the current belt head positioning image on the detection interface. The image contains the bounding box of the belt head and the positioning area, and uses different colors or marks to distinguish the positioning area and the belt head, so that the deviation is intuitively visible. The image also shows the upper surface of the belt head and the upper and lower boundaries of the positioning area, which is convenient for operators to clearly understand the location and degree of deviation. If the positioning deviation exceeds the allowable range, the system generates an alarm message and displays it together with the belt head positioning image. The alarm message contains rich production process data to help operators quickly locate and deal with abnormal situations. The alarm information of the belt head positioning detection interface contains multiple dimensions of key production data, ensuring that operators obtain complete background information to quickly and accurately deal with problems.

[0118] The alarm information displays the specific time period of the positioning detection, including the start and end time of the positioning detection. The specific time point record of the abnormality makes it easier for operators to trace and analyze the cause of the deviation in the lead positioning. At the same time, other records related to this time point can be queried in the log. The alarm information also displays the number of the cold rolling production line where the current abnormal event is located. There may be multiple production line operations on the production line. The accurate production line number helps the system quickly locate the specific operation location, making problem tracking more efficient. In addition, the uncoiler number is also displayed, which is the specific equipment number involved in this lead positioning abnormality event. Because a cold rolling production line is usually equipped with multiple uncoilers, the uncoiler number helps to further narrow the scope of investigation. The operator can directly check the configuration of the uncoiler or the lead entry path to ensure that the problem is quickly located and handled.

[0119] To facilitate intuitive understanding by operators, the lead positioning detection interface adopts a clear visual design, and the image is highlighted on the detection interface. The alarm information contains text prompts, explaining the cause of the abnormal positioning and the recommended processing steps. Detailed information such as "start time", "end time", "cold rolling production line number" and "unwinding machine number" are displayed next to the image in a table or annotation form to ensure that the information is accurate and comprehensive.

[0120] While displaying the alarm information, the system also records the data of the abnormal event into the database for subsequent quality traceability and improvement analysis. The alarm information displayed on the detection interface will be stored to form a detailed abnormal record, including the lead positioning image, alarm time, deviation information, etc., for future retrospective inspection to help the production team analyze the cause of the positioning problem. Through the saved alarm data, the system can carry out historical data analysis to find out the common causes or trends of lead positioning deviations, thereby helping to optimize the lead positioning process of the production line. When the operator handles the abnormal positioning problem and manually releases the alarm, the system will clear the alarm information on the detection interface and reset the lead positioning detection status. After the alarm is released, the system resumes the normal detection process to ensure that the system monitors the status of the lead positioning in real time. This reset function ensures the continuity and efficiency of the detection system.

[0121] See also Figure 4 , which is a schematic structural diagram of a detection device for head positioning of a cold rolling coiler provided in an embodiment of the present application, comprising:

[0122] An image acquisition unit 21 is used to acquire a strip head positioning image of the strip;

[0123] The strip head recognition unit 22 is used to recognize the strip head positioning image based on the detection model to obtain a recognition result, wherein the recognition result is used to determine whether the strip head positioning of the strip is correct;

[0124] The strip coiling unit 23 is used to perform the strip coiling operation when the identification result indicates that the strip head of the strip is correctly positioned.

[0125] See also Figure 5 The embodiment of the present application also provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor. When the processor 320 executes the computer program 311, the steps of any method for detecting the positioning of the head of the cold rolling coiler are implemented.

[0126] Since the electronic device introduced in this embodiment is the equipment used to implement the detection device for the head positioning of the cold rolling coiler in the embodiment of the present application, based on the method introduced in the embodiment of the present application, the technical personnel in this field can understand the specific implementation mode of the electronic device of this embodiment and its various variations. Therefore, how the electronic device implements the method in the embodiment of the present application is not introduced in detail here. As long as the equipment used by the technical personnel in this field to implement the method in the embodiment of the present application is within the scope of protection of this application.

[0127] During the specific implementation process, when the computer program 311 is executed by a processor, any implementation method in the embodiments corresponding to the first aspect can be implemented.

[0128] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0129] Those skilled in the art will appreciate that the embodiments of the present application may provide methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-readable program code.

[0130] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0131] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0132] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0133] The present application also provides a computer program product, which includes computer software instructions. When the computer software instructions are executed on a processing device, the processing device executes Figure 1 The process of the detection method for the cold rolling coiler head positioning in the corresponding embodiment.

[0134] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can store or a data storage device such as a server or a data center that includes one or more available media integration. The available medium can be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state drive (SSD)), etc.

[0135] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0136] In the several embodiments provided in the present application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0137] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0138] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0139] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), disk or optical disk and other media that can store program codes.

[0140] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

[0141] Although the preferred embodiments of this specification have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of this specification.

[0142] Obviously, those skilled in the art can make various changes and modifications to this specification without departing from the spirit and scope of this specification. Thus, if these modifications and variations of this specification fall within the scope of the claims of this specification and their equivalents, this specification is also intended to include these modifications and variations.

Claims

1. A method for detecting the head positioning of a cold rolling coiler, characterized in that: The method comprises: Acquire the strip head positioning image of the strip; Recognize the strip head positioning image based on the detection model to obtain a recognition result, wherein the recognition result is used to determine whether the strip head positioning of the strip is correct; When the identification result indicates that the strip head of the steel strip is positioned correctly, the coiling operation of the steel strip is performed.

2. The method for detecting the head positioning of the cold rolling coiler according to claim 1, characterized in that: Also includes: Acquire a historical strip head positioning image of the strip; Based on the historical leading positioning image, the network model is trained to obtain the detection model.

3. The method for detecting the head positioning of the cold rolling coiler according to claim 2, characterized in that: Also includes: The network model is a YOLOv8 network model, wherein the loss function of the network model is a CIOU loss function.

4. The method for detecting the head positioning of the cold rolling coiler according to claim 2, characterized in that: The step of obtaining the historical strip head positioning image of the strip steel comprises: Based on the speed information of the cold rolling production line, a smooth transition algorithm is used to dynamically adjust the acquisition frame rate of the camera to obtain the historical strip head positioning image of the strip.

5. The method for detecting the head positioning of the cold rolling coiler according to claim 1, characterized in that: The step of identifying the head positioning image based on the detection model to obtain the identification result includes: Based on the output result of the detection model, determine the head bounding box information and the positioning area bounding box information, wherein the head bounding box information includes the head bounding box center point coordinates and the head bounding box height, the head bounding box center point coordinates include the head bounding box center point ordinate, the positioning area bounding box information includes the positioning area bounding box center point coordinates and the positioning area bounding box height, and the positioning area bounding box center point coordinates include the positioning area bounding box center point ordinate; Based on the ordinate of the center point of the tape head boundary box and the height of the tape head boundary box, the ordinate of the upper surface of the tape head is calculated, wherein the ordinate of the upper surface of the tape head is the longitudinal coordinate between the first contact of the tape head with the winding area and the winding machine core shaft; Calculate the ordinate of the upper surface of the positioning area and the ordinate of the lower surface of the positioning area based on the ordinate of the center point of the positioning area boundary box and the height of the positioning area boundary box; Based on the ordinate of the upper surface of the tape head, the ordinate of the upper surface of the positioning area and the ordinate of the lower surface of the positioning area, an identification result is obtained.

6. The method for detecting the head positioning of the cold rolling coiler according to claim 5, characterized in that: The obtaining of the recognition result based on the upper surface ordinate of the tape head, the upper surface ordinate of the positioning area and the lower surface ordinate of the positioning area comprises: When the longitudinal coordinate of the upper surface of the tape head is greater than or equal to the first preset value and less than or equal to the second preset value, the recognition result is normal; When the longitudinal coordinate of the upper surface of the belt head is less than the first preset value or greater than the second preset value, the recognition result is abnormal and an alarm mechanism is triggered; The first preset value is an interpolation value of the ordinate of the upper surface of the positioning area and the deviation value, and the second preset value is a sum of the ordinate of the lower surface of the positioning area and the deviation value.

7. The method for detecting the head positioning of the cold rolling coiler according to claim 6, characterized in that: Also includes: In the case where the recognition result is abnormal, the tape head positioning detection interface is controlled to display the tape head positioning image and alarm information, wherein the tape head positioning detection interface includes the start time, the end time, the cold rolling production line and the uncoiling machine number.

8. A detection device for the head positioning of a cold rolling coiler, characterized in that: include: An image acquisition unit, used for acquiring a strip head positioning image of the strip; A strip head recognition unit, used for recognizing the strip head positioning image based on a detection model to obtain a recognition result, wherein the recognition result is used to determine whether the strip head positioning of the strip is correct; The strip steel coiling unit is used to perform the coiling operation of the strip steel when the identification result indicates that the strip head of the strip steel is correctly positioned.

9. An electronic device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor is used to implement the steps of a method for detecting the head positioning of a cold rolling coiler as described in any one of claims 1 to 7 when executing the computer program stored in the memory.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, a method for detecting the head positioning of a cold rolling coiler according to any one of claims 1 to 7 is implemented.

Images