System and method for identifying coating quality

Abstract

A system for identifying coating quality includes photographic equipment, a process control system, and a computing system. The photographic equipment captures images of the product after coating on the production line. The process control system provides process data for the product on the coating line. The computing system analyzes coating defects based on the post-coating images and generates warning messages based on these defects and the process data. The method for identifying coating quality can be implemented through these components of the system.

Description

Technical Field

[0001] This invention relates to production thread control technology, and in particular to systems and methods for identifying coating quality. Prior Technology

[0002] When producing steel sheets used as core materials for motors, generators, transformers, and other electrical equipment, a coating is applied to the unrolled steel strip from the steel coil to prevent rust, provide insulation, and ensure a smooth surface, thereby maintaining the quality of the steel sheets segmented from the steel strip.

[0003] However, the coating condition on the steel strip may be defective due to the following conditions: 1. When the coating machine rollers suddenly deflect; 2. When changing coatings; or 3. After replacing the coating machine rollers. However, because these defects usually occur randomly, with the rapid progress of the production line, operators often cannot easily identify coating defects on-site. Furthermore, if operators are required to observe the coating condition on-site, the need to approach the top or side edge of the steel strip for extended periods increases the risk of being cut or entangled by the steel strip.

[0004] Therefore, there is an urgent need in the art for a system and method for identifying coating quality in order to solve the above problems. Summary of the Invention

[0005] This disclosure provides a system for identifying coating quality, comprising a photographic device, a process control device, and a computing device. The photographic device is used to capture images of a product after it has been coated with a coating on a coating production line. The process control device is used to provide process data of the product on the coating production line. The computing device is coupled to the photographic device and the process computer, and is used to analyze coating defects of the product based on the post-coating images, and to generate warning messages based on coating defects and process data.

[0006] This disclosure also provides a method for identifying coating quality, comprising: a photographic device capturing a post-coating image of a product after being coated with a coating on a coating production line; a process control device providing process data of the product on the coating production line; a computing device analyzing coating defects of the product based on the post-coating image; and a computing device generating a warning message based on the coating defects and process data. Simple Explanation of the Diagram

[0007] Figure 1 is a schematic diagram of the component configuration of a system for identifying coating quality according to at least one embodiment of the present invention. Figure 2 is a flowchart of the steps of a method for identifying coating quality according to at least one embodiment of the present invention. Figure 3 is a schematic diagram of the coated image of at least one embodiment of the present invention. Figure 4 is a schematic diagram of a coated image marked "coating condition is good" according to at least one embodiment of the present invention. Figure 5 is a schematic diagram of a coated image marked "abnormal coating condition" according to at least one embodiment of the present invention. Figure 6 is a schematic diagram illustrating the operation of accessing defect data according to at least one embodiment of the present invention. Figure 7 is a schematic diagram illustrating the operation of accessing defect data according to at least one embodiment of the present invention. Implementation

[0008] The following embodiments are provided to illustrate the present invention. Those skilled in the art will readily understand the advantages and effects of the present invention upon reading its contents. However, the embodiments of the present invention are not intended to limit the scope of the invention. The present invention can be implemented or applied through other feasible methods, and every detail contained herein can be changed or modified according to different aspects and applications without departing from the essence of the invention.

[0009] The proportions, structures, and dimensions shown in the accompanying illustrations are solely for illustrative purposes and to aid those skilled in the art in understanding the invention, and are not intended to limit the scope of the invention. Therefore, any changes in proportions, modifications to structure, or adjustments to dimensions without affecting the purpose and effects of the invention should fall within the scope of the technical content disclosed herein.

[0010] In this document, when describing an object as "comprising," "including," or "having" technical features, other elements, structures, fields, components, devices, apparatuses, systems, steps, connections, modules, units, etc., may be included unless otherwise stated, and other features shall not be excluded. Furthermore, unless otherwise specified, the singular forms "a" and "the" used herein also include the plural forms, and the terms "or" and "and / or" are used interchangeably.

[0011] Figure 1 is a schematic diagram of the component configuration of the system 1 for identifying coating quality according to the present invention. It includes a photographic device 10, an auxiliary lighting device 11, a programmable controller 20, a computing device 30, and a display device 40, for monitoring the coating status of products on the coating production line PL. In this embodiment, the coating production line PL can be a steel strip coating machine (e.g., a three-roll coating machine), and the product can be a steel strip unwound from a steel coil and about to be segmented into steel sheets. Furthermore, in addition to the signal transmission between the programmable controller 20 and the computing device 30 via a communication module 21 (e.g., a Modbus I / O module), the components of system 1 can be connected to each other via any wired or wireless means to transmit data (e.g., through the TCP / IP communication protocol).

[0012] The camera 10 can be a network-type surveillance camera and can be installed at the output end of the coating production line PL to capture images of the coated products after coating and output to the output end. In addition, auxiliary lighting equipment 11, which is also located on the coating production line PL, can provide auxiliary light sources to ensure the shooting quality of the camera 10.

[0013] The programmable control device 20 can be used to provide process data of the product on the coating production line PL based on the operating status of the coating production line PL. In this embodiment, the programmable control device 20 may consist of a Level-1 programmable control computer and a Level-2 programmable control computer, which are used to record different information on the coating production line PL, respectively. For example, the Level-1 programmable control computer can be used to record the on / off events and production speed of the coating production line PL during operation, while the Level-2 programmable control computer can be used to record information such as the product identification code (e.g., the identification code of the steel coil to which the steel strip belongs) and width (e.g., the width of the steel strip) of the product coated on the coating production line PL. Furthermore, the Level-1 programmable computer can also transmit an alarm signal (e.g., a "Coating_ON" signal) to the computing device 30 via the communication module 21 when it detects that the coating line PL has started coating operations, thereby instructing the system 1 to begin product coating quality identification; when it detects that the coating line PL has stopped coating operations, it can transmit an alarm signal (e.g., a "Coating_OFF" signal) to the computing device 30 via the communication module 21, thereby instructing the system 1 to stop product coating quality identification; or when the computing device 30 returns an alarm signal indicating a coating defect (e.g., a "Coating_Abnormal" signal), it can record the relevant information in a log file for future review. However, the programmable device 20 can also be implemented as a single computer device integrating the functions of the Level-1 programmable computer and the Level-2 programmable computer, and the present invention is not limited thereto.

[0014] The computing device 30 can be coupled to the photographic device 10 and the programmable device 20, and can be equipped with an object detection model. This model analyzes coating defects in the product based on the post-coating image from the photographic device 10. When a coating defect is identified, it generates a warning message by combining the coating defect, the post-coating image, and process data from the programmable device 20. The warning message can be used to immediately notify relevant operators to repair the defective parts of the product or to suspend the coating production line (PL) to inspect the faulty parts. It can also be stored as defect data for future use as evidence to clarify disputes when customers raise concerns about product quality.

[0015] Display device 40 can be coupled to computing device 30 and used to display the coated image from camera device 10 and / or warning messages from computing device 30 in real time for operators to review.

[0016] Figure 2 below illustrates the steps of the method for identifying coating quality using the components of System 1 of the present invention, and can be used in conjunction with Figures 3 to 7 to understand the process and effect of coating quality identification.

[0017] In step S201, the camera device 10 may start capturing and acquiring a coated image in response to a warning signal (e.g., a "Coating_ON" signal) received by the computing device 30 from the programmable control device 20 indicating that the coating line PL has started coating operations. The coated image may be as shown in Figure 3, which describes the output form of the product after the coating operation is performed by the coating line PL.

[0018] In step S202, the computing device 30 can obtain the product's process data from the programmable controller 20. The process data may include the product's identification code and width, as well as the production speed of the coating line PL.

[0019] In step S203, the computing device 30 can use an object detection model to identify coating defects in the post-coating image. At this time, the object detection model can be a You-only-look-once (YOLO) model (e.g., the YOLOv8 model), used to identify features such as color changes and textures of the coating on the product surface, thereby determining whether the coating condition of the product is normal or has coating defects. The training method for the YOLO model may include: labeling the dataset to allow the object detection model to identify possible coating defect categories on the product (including but not limited to: edge coating defects, central coating defects, normal coating, etc.); loading a pre-trained model into the computing device 30; training the pre-trained model 500 times using a dataset containing positive samples (e.g., post-coating images without coating defects) and negative samples (e.g., post-coating images with coating defects); using data augmentation (e.g., sharpness adjustment, flipping, Mosaic processing, Mixup processing, etc.) to increase the generalization ability and learning efficiency of the neural network model; and performing model validation to select the best model for implementation.

[0020] In step S204, the computing device 30 can generate a warning message from the identified coating defects and process data and store it as defect data. At this time, the state of the coating defects identified from the post-coating image can be shown in Figures 4 and 5. In Figure 4, since the coating condition of the product surface is good, the object detection model does not detect the coating defect. The computing device 30 can mark the post-coating image with a warning message of "good coating condition" (such as a green box indicating good coating condition at the edge of the steel strip) and store it as a defect data entry after associating it with the process data. On the other hand, in Figure 5, since the object detection model has detected an abnormal coating condition on the product surface, the computing device 30 can mark the post-coating image with a warning message of "abnormal coating condition" (such as a red box indicating abnormal coating condition at both the edge and center of the steel strip) and store it as a defect data entry after associating it with the process data. In Figures 4 and 5 above, process data may include product identification codes (e.g., the identification code of the steel coil to which the steel strip currently belongs) and the current steel strip length position calculated by integrating time with the production speed (also known as defect location), which can be used to indicate the location of the product and the coating defects on it.

[0021] In step S205, the computing device 30 can transmit the completed coating image and warning message (e.g., the content of Figure 4 or Figure 5) to the display device 40 for real-time display, so that the operator can check the coating status of the product in real time.

[0022] In step S206, the computing device 30 can also retrieve specific defect data in response to a retrieval command input by the operator to the system 1. For example, when the operator inputs a specified product identification code and the location of the defect to be viewed, the computing device 30 can select the corresponding entry in the defect data and transmit it to the display device 40 for display. The retrieval of defect data can be shown in Figures 6 and 7. The defect data can be recorded in a coordinate graph format, showing the distribution of coating defects on the product. This allows the operator to locate the desired coating defect by using the width or length of the product and simultaneously review the content of the corresponding coated image. This function can provide evidence for clarifying disputes when customers raise concerns about product quality after shipment, thus identifying the attribution of responsibility for the dispute.

[0023] In summary, the system and method for identifying coating quality disclosed in this invention can identify coating defects in products on the coating production line in real time through automation, enabling operators to efficiently and promptly grasp coating quality and adjust the operating parameters of the coating production line (e.g., the roller coating parameters of the coating machine). Furthermore, the system and method of this invention can also store defect data, allowing operators to trace the distribution of coating defects throughout the product and ensure accountability, thereby improving quality control capabilities. The above description is only a preferred embodiment of the present invention. All equivalent changes and modifications made in accordance with the claims of the present invention shall be covered by the present invention.

[0024] 1: System

[0025] 10: Photography equipment

[0026] 11: Auxiliary lighting equipment

[0027] 20: Programmable control equipment

[0028] 21: Communication Module

[0029] 30: Computing devices

[0030] 40: Display device

[0031] PL: Coating production line

[0032] S201~S206: Steps

Claims

1. A system for identifying coating quality, comprising: A photographic device is used to capture images of a product after it has been coated with paint on a coating production line; a process control device is used to provide process data of the product on the coating production line, including the production speed of the product on the coating production line; and a computing device is coupled to the photographic device and the process computer, used to analyze coating defects of the product based on the post-coating image, and to generate warning messages based on the coating defects and the process data. The computing device is further used to: associate the post-coating image with the process data, the coating defects and the warning messages and store them as defect data; and in response to a retrieval command, select a record corresponding to the retrieval command from the defect data and transmit it to the display device for display. When the computing device identifies a coating defect in the post-coating image, it generates the defect location of the coating defect on the product based on the production speed relative to time integration and stores it in the defect data.

2. The system as described in claim 1, further comprising: A display device, coupled to the computing device, is used to display the coated image and the warning message in real time.

3. The system as described in claim 1, wherein, The process data also includes the product identification code of the product; the retrieval instruction includes the specified product identification code and the defect location; and the computing device is further used to: select the entry in the defect data that corresponds to the retrieval instruction based on the product identification code and the defect location and send it to the display device for display.

4. A method for identifying coating quality, comprising: A photographic device captures images of a product after it has been coated with paint on a coating production line; a programmable control system provides process data for the product on the coating production line, including the production speed of the product on the coating production line; a computing device analyzes coating defects of the product based on the post-coating images; and the computing device generates warning messages based on the coating defects and the process data. The computing device associates and stores the coated image with the process data, the coating defect, and the warning message as defect data, including: when the computing device identifies the coating defect in the coated image, it generates the defect location of the coating defect on the product based on the production speed relative to time integration and stores it in the defect data; and in response to a retrieval command, the computing device selects a line corresponding to the retrieval command from the defect data and transmits it to the display device for display.

5. The method as described in claim 4, further comprising: The display device instantly displays the coated image and the warning message.

6. The method as described in claim 4, wherein, The process data also includes the product identification code of the product; the retrieval instruction includes the specified product identification code and the defect location; and the computing device responds to the retrieval instruction by selecting the entry corresponding to the retrieval instruction from the defect data and transmitting it to the display device for display, including: the computing device selects the entry corresponding to the retrieval instruction from the defect data according to the product identification code and the defect location and sends it to the display device for display.

Images