Coil size detection method based on visual system

Abstract

The invention provides a coil size detection method based on a visual system, and the method comprises the steps: setting a detection range and a judgment condition, and carrying out the graphic preprocessing which comprises the steps: increasing the contrast ratio, carrying out the binarization processing, and carrying out the real-time shade correction; the camera and the mechanical arm are matched to measure actual data of the raw material fusiform coil, and whether the raw material fusiform coil meets the requirement or not is judged according to the detection range and the judgment condition; if the requirement is met, the manipulator grabs the raw material fusiform coil according to actual data, and expanding operation is carried out through a coil expanding machine, so that the raw material fusiform coil is expanded into an expanded coil; and the camera and the manipulator are matched to measure actual data of the expansion coil, whether the expansion coil meets the requirement or not is judged according to the detection range and the judgment condition, and if the expansion coil meets the requirement, the manipulator carries out the next grabbing operation. According to the scheme, the visual system and manipulator operation can be effectively combined, and through comparison of multiple data, the labor cost is reduced, and the production efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of coil detection, and in particular to a coil size detection method based on a vision system. Background Art

[0002] In modern manufacturing, precision and quality are one of the important indicators of product competitiveness. Especially in the electronics and electrical industries, coils are important basic components, and their dimensional accuracy directly affects the performance and reliability of the final product. Traditional coil size detection methods generally rely on manual measurement, which is not only time-consuming and labor-intensive, but also difficult to ensure the consistency and accuracy of measurement. Manual measurement is limited by the experience and skills of the operator, which can easily cause measurement errors, thereby affecting product quality. In addition, manual operation is also prone to low production efficiency and cannot meet the needs of modern production for high efficiency and high quality. With the expansion of production scale and the improvement of product consistency requirements, traditional detection methods have been difficult to meet market demand. In recent years, with the continuous advancement of digital image processing technology, the use of digital image processing technology to measure and analyze the size and geometric features of workpieces can achieve efficient, accurate and real-time product quality monitoring. At the same time, compared with traditional contact measurement methods, the dimension detection technology based on the visual system does not require direct contact with the product, avoiding possible damage or contamination to the product. At the same time, the visual system can be applied to products of various shapes and sizes, with high adaptability and flexibility. It can optimize the production rhythm and improve production efficiency in the entire production process, and can provide data support for quality improvement and production optimization. Therefore, it is necessary to provide a coil size detection method based on a visual system.

[0003] However, one of the shortcomings of the related prior art is that the visual system is not effectively combined with the robot operation, resulting in low efficiency in actual operation. Summary of the invention

[0004] The present invention aims to solve at least one of the above technical problems.

[0005] To solve the above problems, a first object of the present invention is to provide a coil size detection method based on a vision system.

[0006] To achieve the first objective of the present invention, an embodiment of the present invention provides a coil size detection method based on a visual system, which includes: step S100: setting a detection range, a judgment condition, and performing graphic preprocessing, wherein the graphic preprocessing includes increasing contrast, binarization processing, and real-time shading correction. The contrast is increased by a logarithmic transformation method. The curve of the logarithmic function can transform the grayscale concentrated in the low-brightness area into a low-brightness area that is equally divided into a high-brightness area. The change function is The binarization process converts each pixel in the image into black and white to enhance the contrast and clarity of the image. The algorithm used is the image threshold segmentation algorithm. The core of the binarization process is to select an optimal threshold, compare the pixel points with the threshold, and divide the image according to the result, dividing it into target and background areas. The key point lies in the selection of the threshold. Threshold segmentation is to compare all the pixels on the image, select a threshold T on the pixel point, judge the image information according to the threshold T, and divide the image into regions. The image threshold is selected using the Ostu algorithm. The target and background of the image are segmented according to the grayscale characteristics of the image. If the variance between classes is larger, the difference between the segmented backgrounds is larger and the probability of misclassification is smaller. The threshold is T, the variance between classes is g, and the total average grayscale value μ of the image is obtained: , the formula for the inter-class variance is: , ; Step S200: The camera and the manipulator cooperate to measure the actual data of the raw material shuttle coil, and determine whether the raw material shuttle coil meets the requirements based on the detection range and judgment conditions; Step S300: If it meets the requirements, the manipulator grabs the raw material shuttle coil according to the actual data, and performs expansion operation through the coil expansion machine, so that the raw material shuttle coil is expanded into an expansion coil; Step S400: The camera and the manipulator cooperate to measure the actual data of the expansion coil, and determine whether the expansion coil meets the requirements based on the detection range and judgment conditions. If the expansion coil meets the requirements, the manipulator performs the next step of grabbing operation.

[0007] Through high-performance cameras and advanced image processing algorithms, the coil size can be accurately measured, the error caused by manual measurement can be avoided, and the quality control level of the product can be improved. This method has a high degree of automation and can quickly complete the entire process of measurement, judgment and grasping, significantly reducing the time of manual intervention and greatly improving production efficiency. The system can monitor the qualified status of the coil through the set detection range and judgment conditions, and timely feedback unqualified products, reducing the losses caused by unqualified products flowing into subsequent processes. Automated detection reduces the dependence on manual operation, reduces labor costs, and reduces errors caused by human factors. Different types of coils can be detected as needed, making this method more adaptable and flexible. Through strict detection standards, it is ensured that each coil produced meets the corresponding quality standards and improves the consistency of the overall product. Increasing the contrast can make the features of the coil more obvious, enhance the clarity of the image, and help improve the accuracy of subsequent analysis. Through binarization processing, the image is converted to contain only two colors, which can effectively highlight the edges and structures of the target object, making feature extraction more concise and efficient. Real-time shading correction can dynamically adjust the brightness and contrast of the image according to different lighting conditions and environmental changes to ensure that clear images can be obtained under various conditions. Graphics preprocessing can effectively eliminate background noise and unnecessary interference, improve the accuracy of target detection, and reduce misjudgment. By optimizing image quality and structural features, the complexity of subsequent image processing and analysis is reduced, and the overall detection efficiency is improved. High-quality image delivery makes machine learning and algorithm application more robust, promoting further automated detection and classification. Clear and easy-to-analyze images reduce the need for manual operation, further improving the automation of the production line, and saving time and cost.

[0008] In the above technical solution, after step S200, the method further includes: if the shuttle coil of the raw material does not meet the requirements, the robot arm returns to the initial position.

[0009] The manipulator returns to its initial position to avoid handling or grabbing unqualified coils, thus reducing material waste. By quickly returning to the initial position, the manipulator can quickly prepare for the next inspection and maintain the smoothness of the production line. It avoids the manipulator from performing unnecessary actions after discovering unqualified products, thereby reducing equipment fatigue and extending its service life. Returning to the initial position ensures that the manipulator is in a safe state during subsequent operations and reduces the risk of abnormal situations. If the machine automatically returns to the initial position, the operator can then promptly check the unqualified products and take necessary corrective measures to maintain a good state of production. This mechanism allows for flexible response to unqualified products during the inspection process, thereby enhancing the ability of the production line to adapt to unexpected situations.

[0010] In any of the above technical solutions, after step S400, it also includes: step S410: if the expansion coil does not meet the requirements, the camera will send an error signal to the robot and PLC, the robot will stop the grasping action and return to the initial position.

[0011] By sending an error signal, the system can quickly identify and handle defective products, avoid further operations, and ensure product quality. The robot stops immediately after receiving the stop signal, reducing potential safety hazards caused by incorrect operations and protecting equipment and operators. Stopping the grip and returning to the initial position in time can effectively avoid further processing of defective products and reduce additional losses caused by subsequent processing. The linkage between the camera and PLC forms a real-time feedback mechanism that can quickly respond to problems on the production line, improve production efficiency and response speed. Automatic return to the initial position simplifies the manual intervention process, and operators can identify and handle problem coils more quickly, improving overall production efficiency. The system's automated feedback ensures the continuity and stability of the production line. Even if defective products appear, adjustments can be made quickly to avoid long production interruptions.

[0012] In any of the above technical solutions, after step S410, the following further includes: step S420: checking error information through PLC and adjusting the expansion parameters of the coil expansion machine.

[0013] By checking the error information and making corresponding adjustments through PLC, problems that arise during the production process can be quickly responded to, improving the adaptability and production efficiency of the machinery. Adjustment of the expansion parameters can improve the quality of the coil and ensure that the technical requirements are better met in subsequent production, thereby optimizing the overall production process. Timely adjustment of parameters can shorten the troubleshooting time, reduce the downtime of the production line, and improve the overall production capacity. Through real-time monitoring and adjustment, product quality can be better controlled, the specifications and performance of each coil can be ensured to be consistent, and market competitiveness can be enhanced. Integrating error information and parameter adjustments, and realizing information-based and systematic management through the PLC management system, will help standardize equipment maintenance and production processes. Each adjustment can be recorded in the PLC system, which will help analyze historical data and provide data support for future parameter optimization.

[0014] In any of the above technical solutions, the detection range includes the coil length range, the cross-section width range, the straight edge span range, the coil lead end span range, the non-lead end span range and the cross-section height range.

[0015] Covering multiple important geometric parameters ensures comprehensive inspection of all aspects of the coil, thereby improving product quality. By refining the inspection range, the production process can be controlled more accurately to ensure that each coil meets strict specifications. Comprehensive inspection of multiple parameters helps to detect potential defects early, reduce the production of unqualified products, and reduce rework and scrap rates. Ensure that all coils produced are consistent in all dimensional parameters to improve product consistency and interchangeability. By clarifying the inspection scope, unnecessary inspections and operations can be reduced, improving overall production efficiency, while saving manpower and time. Detailed inspection range data can provide a basis for subsequent quality tracking and analysis, helping to identify the causes of problems and make improvements.

[0016] In any of the above technical solutions, the determination condition is that the actual data is within the detection range.

[0017] By setting clear judgment conditions, the standards for qualified and unqualified can be clearly defined, and the effectiveness and consistency of detection can be improved. Standardized judgment conditions reduce the impact of human subjective judgment and ensure the objectivity and reliability of the test results. Judging within the set range can quickly determine whether the product is qualified, thereby improving the speed and efficiency of the overall detection. Timely identification of unqualified coils can reduce the complexity of subsequent processing and reduce rework and scrap rates caused by unqualified products. By setting judgment conditions, the control logic of machinery and systems can be simplified, which helps to achieve a higher level of automation. Through strict judgment conditions, it is ensured that all products shipped are within the specified range, improving the consistency and reliability of products.

[0018] In any of the above technical solutions, in step S200, the actual data of the raw shuttle coil includes the coil length C of the raw shuttle coil and the cross-sectional width A of the raw shuttle coil.

[0019] Coil length and cross-sectional width are key dimensions that affect coil performance. Accurately measuring these two parameters can effectively evaluate product quality. Ensuring that actual data matches design specifications helps ensure technical and functional compliance of the product and avoid performance issues caused by improper design. Strict testing of key dimensions can improve product consistency and reduce inconvenience and premature failures caused by dimensional differences. Timely detection and control of these core dimensions can significantly reduce the defect rate in the production process, thereby reducing rework and scrap. Through regular data analysis of coil length and cross-sectional width, the production process can be optimized to improve production efficiency.

[0020] In any of the above technical solutions, in step S400, the actual data of the expansion coil includes the straight side span D2 of the expansion coil, the coil lead end span D2 of the expansion coil, the non-lead end span D3 of the expansion coil, and the cross-sectional height B of the expansion coil.

[0021] The straight edge span, lead end span, non-lead end span and cross-sectional height are key parameters that affect coil performance and assembly. Detailed measurements can be used to comprehensively evaluate product quality. Monitoring these key dimensions can ensure the consistency of each expansion coil and reduce unstable performance or assembly difficulties caused by dimensional differences. Timely detection and control of these key dimensions can significantly reduce the defect rate in production, reduce the occurrence of unqualified products and the corresponding rework costs. By analyzing key data, potential problems in the production process can be identified and a basis can be provided for process improvement, thereby improving overall production efficiency. The collected actual data provides data support for subsequent quality management, process optimization and design improvement, forming a data-driven decision-making mechanism. If quality problems occur, these specific actual data can help quickly locate the root cause of the problem, facilitating quality traceability and optimization.

[0022] In any of the above technical solutions, after step S400, the method further includes: if the expansion coil meets the requirements, storing the actual data of the expansion coil and the actual data of the raw material shuttle coil.

[0023] Storing actual data can provide strong support for subsequent quality tracking and help quickly locate and analyze problems when they occur. By systematically recording the actual data of qualified products, it can provide an important basis for quality management and improvement, and promote the standardization and normalization of the process. The saved data provides a basis for subsequent historical comparative analysis, which can identify performance trends, optimize manufacturing processes, and improve the scientific nature of decision-making. Through data accumulation, companies can determine where improvements need to be made in the production process, reduce defect rates, and improve overall benefits. Through data storage and analysis, combined with machine learning and intelligent algorithms, self-optimizing intelligent manufacturing can be gradually realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0025] Figure 1 A flow chart of a coil size detection method based on a visual system provided by an embodiment of the present invention; Figure 2 A schematic diagram of a raw material shuttle coil of a coil size detection method based on a visual system provided by an embodiment of the present invention; Figure 3 A schematic diagram of an expanding coil of a coil size detection method based on a visual system provided in an embodiment of the present invention.

[0026] Description of reference numerals: A-section width of raw shuttle coil, B-section height of expansion coil, C-coil length of raw shuttle coil, D1-straight side span of expansion coil, D2-span of coil lead end of expansion coil, D3-span of non-lead end of expansion coil. DETAILED DESCRIPTION

[0027] In order to make the above-mentioned purpose, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. It should be noted that, in the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other. In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention can also be implemented in other ways different from those described herein, and therefore, the protection scope of the present invention is not limited by the specific embodiments disclosed below.

[0028] like Figure 1 , Figure 2 as well as Figure 3 As shown, an embodiment of the present invention provides a coil size detection method based on a visual system, which includes: step S100: setting a detection range, a judgment condition, and performing graphic preprocessing, wherein the graphic preprocessing includes increasing contrast, binarization processing, and real-time shading correction. The contrast is increased by a logarithmic transformation method. The curve of the logarithmic function can transform the grayscale concentrated in the low-brightness area into a low-brightness area that is equally divided into a high-brightness area. The change function is The binarization process converts each pixel in the image into black and white to enhance the contrast and clarity of the image. The algorithm used is the image threshold segmentation algorithm. The core of the binarization process is to select an optimal threshold, compare the pixel points with the threshold, and divide the image according to the result, dividing it into target and background areas. The key point lies in the selection of the threshold. Threshold segmentation is to compare all the pixels on the image, select a threshold T on the pixel point, judge the image information according to the threshold T, and divide the image into regions. The image threshold is selected using the Ostu algorithm. The target and background of the image are segmented according to the grayscale characteristics of the image. If the variance between classes is larger, the difference between the segmented backgrounds is larger and the probability of misclassification is smaller. The threshold is T, the variance between classes is g, and the total average grayscale value μ of the image is obtained: , the formula for the inter-class variance is: , ; Step S200: The camera and the manipulator cooperate to measure the actual data of the raw material shuttle coil, and judge whether the raw material shuttle coil meets the requirements according to the detection range and judgment conditions; Step S300: If it meets the requirements, the manipulator grabs the raw material shuttle coil according to the actual data, and performs expansion operation through the coil expansion machine to expand the raw material shuttle coil into an expansion coil. If it does not meet the requirements, the manipulator returns to the initial position; Step S400: The camera and the manipulator cooperate to measure the actual data of the expansion coil, And according to the detection range and judgment conditions, it is judged whether the expansion coil meets the requirements. If the expansion coil meets the requirements, the robot will perform the next grasping operation and store the actual data of the expansion coil and the actual data of the raw shuttle coil. If the expansion coil does not meet the requirements, the camera will send an error signal to the robot and PLC, the robot will stop the grasping action and return to the initial position, the operator will view the error information through the PLC, and adjust the expansion parameters of the coil expansion machine; wherein, the detection range includes the coil length range, the cross-sectional width range, the straight line side span range, the coil lead end span range, the non-lead end span range and the cross-sectional height range, and the judgment condition is that the actual data is within the detection range, the actual data of the raw shuttle coil includes the coil length C of the raw shuttle coil and the cross-sectional width A of the raw shuttle coil, and the actual data of the expansion coil includes the straight line side span D1 of the expansion coil, the coil lead end span D2 of the expansion coil, the non-lead end span D3 of the expansion coil, and the cross-sectional height B of the expansion coil.

[0029] In this embodiment, before the coil size detection is performed, the detection range needs to be set according to the various sizes of the finished coil, mainly including the coil length range, the cross-section width range, the straight edge span range, the coil lead end span range, the non-lead end span range and the cross-section height range. And the image is pre-processed through the online detection setting, mainly including increasing contrast, binarization processing, real-time shading correction, etc. This operation can enhance the clarity of the image and reduce noise.

[0030] Increase image contrast: The logarithmic transformation method is used. The logarithmic function curve y=lnx. When x is small, the curve changes greatly, and when x is large, the curve changes little. The grayscale concentrated in the low-brightness area can be transformed into a low-brightness area that is equally divided into a high-brightness area. The change function is Where r is the grayscale value of the input image, s is the grayscale value of the output image, c is usually 1, and the grayscale value range is [0, 255]. After the transformation, the range of s may not be between [0, 255]. It is necessary to use mathematical methods to convert the value of s to between [0, 255]. Mathematical methods often use the minimum and maximum values ​​to scale the s value.

[0031] Image binarization: It converts each pixel in the image into black and white (binary) in order to enhance the contrast and clarity of the image and facilitate more accurate identification of edges and features in the image. The most commonly used algorithm is the image threshold segmentation algorithm. The core of binarization is to select an optimal threshold, compare the pixel points with the threshold, and divide the image into target and background areas based on the results. The key point lies in the selection of the threshold. Threshold segmentation actually compares all the pixels on the image, selects a threshold T on the pixel point, judges the image information based on the threshold T, and divides the image into regions. Among them, represents the original image pixel; represents the image after binarization algorithm processing; T represents the threshold value. The image threshold selection uses the Ostu algorithm. The image is segmented from the target and background according to the grayscale characteristics of the image. If the inter-class variance is larger, the difference in the segmented background is larger. Therefore, the larger the inter-class variance is, the smaller the probability of misclassification is. Take the optimal threshold value as T, the inter-class variance value as g, and get the total average grayscale value of the image: μ: , Indicates the pixel ratio; represents the average grayscale; Indicates the background ratio; Represents the average gray value. The formula for the inter-class variance is: , The larger g is, the larger the number of pixels between the target and the background is, and the lower the probability of segmentation error is. The T value selected when g is maximum is the optimal segmentation threshold.

[0032] Real-time shading correction: A pre-processing function that can effectively eliminate the interference of uneven background brightness. When the detection is affected by lighting or shadows and the defects cannot be detected, correction can be performed. According to the size of the part to be extracted set on the software, the extraction size is confirmed by the square area of ​​the image screen, and the brightness of the part to be extracted is selected. If only the part brighter than the background is extracted, select light, otherwise, select dark.

[0033] In addition, it is necessary to set the judgment condition. In this embodiment, the judgment condition is set to be qualified if the actual data is within the range of the detection range, and unqualified if it is not within the range. In this embodiment, the camera is connected to the manipulator, and the position and angle of the camera can be adjusted by moving the manipulator. The coil length C and the cross-sectional width A of the raw shuttle coil are measured by the camera and the manipulator. The measurement of the coil length C is by moving the camera to the side view direction of the raw shuttle coil to take a photo, and then identifying the leftmost and rightmost endpoints of the raw shuttle coil in the photo, and calculating the horizontal distance difference between the two endpoints, which is the coil length C. The two leads of the raw shuttle coil are in a vertical state, and will not affect the identification of the coil length C of the raw shuttle coil at the side view angle. The measurement of the cross-sectional width A is by identifying the height difference between the upper end and the lower end of the lower layer of the raw shuttle coil. After the identification is completed, the coil length C and cross-sectional width A are compared with the coil length range and interface width range in the detection range. If they do not meet the range, the robot returns to the initial position and does not grab it. If they meet the range, the size of the raw shuttle coil meets the standard. The robot grabs it according to the actual data of the raw shuttle and the offset of the raw shuttle coil position, and places it on the coil expansion machine for expansion operation. When the coil is expanded and becomes an expanding coil, the camera and the manipulator cooperate to move to the side of the expanding coil to measure the straight side span D1 of the expanding coil, the coil lead end span D2 of the expanding coil, the non-lead end span D3 of the expanding coil, and the cross-sectional height B of the expanding coil. The cross-sectional height B is the height difference between the upper end and the lower end of the lower layer of the expanding coil, the straight side span D1 is the height difference between the lower end of the upper layer and the upper end of the lower layer in the middle part of the expanding coil, the coil lead end span D2 is the height difference between the lower end of the upper layer close to the lead part and the upper end of the lower layer of the expanding coil, and the non-lead end span D3 is the height difference between the lower end of the upper layer away from the lead part and the upper end of the lower layer of the expanding coil. After the measurement is completed, the actual data of each expansion coil is compared with the straight edge span range, coil lead end span range, non-lead end span range and cross-section height range in the detection range. If it does not meet the requirements, the camera will send an error signal to the manipulator and PLC, the manipulator will stop grabbing and return to the initial position, the staff will check the error information through the PLC and adjust the expansion parameters of the coil expansion machine. If it meets the requirements, the manipulator will proceed to the next grabbing operation, and the staff can scan the QR code on the coil to store and bind the actual data of the expansion coil and the actual data of the raw shuttle coil for subsequent quality traceability.

[0034] In the present invention, the terms "install", "connect", "connect", "fix" and the like should be understood in a broad sense. For example, "connect" can be a fixed connection, a detachable connection, or an integral connection; "connect" can be a direct connection or an indirect connection through an intermediate medium. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0035] In the description of the present invention, it is necessary to understand that the directions or positional relationships indicated by the terms "upper" and "lower" etc. are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or unit referred to must have a specific direction, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0036] In the description of this specification, the description of the terms "one embodiment", "some embodiments", "specific embodiments", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0037] Although the present invention is disclosed as above, the present invention is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the scope defined by the claims.

Claims

1. A coil size detection method based on a visual system, characterized in that: include: Step S100: Setting the detection range, determination conditions, and performing graphic preprocessing, wherein the graphic preprocessing includes increasing contrast, binarization, and real-time shading correction. The contrast is increased by logarithmic transformation. The logarithmic function curve can transform the grayscale concentrated in the low brightness area into a low brightness area that is equally divided into high brightness areas. The change function is: , The binarization process converts each pixel in the image into black and white to enhance the contrast and clarity of the image. The algorithm used is the image threshold segmentation algorithm. The core of the binarization process is to select an optimal threshold, compare the pixel points with the threshold, and divide the image according to the result, dividing it into target and background areas. The key point lies in the selection of the threshold. Threshold segmentation is to compare all the pixel points on the image, select a threshold T on the pixel point, judge the image information according to the threshold T, and divide the image into regions. The image threshold is selected using the Ostu algorithm. The target and background of the image are segmented according to the grayscale characteristics of the image. If the variance between classes is larger, the difference between the segmented backgrounds is larger and the probability of misclassification is smaller. The threshold is T, the variance between classes is g, and the total average grayscale value μ of the image is obtained: , the formula for the inter-class variance is: , ; Step S200: the camera and the manipulator cooperate to measure the actual data of the raw material shuttle coil, and determine whether the raw material shuttle coil meets the requirements according to the detection range and the determination conditions; Step S300: If the requirements are met, the robot grabs the raw shuttle coil according to the actual data, and performs expansion operation through the coil expansion machine, so that the raw shuttle coil is expanded into an expanded coil; Step S400: The camera and the manipulator cooperate to measure the actual data of the expansion coil, and determine whether the expansion coil meets the requirements according to the detection range and the judgment conditions. If the expansion coil meets the requirements, the manipulator performs the next grasping operation.

2. The coil size detection method based on a visual system according to claim 1, characterized in that: After step S200, the method further includes: If the raw material shuttle coil does not meet the requirements, the robot returns to the initial position.

3. The coil size detection method based on a visual system according to claim 1, characterized in that: After step S400, the method further includes: Step S410: If the expansion coil does not meet the requirements, the camera will send an error signal to the robot and PLC, and the robot will stop the grasping action and return to the initial position.

4. The coil size detection method based on a visual system according to claim 3 is characterized in that: After step S410, the following steps are further included: Step S420: Check the error information through the PLC and adjust the expansion parameters of the coil expansion machine.

5. The coil size detection method based on a visual system according to claim 1, characterized in that: The detection range includes a coil length range, a cross-section width range, a straight edge span range, a coil lead end span range, a non-lead end span range, and a cross-section height range.

6. The coil size detection method based on a visual system according to claim 1, characterized in that: The determination condition is that the actual data is within the range of the detection range.

7. The coil size detection method based on a visual system according to claim 1, characterized in that: In the step S200, the actual data of the raw shuttle coil includes the coil length C of the raw shuttle coil and the cross-sectional width A of the raw shuttle coil.

8. The coil size detection method based on a visual system according to claim 1, characterized in that: In the step S400, the actual data of the expansion coil includes the straight side span D1 of the expansion coil, the coil lead end span D2 of the expansion coil, the non-lead end span D3 of the expansion coil, and the cross-sectional height B of the expansion coil.

9. The coil size detection method based on a visual system according to claim 1, characterized in that: After step S400, the method further includes: If the expansion coil meets the requirements, the actual data of the expansion coil and the actual data of the raw material shuttle coil are stored.

Images