Metal sheet laminating process monitoring method based on data analysis

Abstract

The invention relates to the field of metal sheet laminating process monitoring, in particular to a metal sheet laminating process monitoring method based on data analysis, which comprises the following steps of: establishing multi-level characteristics of a metal sheet laminating process by using a real-time image of the metal sheet laminating process; carrying out verification processing on the basis of data analysis by utilizing the multi-stage characteristics of the metal sheet laminating process to obtain a multi-stage characteristic verification result of the metal sheet laminating process; the monitoring result of the metal sheet laminating process is obtained by using the multi-stage feature verification result of the metal sheet laminating process, and feature extraction is performed on real-time image data through data analysis, which is a key step for improving the monitoring accuracy, and useful information is extracted from the image so as to facilitate subsequent comparison and analysis tasks. The features can express the deep meaning of the image, meanwhile, a multi-stage comparison mechanism is established, error correction iteration is carried out through self-circulation in the scheme, and stable execution and implementation are ensured.

Description

Technical Field

[0001] The present invention relates to the field of metal sheet bonding process monitoring, and particularly to a method for monitoring metal sheet bonding process based on data analysis. Background Art

[0002] In traditional metal sheet processing, real-time monitoring of the bonding yield is carried out manually. However, a large number of defective bonded products still cannot be selected solely by manual inspection. At the same time, algorithms relying on precise visual analysis are also not fully applicable to metal sheet bonding processing to a certain extent. Therefore, there is an urgent need for a practical monitoring method for the metal sheet bonding process. Summary of the Invention

[0003] Aiming at the deficiencies of the prior art, the present invention provides a method for monitoring metal sheet bonding process based on data analysis, which extracts features from real-time image data through data analysis, provides a good data basis for subsequent comparison, and improves the monitoring accuracy.

[0004] To achieve the above object, the present invention provides a method for monitoring metal sheet bonding process based on data analysis, including: S1. Establish multi-level features of the metal sheet bonding process by using real-time images of the metal sheet bonding process; S2. Perform verification processing based on data analysis by using the multi-level features of the metal sheet bonding process to obtain a verification result of the multi-level features of the metal sheet bonding process; S3. Obtain a monitoring result of the metal sheet bonding process by using the verification result of the multi-level features of the metal sheet bonding process.

[0005] Preferably, establishing multi-level features of the metal sheet bonding process by using real-time images of the metal sheet bonding process includes: S1-1. Collect real-time images of the metal sheet bonding process; S1-2. Establish image features of the metal sheet bonding process by using the real-time images of the metal sheet bonding process; S1-3. Establish target features of the metal sheet bonding process by using the real-time images of the metal sheet bonding process; S1-4. Use the image features and target features of the metal sheet bonding process as multi-level features of the metal sheet bonding process.

[0006] Furthermore, establishing image features of the metal sheet bonding process by using real-time images of the metal sheet bonding process includes: S1-2-1. Obtain the image pixels corresponding to the real-time images of the metal sheet bonding process as the first image feature; S1-2-2. Obtain the image area corresponding to the metal sheet by using the real-time images of the metal sheet bonding process as the second image feature; S1-2-3. Obtain the remaining area corresponding to the real-time image of the metal sheet bonding process according to the second image feature as the third image feature; S1-2-4. Use the first image feature, the second image feature, and the third image feature as the image features of the metal sheet bonding process.

[0007] Furthermore, establishing the target features of the metal sheet bonding process by using the real-time image of the metal sheet bonding process includes: Establish a two-dimensional coordinate system of the real-time image with the image center corresponding to the real-time image of the metal sheet bonding process as the origin; Obtain the corresponding vertex coordinates of the metal sheet by using the two-dimensional coordinate system of the real-time image to establish a metal sheet vertex coordinate set; Obtain the metal sheet vertex vectors by using the metal sheet vertex coordinate set according to the two-dimensional coordinate system of the real-time image to establish a metal sheet vertex vector set; Use the metal sheet vertex coordinate set and the metal sheet vertex vector set as the target features of the metal sheet bonding process.

[0008] Furthermore, performing verification processing based on data analysis by using the multi-level features of the metal sheet bonding process to obtain the multi-level feature verification result of the metal sheet bonding process includes: S2-1. Obtain the associated data features of the metal sheet bonding process based on data analysis by using the multi-level features of the metal sheet bonding process; S2-2. Perform verification processing by using the associated data features of the metal sheet bonding process to obtain the multi-level feature verification result of the metal sheet bonding process.

[0009] Furthermore, obtaining the associated data features of the metal sheet bonding process based on data analysis by using the multi-level features of the metal sheet bonding process includes: S2-1-1. Determine whether the second image feature corresponding to the image feature of the metal sheet bonding process in the multi-level features of the metal sheet bonding process corresponds to the metal sheet vertex coordinate set corresponding to the target feature of the metal sheet bonding process in the multi-level features of the metal sheet bonding process. If so, execute S2-1-2; otherwise, return to S1-2-2; S2-1-2. Obtain the metal sheet contour by using the second image feature corresponding to the image feature of the metal sheet bonding process in the multi-level features of the metal sheet bonding process; S2-1-3. Determine whether the metal sheet contour corresponds to the third image feature corresponding to the image feature of the metal sheet bonding process in the multi-level features of the metal sheet bonding process. If so, execute S2-1-4; otherwise, return to S1-2-3; S2-1-4. Establish a metal sheet contour coordinate set by using the metal sheet contour according to the two-dimensional coordinate system of the real-time image; S2-1-5. Determine whether the metal sheet contour coordinate set contains the metal sheet vertex coordinate set corresponding to the target feature of the multi-level feature of the metal sheet bonding process. If so, use the metal sheet contour coordinate set as the associated data feature of the metal sheet bonding process. Otherwise, return to S2-1-1.

[0010] Further, the multi-level feature verification result of the metal sheet bonding process obtained by performing verification processing using the associated data feature of the metal sheet bonding process includes: S2-2-1. Establish a standard verification time t using the real-time image corresponding time of the metal sheet bonding process. S2-2-2. Obtain the associated data features at time t-1 and time t+1 respectively. S2-2-3. Determine whether the region corresponding to the associated data feature at the standard verification time t is consistent with the region corresponding to the associated data feature at time t-1. If so, execute S2-2-4. Otherwise, directly execute S2-2-5. S2-2-4. Determine whether the region corresponding to the associated data feature at the standard verification time t is consistent with the region corresponding to the associated data feature at time t+1. If so, the multi-level feature verification result of the metal sheet bonding process is normal. Otherwise, execute S2-2-5. S2-2-5. Determine whether the region corresponding to the associated data feature at time t-1 is consistent with the region corresponding to the associated data feature at time t+1. If so, execute S2-2-6. Otherwise, the multi-level feature verification result of the metal sheet bonding process is abnormal. S2-2-6. Determine whether the associated data feature at time t-1 is consistent with the associated data feature at time t+1. If so, the multi-level feature verification result of the metal sheet bonding process is normal. Otherwise, the multi-level feature verification result of the metal sheet bonding process is at risk, and output the associated data features at time t-1, the standard verification time t, and time t+1.

[0011] Further, the metal sheet bonding process monitoring result obtained using the multi-level feature verification result of the metal sheet bonding process includes: S3-1. Determine whether the multi-level feature verification result of the metal sheet bonding process is normal. If so, the metal sheet bonding process monitoring result is normal, update the current time, and return to S1-1. Otherwise, execute S3-2. S3-2. Determine whether the multi-level feature verification result of the metal sheet bonding process is at risk. If so, execute S3-3. Otherwise, directly execute S3-4. S3-3. Determine whether the vertex vector set of the metal sheet corresponding to the target feature of the metal sheet bonding process corresponds to the first image feature corresponding to the image feature of the metal sheet bonding process. If so, the monitoring result of the metal sheet bonding process is normal, update the current time, and return to S1-1. Otherwise, the monitoring result of the metal sheet bonding process is risky, and output the real-time image of the metal sheet bonding process. S3-4. Determine whether the vertex vector set of the metal sheet corresponding to the metal sheet bonding process at the standard verification time t is in correspondence with the vertex vector set of the metal sheet corresponding to the metal sheet bonding process at time t-1. If so, execute S3-5. Otherwise, the monitoring result of the metal sheet bonding process is abnormal. S3-5. Determine whether the vertex vector set of the metal sheet corresponding to the metal sheet bonding process at the standard verification time t is in correspondence with the vertex vector set of the metal sheet corresponding to the metal sheet bonding process at time t+1. If so, the monitoring result of the metal sheet bonding process is normal. Otherwise, the monitoring result of the metal sheet bonding process is abnormal.

[0012] Compared with the closest prior art, the beneficial effects of the present invention are as follows: Feature extraction of real-time image data through data analysis is a key step to improve the monitoring accuracy. Extract useful information from the image for subsequent comparison and analysis tasks. These features can express the deep meaning of the image, and at the same time establish a multi-level comparison mechanism and a self-loop within the scheme for error correction iteration to ensure stable implementation. Description of the Drawings

[0013] Figure 1 is a flowchart of a method for monitoring a metal sheet bonding process based on data analysis provided by the present invention. Detailed Embodiments

[0014] The following further details the specific embodiments of the present invention with reference to the accompanying drawings.

[0015] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0016] Embodiment 1: The present invention provides a method for monitoring a metal sheet bonding process based on data analysis, as Figure 1 shown, including: S1. Establish multi-level features of the metal sheet bonding process using the real-time image of the metal sheet bonding process. S2. Verify and process the multi-level features of the metal sheet bonding process based on data analysis to obtain the verification result of the multi-level features of the metal sheet bonding process; S3. Obtain the monitoring result of the metal sheet bonding process by using the verification result of the multi-level features of the metal sheet bonding process.

[0017] S1 specifically includes: S1-1. Collect the real-time image of the metal sheet bonding process; S1-2. Establish the image features of the metal sheet bonding process by using the real-time image of the metal sheet bonding process; S1-3. Establish the target features of the metal sheet bonding process by using the real-time image of the metal sheet bonding process; S1-4. Use the image features and target features of the metal sheet bonding process as the multi-level features of the metal sheet bonding process.

[0018] S1-2 specifically includes: S1-2-1. Obtain the image pixels corresponding to the real-time image of the metal sheet bonding process as the first image feature; S1-2-2. Obtain the image area corresponding to the metal sheet by using the real-time image of the metal sheet bonding process as the second image feature; S1-2-3. Obtain the remaining area corresponding to the real-time image of the metal sheet bonding process according to the second image feature as the third image feature; S1-2-4. Use the first image feature, the second image feature and the third image feature as the image features of the metal sheet bonding process.

[0019] S1-3 specifically includes: S1-3-1. Establish a two-dimensional coordinate system of the real-time image with the image center corresponding to the real-time image of the metal sheet bonding process as the origin; S1-3-2. Obtain the vertex coordinates corresponding to the metal sheet by using the two-dimensional coordinate system of the real-time image to establish a set of metal sheet vertex coordinates; S1-3-3. Obtain the metal sheet vertex vectors according to the two-dimensional coordinate system of the real-time image by using the set of metal sheet vertex coordinates to establish a set of metal sheet vertex vectors; S1-3-4. Use the set of metal sheet vertex coordinates and the set of metal sheet vertex vectors as the target features of the metal sheet bonding process.

[0020] S2 specifically includes: S2-1. Obtain the associated data features of the metal sheet bonding process by using the multi-level features of the metal sheet bonding process based on data analysis; S2-2. Use the associated data features of the metal sheet bonding process to perform verification processing to obtain the multi-level feature verification result of the metal sheet bonding process.

[0021] S2-1 specifically includes: S2-1-1. Determine whether the image features corresponding to the multi-level features of the metal sheet bonding process, i.e., the second image features, correspond to the metal sheet vertex coordinate sets corresponding to the target features of the metal sheet bonding process in the multi-level features of the metal sheet bonding process. If so, execute S2-1-2; otherwise, return to S1-2-2. S2-1-2. Use the second image features corresponding to the multi-level features of the metal sheet bonding process to obtain the metal sheet contour. S2-1-3. Determine whether the metal sheet contour corresponds to the third image features corresponding to the multi-level features of the metal sheet bonding process. If so, execute S2-1-4; otherwise, return to S1-2-3. S2-1-4. Establish a metal sheet contour coordinate set based on the real-time image two-dimensional coordinate system using the metal sheet contour. S2-1-5. Determine whether the metal sheet contour coordinate set contains the metal sheet vertex coordinate sets corresponding to the target features of the metal sheet bonding process in the multi-level features of the metal sheet bonding process. If so, use the metal sheet contour coordinate set as the associated data features of the metal sheet bonding process; otherwise, return to S2-1-1.

[0022] S2-2 specifically includes: S2-2-1. Use the moment corresponding to the real-time image of the metal sheet bonding process to establish the standard verification moment t. S2-2-2. Obtain the associated data features at the t-1 moment and the t+1 moment respectively. S2-2-3. Determine whether the associated data feature corresponding area at the standard verification moment t is consistent with the associated data feature corresponding area at the t-1 moment. If so, execute S2-2-4; otherwise, directly execute S2-2-5. S2-2-4. Determine whether the associated data feature corresponding area at the standard verification moment t is consistent with the associated data feature corresponding area at the t+1 moment. If so, the multi-level feature verification result of the metal sheet bonding process is normal; otherwise, execute S2-2-5. S2-2-5. Determine whether the associated data feature corresponding area at the t-1 moment is consistent with the associated data feature corresponding area at the t+1 moment. If so, execute S2-2-6; otherwise, the multi-level feature verification result of the metal sheet bonding process is abnormal. S2-2-6. Determine whether the associated data features at time t-1 are the same as those at time t+1. If so, the multi-level feature verification result of the metal sheet bonding process is normal; otherwise, the multi-level feature verification result of the metal sheet bonding process is at risk, and output the associated data features at time t-1, the standard verification time t, and time t+1.

[0023] S3 specifically includes: S3-1. Determine whether the multi-level feature verification result of the metal sheet bonding process is normal. If so, the monitoring result of the metal sheet bonding process is normal, update the current time, and return to S1-1; otherwise, execute S3-2. S3-2. Determine whether the multi-level feature verification result of the metal sheet bonding process is at risk. If so, execute S3-3; otherwise, directly execute S3-4. S3-3. Determine whether the metal sheet vertex vector set corresponding to the target feature of the metal sheet bonding process corresponds to the first image feature corresponding to the image feature of the metal sheet bonding process. If so, the monitoring result of the metal sheet bonding process is normal, update the current time, and return to S1-1; otherwise, the monitoring result of the metal sheet bonding process is at risk, and output the real-time image of the metal sheet bonding process. S3-4. Determine whether the metal sheet vertex vector sets corresponding to the metal sheet bonding process at the standard verification time t and the metal sheet bonding process at time t-1 are both correspondingly consistent. If so, execute S3-5; otherwise, the monitoring result of the metal sheet bonding process is abnormal. S3-5. Determine whether the metal sheet vertex vector sets corresponding to the metal sheet bonding process at the standard verification time t and the metal sheet bonding process at time t+1 are both correspondingly consistent. If so, the monitoring result of the metal sheet bonding process is normal; otherwise, the monitoring result of the metal sheet bonding process is abnormal.

[0024] Those skilled in the art should understand that the embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0025] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in one or more flows and / or blocks. Figure 1 in one or more flows and / or blocks Figure 1 in one or more blocks.

[0026] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in one or more flows and / or blocks. Figure 1 in one or more flows and / or blocks Figure 1 in one or more blocks.

[0027] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more flows and / or blocks. Figure 1 in one or more flows and / or blocks Figure 1 in one or more blocks.

[0028] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that modifications or equivalent replacements can still be made to the specific embodiments of the present invention. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention shall be covered by the protection scope of the claims of the present invention.

Claims

1. A method for monitoring the metal sheet bonding process based on data analysis, characterized in that, Including: S1. Establish multi-level features of the metal sheet bonding process using real-time images of the metal sheet bonding process; S2. Based on data analysis, use the multi-level features of the metal sheet bonding process to perform verification processing to obtain the verification result of the multi-level features of the metal sheet bonding process; S3. Use the verification result of the multi-level features of the metal sheet bonding process to obtain the monitoring result of the metal sheet bonding process.

2. The method for monitoring the metal sheet bonding process based on data analysis according to claim 1, characterized in that, The establishment of multi-level features of the metal sheet bonding process using real-time images of the metal sheet bonding process includes: S1-1. Collect real-time images of the metal sheet bonding process; S1-2. Use the real-time images of the metal sheet bonding process to establish image features of the metal sheet bonding process; S1-3. Use the real-time images of the metal sheet bonding process to establish target features of the metal sheet bonding process; S1-4. Use the image features and target features of the metal sheet bonding process as multi-level features of the metal sheet bonding process.

3. The method for monitoring the metal sheet bonding process based on data analysis according to claim 2, characterized in that, The establishment of image features of the metal sheet bonding process using real-time images of the metal sheet bonding process includes: S1-2-1. Obtain the image pixels corresponding to the real-time images of the metal sheet bonding process as the first image feature; S1-2-2. Use the real-time images of the metal sheet bonding process to obtain the corresponding image area of the metal sheet as the second image feature; S1-2-3. According to the second image feature, obtain the remaining area corresponding to the real-time images of the metal sheet bonding process as the third image feature; S1-2-4. Use the first image feature, the second image feature, and the third image feature as the image features of the metal sheet bonding process.

4. The method for monitoring the metal sheet bonding process based on data analysis according to claim 3, characterized in that, The establishment of target features of the metal sheet bonding process using real-time images of the metal sheet bonding process includes: Use the image center corresponding to the real-time images of the metal sheet bonding process as the origin to establish a two-dimensional coordinate system for the real-time images; Use the two-dimensional coordinate system of the real-time images to obtain the vertex coordinates of the metal sheet and establish a set of metal sheet vertex coordinates; Use the set of metal sheet vertex coordinates to obtain the metal sheet vertex vectors according to the two-dimensional coordinate system of the real-time images and establish a set of metal sheet vertex vectors; Use the set of metal sheet vertex coordinates and the set of metal sheet vertex vectors as the target features of the metal sheet bonding process.

5. The method for monitoring the metal sheet bonding process based on data analysis according to claim 2, characterized in that, Based on data analysis, use the multi-level features of the metal sheet bonding process to perform verification processing to obtain the verification result of the multi-level features of the metal sheet bonding process includes: S2-1. Based on data analysis, use the multi-level features of the metal sheet bonding process to obtain the associated data features of the metal sheet bonding process; S2-2. Use the associated data features of the metal sheet bonding process to perform verification processing to obtain the verification result of the multi-level features of the metal sheet bonding process.

6. The method for monitoring the metal sheet bonding process based on data analysis according to claim 5, characterized in that, Based on data analysis, use the multi-level features of the metal sheet bonding process to obtain the associated data features of the metal sheet bonding process includes: S2-1-1. Judge whether the second image feature corresponding to the image feature of the metal sheet bonding process in the multi-level features of the metal sheet bonding process corresponds to the set of metal sheet vertex coordinates corresponding to the target feature of the metal sheet bonding process in the multi-level features of the metal sheet bonding process. If so, execute S2-1-2; otherwise, return to S1-2-2; S2-1-2. Obtain the metal sheet contour based on the second image feature corresponding to the image feature of the multi-level feature of the metal sheet laminating process using the metal sheet laminating process; S2-1-3. Determine whether the metal sheet contour corresponds to the third image feature corresponding to the image feature of the multi-level feature of the metal sheet laminating process. If so, execute S2-1-4; otherwise, return to S1-2-3; S2-1-4. Establish a metal sheet contour coordinate set based on the metal sheet contour in the real-time image two-dimensional coordinate system; S2-1-5. Determine whether the metal sheet contour coordinate set contains the metal sheet vertex coordinate set corresponding to the target feature of the multi-level feature of the metal sheet laminating process. If so, use the metal sheet contour coordinate set as the associated data feature of the metal sheet laminating process; otherwise, return to S2-1-1.

7. The method for monitoring the metal sheet bonding process based on data analysis according to claim 6, characterized in that, The verification process using the associated data feature of the metal sheet laminating process to obtain the multi-level feature verification result of the metal sheet laminating process includes: S2-2-1. Establish a standard verification time t based on the corresponding time of the real-time image of the metal sheet laminating process; S2-2-2. Obtain the associated data features at t-1 and t+1 respectively; S2-2-3. Determine whether the region corresponding to the associated data feature at the standard verification time t is consistent with the region corresponding to the associated data feature at t-1. If so, execute S2-2-4; otherwise, directly execute S2-2-5; S2-2-4. Determine whether the region corresponding to the associated data feature at the standard verification time t is consistent with the region corresponding to the associated data feature at t+1. If so, the multi-level feature verification result of the metal sheet laminating process is normal; otherwise, execute S2-2-5; S2-2-5. Determine whether the region corresponding to the associated data feature at t-1 is consistent with the region corresponding to the associated data feature at t+1. If so, execute S2-2-6; otherwise, the multi-level feature verification result of the metal sheet laminating process is abnormal; S2-2-6. Determine whether the associated data feature at t-1 is consistent with the associated data feature at t+1. If so, the multi-level feature verification result of the metal sheet laminating process is normal; otherwise, the multi-level feature verification result of the metal sheet laminating process is at risk, and output the associated data features at t-1, the standard verification time t, and t+1.

8. The method for monitoring a metal sheet bonding process based on data analysis according to claim 7, wherein, The process of obtaining the metal sheet laminating process monitoring result using the multi-level feature verification result of the metal sheet laminating process includes: S3-1. Determine whether the multi-level feature verification result of the metal sheet laminating process is normal. If so, the metal sheet laminating process monitoring result is normal, update the current time, and return to S1-1; otherwise, execute S3-2; S3-2. Determine whether the multi-level feature verification result of the metal sheet laminating process is at risk. If so, execute S3-3; otherwise, directly execute S3-4; S3-3. Determine whether the vertex vector set of the metal sheet corresponding to the target feature of the metal sheet bonding process corresponds to the first image feature corresponding to the image feature of the metal sheet bonding process. If so, the monitoring result of the metal sheet bonding process is normal, update the current time, and return to S1-1. Otherwise, the monitoring result of the metal sheet bonding process is risky, and output the real-time image of the metal sheet bonding process; S3-4. Determine whether the vertex vector set of the metal sheet corresponding to the metal sheet bonding process at the standard verification time t is in correspondence with the vertex vector set of the metal sheet corresponding to the metal sheet bonding process at time t-1. If so, execute S3-5. Otherwise, the monitoring result of the metal sheet bonding process is abnormal; S3-5. Determine whether the vertex vector set of the metal sheet corresponding to the metal sheet bonding process at the standard verification time t is in correspondence with the vertex vector set of the metal sheet corresponding to the metal sheet bonding process at time t+1. If so, the monitoring result of the metal sheet bonding process is normal. Otherwise, the monitoring result of the metal sheet bonding process is abnormal.

Images