A method and system for acquiring die-cutting parameters of a composite film

Abstract

The application relates to the technical field of data processing, in particular to a die-cutting parameter acquisition method and system for a composite film, which comprises the following steps: acquiring a standard gray image and a continuous frame real-time gray image of the composite film; for any frame real-time gray image, the background probability and the foreground probability of any pixel point in the real-time gray image are calculated in combination with the standard gray image; the edge probability of any pixel point is calculated based on the foreground probability and the background probability; the edge probability is subjected to density clustering based on the edge probability of all pixel points in the real-time gray image, and a real-time edge pixel point set for the real-time gray image is acquired; a standard edge pixel point set of the standard gray image is acquired, and a die-cutting deviation parameter of the composite film is calculated based on the real-time edge pixel point set and the standard edge pixel point set. The die-cutting deviation parameter of the composite film can be dynamically calculated to identify a cutting path and improve the yield.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method and system for obtaining die-cutting parameters of a composite film. Background Technology

[0002] Composite film materials are widely used in packaging, construction, medical, electronics, and new energy fields due to their excellent physicochemical properties. These materials typically require high-precision die-cutting processes, and the quality of the cut edges directly determines the product yield. However, two major technical bottlenecks exist in actual die-cutting production:

[0003] 1. Cutting deviation caused by fixture displacement: Composite films are thin and smooth, making them prone to micro-displacement during high-speed die cutting due to insufficient clamping force. This displacement causes the cutting head to deviate from the preset path, resulting in defects such as uneven edge cutting, burrs, or delamination, significantly increasing the scrap rate and reducing production efficiency.

[0004] Current solutions mostly involve offline sampling inspection, which uses manual sampling tools to check the dimensions of cut parts. This has a serious lag and cannot correct production line deviations in real time.

[0005] 2. Edge recognition failure caused by transparency: Composite films are usually highly transparent or semi-transparent, making it difficult to form clear edge contrast under industrial vision systems. Traditional grayscale image processing algorithms (such as Canny and Sobel) rely on obvious gradient changes, while the grayscale gradient features of transparent edges are weak and easily affected by ambient light, resulting in a significant deviation between the real-time edge position and the actual physical edge, which cannot provide reliable feedback for cutting accuracy.

[0006] Current solutions mostly use fixed threshold segmentation, which is based on static grayscale threshold segmentation of edges. This has poor adaptability to transparent materials and cannot distinguish between real edges and noise (such as scratches and bubbles on the film surface).

[0007] It can be found that the core defect of the above-mentioned composite film cutting technology bottleneck lies in the inability to dynamically calculate the die-cutting deviation parameters of the composite film, resulting in a lack of robust identification capability for the edge of the film material during dynamic production, and an inability to accurately identify whether the cutting path of the composite film deviates from the preset path during the cutting process. Summary of the Invention

[0008] In order to dynamically calculate the die-cutting deviation parameters during the cutting process of composite films, so as to identify the cutting path of composite films in real time and improve the yield of composite film cutting, this application provides a method and system for obtaining die-cutting parameters of composite films.

[0009] In a first aspect, this application provides a method for obtaining die-cutting parameters of a composite film, employing the following technical solution:

[0010] A method for obtaining die-cutting parameters of a composite film, comprising:

[0011] Acquire standard grayscale images and continuous frame real-time grayscale images of the composite film during the cutting process;

[0012] For any frame of the real-time grayscale image, the background probability and foreground probability of any pixel in the real-time grayscale image are calculated in combination with the standard grayscale image.

[0013] Based on the foreground probability of any pixel and the foreground and background probabilities of its neighboring pixels, the edge probability of any pixel is calculated.

[0014] Based on the edge probabilities of all pixels in the real-time grayscale image, density clustering is performed on the edge probabilities to obtain a set of real-time edge pixels for the real-time grayscale image.

[0015] Obtain the standard edge pixel set of the standard grayscale image, and calculate the die-cutting deviation parameter of the composite film based on the real-time edge pixel set and the standard edge pixel set.

[0016] Optionally, calculating the background probability and foreground probability of any pixel in the real-time grayscale image includes:

[0017] Obtain the standard composite film pixel values ​​and standard background pixel values ​​from the standard grayscale image;

[0018] Obtain the position coordinates of the edge pixels of the previous frame of the grayscale image of the real-time grayscale image;

[0019] Based on the position coordinates and pixel value of any pixel in the real-time grayscale image, the pixel value of the standard composite film, the pixel value of the standard background, and the position coordinates of the edge pixel in the previous frame, the background probability and foreground probability of any pixel are calculated.

[0020] Optionally, calculating the edge probability of any pixel based on the foreground probability of any pixel and the foreground and background probabilities of its neighboring pixels includes:

[0021] Obtain the first probability difference between any pixel and its neighboring pixels with respect to the foreground probability;

[0022] Obtain the second probability difference between the foreground probability and the background probability of the neighboring pixels of any given pixel;

[0023] The edge probability of any pixel is calculated based on the number of neighboring pixels of any pixel, the first probability difference, and the second probability difference.

[0024] Optionally, performing density clustering on the edge probabilities to obtain a real-time edge pixel set for the real-time grayscale image includes:

[0025] Density clustering is performed on the edge probabilities to obtain clusters of pixels with high edge probabilities;

[0026] All pixels in the cluster are obtained as the real-time edge pixel set for the real-time grayscale image.

[0027] Optionally, the edge probabilities can be density-clustered using the DBSCAN clustering algorithm.

[0028] Optionally, the step of calculating the die-cutting deviation parameter of the composite film based on the real-time edge pixel set and the standard edge pixel set includes:

[0029] Calculate the minimum Euclidean distance between any pixel in the real-time edge pixel set and all pixels in the standard edge pixel set to obtain the deviation set;

[0030] The maximum value is obtained from the set of deviations and used as the die-cutting deviation parameter of the composite film.

[0031] Optionally, after calculating the die-cutting deviation parameter of the composite film, the method further includes:

[0032] Obtain a preset die-cutting deviation threshold, and obtain a comparison result between the die-cutting deviation parameter and the die-cutting deviation threshold;

[0033] If the die-cutting deviation parameter is greater than the die-cutting deviation threshold, it is determined that the composite film cutting has deviated, and an alarm message is generated.

[0034] Secondly, this application provides a system for obtaining die-cutting parameters of a composite film, which adopts the following technical solution:

[0035] A system for obtaining die-cutting parameters of a composite film includes a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the method for obtaining die-cutting parameters of the composite film described above is implemented.

[0036] The beneficial effect is that the die-cutting parameters of the above-mentioned composite film are obtained to generate a computer program and stored in the memory so that it can be loaded and executed by the processor. Thus, the system can be made according to the memory and the processor, which is convenient to use.

[0037] This application has the following technical advantages:

[0038] 1. By acquiring continuous frames of real-time grayscale images and combining them with standard grayscale images, the background probability and foreground probability of pixels in any frame of real-time grayscale image are calculated, dynamically adapting to the subtle grayscale gradient characteristics of the transparent composite film.

[0039] 2. The DBSCAN clustering algorithm is used to cluster the edge probabilities of all pixels in the real-time grayscale image. This can extract a set of spatially continuous real edge pixels from the edge probability map and overcome the edge fragmentation problem caused by fixture displacement.

[0040] 3. The real-time edge pixel set and the standard edge pixel set are spatially compared using the Euclidean distance algorithm to output the die-cutting deviation parameter, which facilitates real-time judgment of whether there is a deviation in the cutting path of the composite film.

[0041] 4. When a cutting deviation is detected in the die-cutting machine, an alarm message can be generated to prompt relevant personnel to make corrections, or the die-cutting machine can automatically adjust according to the die-cutting deviation parameters, switching from offline sampling inspection mode to online correction mode, thereby improving the yield of composite films. Attached Figure Description

[0042] The above and other objects, features, and advantages of exemplary embodiments of this application will become readily understood by reading the following detailed description with reference to the accompanying drawings. Several embodiments of this application are illustrated in the drawings by way of example and not limitation, and the same or corresponding reference numerals denote the same or corresponding parts.

[0043] Figure 1 This is a schematic diagram of the overall process of a method for obtaining die-cutting parameters of a composite film according to an embodiment of this application.

[0044] Figure 2 This is a flowchart illustrating steps S201-S203 of a method for obtaining die-cutting parameters of a composite film according to an embodiment of this application.

[0045] Figure 3 This is a flowchart illustrating steps S301-S303 in a method for obtaining die-cutting parameters of a composite film according to an embodiment of this application.

[0046] Figure 4 This is a flowchart illustrating steps S401-S402 of a method for obtaining die-cutting parameters of a composite film according to an embodiment of this application.

[0047] Figure 5 This is a flowchart illustrating steps S501-S502 in a method for obtaining die-cutting parameters of a composite film according to an embodiment of this application.

[0048] Figure 6 This is a flowchart illustrating steps S601-S602 of a method for obtaining die-cutting parameters of a composite film according to an embodiment of this application. Detailed Implementation

[0049] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0050] It should be understood that when the terms "first," "second," etc., are used in the claims, description, and drawings of this application, they are only used to distinguish different objects and not to describe a specific order. The terms "comprising" and "including" used in the description and claims of this application indicate the presence of the described features, integrals, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or collections thereof.

[0051] This application discloses a method for obtaining die-cutting parameters of a composite film, referring to... Figure 1 ,include:

[0052] S101. Obtain standard grayscale images and real-time grayscale images of continuous frames of the composite film during the cutting process.

[0053] In this embodiment, the standard grayscale image is an image in which the composite film and other objects can be clearly distinguished. The foreground of this image is the composite film area, and the background is the area of ​​other objects in the image, such as conveyor belts, cutting equipment, etc. Furthermore, the foreground and background areas of the composite film are marked in the standard grayscale image.

[0054] During the actual cutting process of the composite film, continuous frames of real-time grayscale images of the composite film are acquired, for example, 30 or 60 frames of grayscale images are acquired per second. For each frame of real-time grayscale image, the die-cutting deviation parameter between the real-time grayscale image and the standard grayscale image needs to be calculated to determine whether any deviation has occurred during the cutting process of the composite film.

[0055] S102. For any frame of real-time grayscale image, combine it with the standard grayscale image to calculate the background probability and foreground probability of any pixel in the real-time grayscale image.

[0056] Taking a single frame of real-time grayscale image as an example, calculating the die-cutting deviation parameter between the real-time grayscale image and the standard grayscale image requires obtaining the edge pixels in the real-time grayscale image for comparison with those in the standard grayscale image. Obtaining the edge pixels in the real-time grayscale image requires distinguishing the foreground and background of all pixels in the real-time grayscale image using the composite film, thereby identifying the edge pixels.

[0057] In this embodiment, calculating the background probability and foreground probability of any pixel in a real-time grayscale image can roughly classify any pixel in the real-time grayscale image into a foreground region or a background region.

[0058] This embodiment calculates the background and foreground probabilities of pixels in any frame of real-time grayscale image by combining the acquired continuous frames of real-time grayscale images with standard grayscale images, and dynamically adapts to the subtle grayscale gradient characteristics of the transparent composite film.

[0059] Reference Figure 2 Calculate the background and foreground probabilities of any pixel in a real-time grayscale image, including:

[0060] S201. Obtain the standard composite film pixel value and the standard background pixel value in the standard grayscale image;

[0061] S202. Obtain the position coordinates of the edge pixels of the previous frame of the real-time grayscale image;

[0062] S203. Based on the position coordinates and pixel value of any pixel in the real-time grayscale image, the standard composite film pixel value, the standard background pixel value, and the position coordinates of the edge pixel in the previous frame, calculate the background probability and foreground probability of any pixel.

[0063] Based on the distinguishing markers for the foreground and background of the composite film in the standard grayscale image, the average pixel value of the pixels in the foreground region is selected as the standard composite film pixel value in the standard grayscale image, and the average pixel value of the pixels in the background region is selected as the standard background pixel value.

[0064] Based on the previous frame grayscale image of the real-time grayscale image, determine the position coordinates of the edge pixels of the previous frame grayscale image.

[0065] In this embodiment, since the acquired real-time grayscale image is continuous, the position of the composite film is fixed at the initial moment. For the first frame of the real-time grayscale image, the initial cutting point in the first frame of the real-time grayscale image can be directly marked as the edge pixel point of the first frame of the real-time grayscale image, that is, the previous frame edge pixel point of the second frame of the real-time grayscale image, and the position coordinates of the previous frame edge pixel point can be obtained.

[0066] Based on the standard composite film pixel value, standard background pixel value, and previous frame edge pixel position coordinates obtained from the above steps, combined with the position coordinates and pixel value of any pixel in the real-time grayscale image, the background probability and foreground probability of any pixel in the real-time grayscale image are calculated.

[0067] Specifically, in this embodiment, the method for calculating the foreground probability and background probability of any pixel in a real-time grayscale image is as follows:

[0068]

[0069] In the formula, This represents the background probability of any pixel in a real-time grayscale image. This represents the foreground probability of any pixel in a real-time grayscale image. This represents the pixel value of any pixel in a real-time grayscale image. Standard background pixel values, For standard composite film pixel values, These are the coordinates of the edge pixels in the previous grayscale image. The coordinates of any pixel in a real-time grayscale image. For Euclidean distance, For An exponential function with base 0. This represents the minimum value function.

[0070] The above calculations can determine the probability that any pixel in a real-time grayscale image is close to the background or foreground. However, during the calculation process, pixels with similar probabilities may appear, making it impossible to effectively distinguish whether a pixel belongs to the foreground or background. It is the nearest pixel from a pixel in a real-time grayscale image to the edge pixel of the previous frame in the previous grayscale image. Since it is a continuously acquired real-time image, the distance can be used to separate pixels with similar local probabilities.

[0071] This allows us to obtain the background and foreground probabilities of each pixel in the real-time grayscale image. It's important to note that the background and foreground probabilities of each pixel are calculated frame-by-frame, i.e., in real-time.

[0072] S103. Based on the foreground probability of any pixel and the foreground and background probabilities of its neighboring pixels, calculate the edge probability of any pixel.

[0073] After obtaining the foreground and background probabilities of all pixels in a real-time grayscale image, the edge probability of a pixel can be calculated based on the foreground probability of one pixel and the foreground and background probabilities of its neighboring pixels, in order to determine whether the pixel is an edge pixel in the real-time grayscale image.

[0074] Specifically, refer to Figure 3 In this embodiment, the edge probability of any pixel is calculated based on the foreground probability of any pixel and the foreground and background probabilities of its neighboring pixels, including:

[0075] S301. Obtain the first probability difference between any pixel and its neighboring pixels with respect to the foreground probability.

[0076] S302. Obtain the second probability difference between the foreground probability and background probability of adjacent pixels of any pixel;

[0077] S303. Calculate the edge probability of any pixel based on the number of adjacent pixels of any pixel, the first probability difference, and the second probability difference.

[0078] In this embodiment, any pixel is selected as the target pixel, and the method for calculating the edge probability of the target pixel is as follows:

[0079]

[0080] in, The edge probability of the target pixel. The foreground probability of the target pixel. The first in the neighborhood of the target pixel Foreground probability of each pixel The first in the neighborhood of the target pixel The background probability of a pixel is 8, where 8 represents the number of all neighboring pixels in the neighborhood of the target pixel in this embodiment, i.e., the range of the neighborhood is an 8-neighborhood. The number of neighboring pixels in the neighborhood of the target pixel can be actively adjusted according to the computing power of the device; in principle, the more the better, and the shorter the calculation time, the better.

[0081] in, The target pixel and the target pixel's neighborhood. The first probability difference between adjacent pixels relative to the foreground probability indicates a greater degree of differentiation in the foreground and a higher probability that the target pixel is an edge point.

[0082] in, The first in the neighborhood of the target pixel The larger the second probability difference between the foreground probability and background probability of the nth adjacent pixels, the higher the probability difference. The more distinct the distinction between foreground and background among adjacent pixels, the higher the reliability of the foreground and background probabilities, and vice versa. This difference can be corrected. Because the target pixel's neighborhood contains the first The less distinct the foreground and background of adjacent pixels, the better the foreground and background differentiation of the target pixel within its neighborhood. The lower the realism of the foreground probability provided by adjacent pixels, the lower the probability, and vice versa.

[0083] Thus, this embodiment is able to obtain the edge probability of any pixel in a real-time grayscale image.

[0084] S104. Based on the edge probabilities of all pixels in the real-time grayscale image, perform density clustering on the edge probabilities to obtain the real-time edge pixel set for the real-time grayscale image.

[0085] After obtaining the edge probabilities of all pixels in the real-time grayscale image, density clustering is performed on the edge probabilities of all pixels to obtain the real-time edge pixel set of the real-time grayscale image.

[0086] Reference Figure 4 Density clustering is performed on the edge probabilities to obtain a real-time set of edge pixels for a real-time grayscale image, including:

[0087] S401. Perform density clustering on the edge probabilities to obtain clusters of pixels with high edge probabilities;

[0088] S402. Obtain all pixels in the cluster as a set of real-time edge pixels for the real-time grayscale image.

[0089] Specifically, for the edge probabilities of all pixels in a real-time grayscale image, density clustering is performed on all edge probabilities. In this embodiment, the DBSCAN clustering algorithm is used. First, the position coordinates of all pixels are treated as data points. A similarity matrix (defining a distance metric) is constructed based on the edge probability values. Then, core points (points in the neighborhood that satisfy the MinPts number) are found. Finally, connected regions are merged to form clusters. The specific steps of clustering using the DBSCAN clustering algorithm are existing technologies and will not be described in detail here.

[0090] This process yields clusters of pixels with high edge probability, and all pixels within these clusters are used as the real-time edge pixel set for the real-time grayscale image. The real-time edge pixel set represents both the boundary between the foreground and background of the composite film in the real-time grayscale image and the cutting path of the composite film.

[0091] This embodiment uses the DBSCAN clustering algorithm to cluster the edge probabilities of all pixels in the real-time grayscale image, which can extract a set of spatially continuous real edge pixels from the edge probability map, overcoming the edge fragmentation problem caused by the displacement of the die-cutting machine fixture.

[0092] S105. Obtain the standard edge pixel set of the standard grayscale image, and calculate the die-cutting deviation parameter of the composite film based on the real-time edge pixel set and the standard edge pixel set.

[0093] After obtaining the real-time edge pixel set, it is necessary to compare and analyze it with the standard edge pixel set of the standard grayscale image to determine whether there is a cutting deviation in the composite film. First, the standard edge pixel set is obtained from the standard grayscale image, and then calculations are performed.

[0094] In this embodiment, refer to Figure 5 Based on the real-time edge pixel set and the standard edge pixel set, the die-cutting deviation parameters of the composite film are calculated, including:

[0095] S501. Calculate the minimum Euclidean distance between any pixel in the real-time edge pixel set and all pixels in the standard edge pixel set to obtain the deviation set;

[0096] S502. Obtain the maximum value from the deviation set and use it as the die-cutting deviation parameter of the composite film.

[0097] This embodiment calculates the Euclidean distance between any pixel in the real-time edge pixel set and all pixels in the standard edge pixel set, and obtains the minimum value as the effective Euclidean distance value for that pixel. This process continues until all effective Euclidean distance values ​​for all pixels are obtained, forming a deviation set for the real-time edge pixels. This deviation set represents the deviation between the real-time edge pixels and the standard edge pixels. Therefore, the maximum value from this deviation set is used as the die-cutting deviation parameter, which can be used to determine whether there is a serious deviation in the cutting of the composite film.

[0098] In this embodiment, the real-time edge pixel set and the standard edge pixel set are spatially compared using the Euclidean distance algorithm to output the die-cutting deviation parameter, which facilitates real-time judgment of whether the cutting path of the composite film has deviated.

[0099] Furthermore, referring to Figure 6 After calculating the die-cutting deviation parameters of the composite film, the following steps are also included:

[0100] S601. Obtain the preset die-cutting deviation threshold, and obtain the comparison result between the die-cutting deviation parameter and the die-cutting deviation threshold;

[0101] S602. If the die-cutting deviation parameter is greater than the die-cutting deviation threshold, it is determined that the composite film cutting has deviated, and an alarm message is generated.

[0102] After obtaining the die-cutting deviation parameters, this embodiment can also compare the die-cutting deviation parameter threshold with a preset die-cutting deviation threshold to determine whether there is a deviation in the cutting of the composite film. For example, if the die-cutting deviation threshold is set to 1.8, and the die-cutting deviation parameter is greater than 1.8, it can be determined that the cutting path of the composite film deviates from the standard grayscale image. At this time, an alarm message is generated to prompt relevant personnel to correct the deviation and adjust the cutting accuracy of the die-cutting machine.

[0103] This embodiment can also feed back the die-cutting deviation parameters to the die-cutting machine PLC system in real time, drive the servo motor to dynamically compensate the cutter head position, realize automatic deviation correction, replace manual sampling inspection, and form a closed-loop control chain of "acquisition-calculation-feedback-execution".

[0104] This application also discloses a system for obtaining die-cutting parameters of a composite film, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, a method for obtaining die-cutting parameters of a composite film according to this application is implemented.

[0105] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and will not be described in detail here.

[0106] In this application, the aforementioned memory can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium can be any suitable magnetic or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc., or any other medium that can be used to store required information and can be accessed by an application, module, or both. Any such computer storage medium can be part of a device or accessible to or connected to a device.

[0107] While this specification has shown and described numerous embodiments of the present application, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will arise for those skilled in the art without departing from the spirit and intent of the present application. It should be understood that various alternatives to the embodiments of the present application described herein may be employed in the practice of this application.

[0108] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A method of acquiring a die-cutting parameter of a composite film, characterized by, The method comprises the following steps: acquiring a standard gray image and a continuous frame real-time gray image of the composite film in the cutting process; for any frame of the real-time gray image, combining the standard gray image to calculate the background probability and the foreground probability of any pixel point in the real-time gray image; acquiring a first probability difference of any pixel point and its adjacent pixel points for the foreground probability; acquiring a second probability difference of the foreground probability and the background probability of the adjacent pixel points of any pixel point; based on the number of pixel points of the adjacent pixel points of any pixel point, the first probability difference and the second probability difference, calculating the edge probability of any pixel point; based on the edge probability of all pixel points in the real-time gray image, performing density clustering on the edge probability to acquire a real-time edge pixel point set for the real-time gray image; acquiring a standard edge pixel point set of the standard gray image, and based on the real-time edge pixel point set and the standard edge pixel point set, calculating a die-cutting deviation parameter of the composite film.

2. The method of claim 1, wherein, The calculation of the background probability and the foreground probability of any pixel point in the real-time gray image comprises the following steps: acquiring a standard composite film pixel value and a standard background pixel value in the standard gray image; acquiring a front frame edge pixel point position coordinate of a front frame gray image of the real-time gray image; based on the position coordinate and the pixel value of any pixel point in the real-time gray image, the standard composite film pixel value, the standard background pixel value and the front frame edge pixel point position coordinate, calculating the background probability and the foreground probability of any pixel point.

3. The method of claim 1, wherein, The density clustering of the edge probability to acquire the real-time edge pixel point set for the real-time gray image comprises the following steps: performing density clustering on the edge probability to acquire a clustering cluster of pixel points with high edge probability; acquiring all pixel points in the clustering cluster as the real-time edge pixel point set for the real-time gray image.

4. The method of claim 3, wherein, The density clustering of the edge probability adopts a DBSCAN clustering algorithm.

5. The method of claim 1, wherein, The calculation of the die-cutting deviation parameter of the composite film based on the real-time edge pixel point set and the standard edge pixel point set comprises the following steps: calculating the minimum value of the Euclidean distance between any pixel point in the real-time edge pixel point set and all pixel points in the standard edge pixel point set to acquire a deviation set; acquiring the maximum value from the deviation set as the die-cutting deviation parameter of the composite film.

6. The method of claim 1-5, wherein, After the calculation of the die-cutting deviation parameter of the composite film, the following steps are further included: acquiring a preset die-cutting deviation threshold value, and acquiring a comparison result of the die-cutting deviation parameter and the die-cutting deviation threshold value; if the die-cutting deviation parameter is greater than the die-cutting deviation threshold value, it is determined that the composite film cutting deviates, and an alarm information is generated.

7. A die-cutting parameter acquisition system for a composite film, characterized by, The device comprises a processor and a memory, and the memory stores computer program instructions which, when executed by the processor, implement the die-cutting parameter acquisition method of the composite film according to any one of claims 1-6. ​

Images