A system and method for online monitoring of aluminum-plastic film quality based on comparative analysis

Abstract

The present invention discloses an online quality monitoring system and method for aluminum-plastic film based on comparative analysis, belonging to the technical field of aluminum-plastic film detection. The system comprises a data acquisition module, a database, a defect recognition module, a three-dimensional model analysis module, an intelligent calculation and evaluation module, an interactive display module, and an intelligent monitoring module. The data acquisition module is used to collect image data and camera parameter data during the quality monitoring process of the aluminum-plastic film; the database is used to store image data of different aluminum-plastic film defect types; the defect recognition module is used to establish and train a CNN model; the three-dimensional model analysis module is used to establish a three-dimensional spatial model; the intelligent calculation and evaluation module is used to calculate the defect degree corresponding to different aluminum-plastic film defects and calculate the comprehensive quality value of the aluminum-plastic film; the interactive display module is used to provide a human-computer interaction platform; and the intelligent monitoring module is used to monitor the defect degree and quality evaluation value.

Description

Technical Field

[0001] The present invention relates to the technical field of aluminum-plastic film detection, and in particular to an aluminum-plastic film quality online monitoring system and method based on comparative analysis. Background Art

[0002] As a key material for lithium battery packaging, aluminum-plastic film's quality stability and production efficiency have become core elements of market competitiveness. Traditional methods for monitoring aluminum-plastic film quality often rely on manual spot checks and offline testing. This method is not only inefficient but also fails to fully cover every link in the production process, making it difficult to promptly detect and correct quality problems. Against this backdrop, online aluminum-plastic film quality monitoring technology has emerged. By integrating advanced image processing algorithms and intelligent analysis systems, this technology achieves real-time and continuous monitoring of the entire aluminum-plastic film production process. By comparing and analyzing with preset standards, it can accurately identify surface defects and promptly detect and warn of potential quality issues. In existing technologies, the degree of aluminum-plastic film defects is often determined based on the grayscale of the captured image. When online monitoring of aluminum-plastic film quality, the analysis results of some tiny scratches and voids that are difficult to determine are often unable to be verified, resulting in low system accuracy. In addition, the single image data makes it impossible for the system to effectively and comprehensively assess the severity of aluminum-plastic film defects, affecting production efficiency and the level of aluminum-plastic film quality control. Summary of the Invention

[0003] The object of the present invention is to provide an online monitoring system and method for aluminum-plastic film quality based on comparative analysis, so as to solve the problems raised in the above background technology.

[0004] In order to solve the above technical problems, the present invention provides the following technical solution: an online monitoring method for aluminum-plastic film quality based on comparative analysis, the method comprising the following steps:

[0005] S10, establishing a three-dimensional space model, using an inspection instrument to monitor the quality of the aluminum-plastic film, and collecting image data and camera parameter data during the aluminum-plastic film quality inspection process; the camera parameters include built-in parameters and external parameters of the CCD camera;

[0006] S20, establishing a CNN model based on the image data of aluminum-plastic film with defects stored in the database, and training the CNN model; analyzing and processing the collected image data, inputting the processed image data into the trained CNN model, and identifying the aluminum-plastic film defects and corresponding defect types in the image data;

[0007] S30, based on the recognized aluminum-plastic film defects in the image data and the camera parameter data, determining the position information of the feature points corresponding to each aluminum-plastic film defect in the three-dimensional space model based on the principle of triangulation, and obtaining the surface functions of the different aluminum-plastic film defects in the three-dimensional space model;

[0008] S40, calculating the defect degrees corresponding to the different aluminum-plastic film defects based on the surface functions of the different aluminum-plastic film defects in the three-dimensional space model; performing an online assessment of the quality of the aluminum-plastic film based on the defect types corresponding to the different aluminum-plastic film defects and the weights of the impact of the different defect types on the quality of the aluminum-plastic film, and calculating a comprehensive quality value of the aluminum-plastic film;

[0009] S50. Provide a human-computer interaction platform to digitally display the calculated defect degrees corresponding to different aluminum-plastic film defects and the quality assessment values of the aluminum-plastic film; wherein, determine the unqualified threshold values of the defect degrees and the unqualified threshold values of the comprehensive quality values of the aluminum-plastic film under different aluminum-plastic film defect types; when the defect degree and the comprehensive quality value reach the unqualified threshold values, determine that the quality of the aluminum-plastic film is unqualified, and send the unqualified information to the management personnel.

[0010] An online monitoring system for aluminum-plastic film quality based on comparative analysis, the system includes a data acquisition module, a database, a defect recognition module, a three-dimensional model analysis module, an intelligent calculation and evaluation module, an interactive display module and an intelligent monitoring module;

[0011] The data acquisition module is used to collect image data and camera parameter data during the quality monitoring process of the aluminum-plastic film; send the collected image data to the defect recognition module; and send the collected camera parameter data to the three-dimensional model analysis module;

[0012] The database is used to store image data of different aluminum-plastic film defect types;

[0013] The defect recognition module is used to establish a CNN model based on the image data of aluminum-plastic film with defects stored in the database and train the CNN model; analyze and process the collected image data, input the processed image data into the trained CNN model, identify the aluminum-plastic film defects in the image data and the corresponding defect types; send the identified aluminum-plastic film defects in the image data to the three-dimensional model analysis module; and send the identified aluminum-plastic film defects in the image data and the corresponding defect types to the intelligent calculation and evaluation module;

[0014] The three-dimensional model analysis module is used to establish a three-dimensional space model. Based on the aluminum-plastic film defects in the identified image data and the camera parameter data, the module determines the position information of the feature points corresponding to each aluminum-plastic film defect in the three-dimensional space model based on the principle of triangulation, and obtains the surface functions of different aluminum-plastic film defects in the three-dimensional space model; the surface functions of different aluminum-plastic film defects in the three-dimensional space model are sent to the intelligent calculation and evaluation module;

[0015] The intelligent calculation and evaluation module is used to calculate the defect degrees corresponding to different aluminum-plastic film defects based on the surface functions of different aluminum-plastic film defects in the three-dimensional space model; perform online evaluation of the quality of the aluminum-plastic film based on the defect types corresponding to the different aluminum-plastic film defects and the influence weights of the different defect types on the quality of the aluminum-plastic film, and calculate the comprehensive quality value of the aluminum-plastic film; and send the calculated defect degrees corresponding to the different aluminum-plastic film defects and the quality evaluation values of the aluminum-plastic film to the interactive display module;

[0016] The interactive display module is used to provide a human-computer interaction platform to digitally display the defect degrees corresponding to different calculated aluminum-plastic film defects and the quality assessment values of the aluminum-plastic film;

[0017] The intelligent monitoring module is used to determine the unqualified threshold values of the defect degree and the unqualified threshold values of the comprehensive quality value of the aluminum-plastic film under different aluminum-plastic film defect types; monitor the defect degree and quality assessment value of the aluminum-plastic film corresponding to different aluminum-plastic film defects displayed in the interactive display module; when the defect degree and the comprehensive quality value reach the unqualified threshold value, the aluminum-plastic film quality is determined to be unqualified, and the unqualified information is sent to the management personnel.

[0018] Compared with the existing technology, the beneficial effects achieved by the present invention are: by establishing a three-dimensional space model, based on the principle of triangulation, the surface functions of different aluminum-plastic film defects in the three-dimensional space model are determined respectively, and the defect degrees corresponding to different aluminum-plastic film defects and the comprehensive quality value of the aluminum-plastic film are calculated, thereby more accurately reflecting the current quality problems of the aluminum-plastic film, improving the accuracy of the system's online monitoring and the level of aluminum-plastic film quality control; by using two CCD cameras to respectively collect image data and perform image comparison analysis, the fault tolerance of the system is improved, thereby improving the production efficiency of the aluminum-plastic film. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 Schematic diagram of the steps of an online monitoring method for aluminum-plastic film quality based on comparative analysis of the present invention;

[0020] Figure 2 It is a structural schematic diagram of an aluminum-plastic film quality online monitoring system based on comparative analysis of the present invention. DETAILED DESCRIPTION

[0021] The following will provide a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0022] See also Figure 1-2 , the present invention provides a technical solution:

[0023] See also Figure 1 In this first embodiment, a method for online monitoring of aluminum-plastic film quality based on comparative analysis is provided. The method uses an inspection instrument to perform online monitoring of the aluminum-plastic film quality. Two industrial CCD cameras are mounted on the left and right sides of the aluminum-plastic film. A high-brightness LED linear focused cold light source is used to backlight the aluminum-plastic film. The CCD cameras and a high-speed image processing system perform real-time online scanning to monitor online whether the aluminum-plastic film has scratches and pits. This replaces manual visual inspection, improves detection efficiency, and detects extremely subtle defects that are invisible to the naked eye. The method includes the following steps:

[0024] S10, establishing a three-dimensional space model, using an inspection instrument to monitor the quality of the aluminum-plastic film, and collecting image data and camera parameter data during the aluminum-plastic film quality monitoring process; the camera parameters include built-in parameters and external parameters of the CCD camera.

[0025] Furthermore, the inspection instrument includes two CCD cameras; wherein, the two CCD cameras are respectively mounted above the aluminum-plastic film, for collecting image data during the quality monitoring process of the aluminum-plastic film, and the two CCD cameras are respectively installed at different positions to form a viewing angle difference; a three-dimensional space model about the X-axis, Y-axis and Z-axis is established, and when the inspection instrument is used to monitor the quality of the aluminum-plastic film, the corresponding camera parameters of the two CCD cameras in the three-dimensional space model are determined respectively; wherein, the built-in parameters include focal length and principal point; the external parameters include rotation matrix and translation vector.

[0026] It should be noted that the built-in parameters include the focal length and principal point coordinates of the camera in the three-dimensional space model; the external parameters include the rotation matrix and translation vector of the camera in the three-dimensional space model; two CCD cameras are used to simultaneously collect image data, the collected image data is compared and analyzed, and input into the CNN model for aluminum-plastic film defect judgment and identification, thereby improving the system's fault tolerance.

[0027] S20. Based on the image data of aluminum-plastic film with defects stored in the database, a CNN model is established and the CNN model is trained; the collected image data is analyzed and processed, and the processed image data is input into the trained CNN model to identify the aluminum-plastic film defects and the corresponding defect types in the image data.

[0028] Furthermore, the method for training the CNN model is as follows: establishing a database, storing image data of different types of aluminum-plastic film defects in the database, and dividing the image data stored in the database into a training set and a validation set, establishing a CNN model, and using the training set to train the CNN model so that the CNN model can identify aluminum-plastic film defects and corresponding defect types in the image data, and using the validation set to evaluate the performance of the trained CNN model, and adjusting the model parameters according to the recognition accuracy of the CNN model.

[0029] It should be noted that the CNN model is trained based on the image data of defective aluminum-plastic film stored in the database, and the image data of defective aluminum-plastic film in the database is continuously updated; in this embodiment, when the quality of the aluminum-plastic film is determined to be unqualified, the image data corresponding to the aluminum-plastic film is labeled with the defect type and stored in the database, thereby improving the accuracy of the model.

[0030] S30, according to the aluminum-plastic film defects in the identified image data and the camera parameter data, based on the principle of triangulation, respectively determine the position information of the feature points corresponding to each aluminum-plastic film defect in the three-dimensional space model, and obtain the surface functions of different aluminum-plastic film defects in the three-dimensional space model.

[0031] Specifically, the method steps are:

[0032] S301. Analyze the collected camera parameter data, determine the intrinsic parameter matrices K1 and K2 of the two CCD cameras based on the built-in parameters of the CCD cameras, and determine the extrinsic parameters [R1|P1] and [R2|P2] of the two CCD cameras based on the extrinsic parameters of the CCD cameras; wherein R1 and R2 represent the rotation matrices of the two CCD cameras, respectively; and P1 and P2 represent the translation vectors of the two CCD cameras, respectively.

[0033] S302, according to the aluminum-plastic film defects in the identified image data, respectively determine the position information of the feature points corresponding to the aluminum-plastic film defects in the three-dimensional space model, and obtain (X ij ,Y ij ,Z ij ); where X ij 、Y ij 、Z ij The formula to meet the conditions is:

[0034]

[0035] Among them, H1 and H2 represent the distance from the principal point of the two CCD cameras to the coordinate point (X ij ,Y ij ,Z ij ) rays; and Respectively represent the inverse of the internal parameter matrix; [X ij ,Y ij ,X ij ,1] means (X ij ,Y ij ,Z ij ) is a homogeneous coordinate in a three-dimensional space model; i = 1, 2, ... N; j = 1, 2, ... M i ; N represents the number of defects in the aluminum-plastic film; M i represents the number of characteristic points in the i-th aluminum-plastic film defect;

[0036] It should be noted that the feature points correspond to pixel points in the image data; one pixel point represents one feature point; by analyzing the pixel point position information of the aluminum-plastic film defect in the image data and based on the triangulation principle, the position information of the pixel point in the corresponding three-dimensional space model is determined, thereby facilitating the subsequent analysis of the degree of aluminum-plastic film defects, replacing the inaccurate analysis results based on image grayscale.

[0037] In this embodiment, the internal parameter matrix K represents the internal characteristics of the CCD camera and is a 3x3 matrix, expressed as: Where f1 and f2 represent the focal lengths of the CCD camera on the X-axis and Y-axis in the three-dimensional space model; C1 and C2 represent the coordinates of the principal point of the CCD camera in the three-dimensional space model; the external parameter [R|P] represents the position and orientation of the CCD camera in the three-dimensional space model, which is a 3x4 matrix; it is expressed as: Among them, R represents the rotation matrix, which is a 3x3 rotation matrix used to rotate the point in the three-dimensional space model coordinate system to the camera coordinate system; P represents the translation vector, which is a 3x1 translation vector used to displace the camera coordinate system point in the three-dimensional space model coordinate system point; through comparative analysis, the corresponding identical feature points in the image data collected by the two CCD cameras are determined, the identical feature points are matched, and converted into spatial coordinates in the same three-dimensional space model; Among them, H1 and H2 are the rays from the principal point coordinates of the two CCD cameras in the three-dimensional space model to the feature points, and the homogeneous coordinates corresponding to the intersection of the two rays are calculated [X ij ,Y ij ,Z ij ,1], thus obtaining the position information of the characteristic points corresponding to each aluminum-plastic film defect in the three-dimensional space model.

[0038] S303, according to (X ij ,Y ij ,Y ij ), determine the characteristic point sets corresponding to different aluminum-plastic film defects, and obtain the surface functions F1(X, Y, Z), F2(X, Y, Z), ..., F of different aluminum-plastic film defects in the three-dimensional space model. N (X,Y,Z).

[0039] It should be noted that the surface functions F1(X,Y,Z), F2(X,Y,Z), ..., F N X, Y, and Z in (X, Y, Z) have corresponding definition domains for different aluminum-plastic film defects. By obtaining the feature point sets corresponding to different aluminum-plastic film defects, the surface functions of different aluminum-plastic film defects in the three-dimensional space model are determined based on the obtained feature point sets, thereby further analyzing these tiny scratches and voids that are difficult to pass the qualification judgment, thereby improving the accuracy of the system analysis. In this implementation, the surface function is used to determine the shape of the aluminum-plastic film defect, and the aluminum-plastic film defect and defect type are verified, thereby assisting the CNN model in judging the aluminum-plastic film defect.

[0040] S40. Calculate the defect degrees corresponding to different aluminum-plastic film defects based on the surface functions of the different aluminum-plastic film defects in the three-dimensional space model; conduct an online evaluation of the quality of the aluminum-plastic film based on the defect types corresponding to the different aluminum-plastic film defects and the weights of the impact of different defect types on the quality of the aluminum-plastic film, and calculate the comprehensive quality value of the aluminum-plastic film.

[0041] Specifically, the method steps are:

[0042] S401, determine the plane function F(X, Y, Z0) of the aluminum-plastic film in the three-dimensional space model, and according to F1(X, Y, Z), F2(X, Y, Z), ..., F N (X, Y, Z), respectively determine the defect degree corresponding to different aluminum-plastic film defects, according to the calculation formula:

[0043]

[0044] Among them, S i Indicates the defect degree corresponding to the i-th aluminum-plastic film defect; F i (X, Y, ε) represents the surface function of the i-th aluminum-plastic film defect in the three-dimensional space model; ε represents the integral variable of the height in the three-dimensional space model; G i represents the defect area corresponding to the i-th aluminum-plastic film defect; Z0 represents the height of the aluminum-plastic film plane in the three-dimensional space model, and Z0 is a constant;

[0045] In this embodiment, the Z axis on the three-dimensional space model is the horizontal height, the inspection instrument is fixed, the aluminum-plastic film moves through the conveyor belt on the inspection instrument, and the conveyor belt is at the same horizontal height. By determining the horizontal height of the conveyor belt and the thickness of the aluminum-plastic film, the height Z0 of the aluminum-plastic film plane in the three-dimensional space model is determined; by calculating the defect degree corresponding to different aluminum-plastic film defects, under the condition that the defect area of the aluminum-plastic film remains unchanged, the calculated The larger the value, the greater the degree of defects in the aluminum-plastic film, that is, the worse the quality.

[0046] S402, respectively determine the influence weights w1, w2, ..., w on the quality of the aluminum-plastic film under different types of aluminum-plastic film defects. L , and according to the defect types corresponding to different aluminum-plastic film defects, the quality of the aluminum-plastic film is evaluated online to obtain the comprehensive quality value K of the aluminum-plastic film:

[0047]

[0048] in, It represents the influence weight of the defect type corresponding to the i-th aluminum-plastic film defect on the quality of the aluminum-plastic film; L represents the number of aluminum-plastic film defect types.

[0049] S50. Provide a human-computer interaction platform to digitally display the calculated defect degrees corresponding to different aluminum-plastic film defects and the quality assessment values of the aluminum-plastic film; wherein, determine the unqualified threshold values of the defect degrees and the unqualified threshold values of the comprehensive quality values of the aluminum-plastic film under different aluminum-plastic film defect types; when the defect degree and the comprehensive quality value reach the unqualified threshold values, determine that the quality of the aluminum-plastic film is unqualified, and send the unqualified information to the management personnel.

[0050] In this embodiment, the unqualified threshold values of defect levels under different aluminum-plastic film defect types are determined respectively. And the unqualified threshold value K of the comprehensive value of aluminum-plastic film quality max , when there is S i Or when K reaches the unqualified threshold, the aluminum-plastic film is judged to be unqualified and the unqualified information is sent to the management personnel. The management personnel can view the defect degree corresponding to different aluminum-plastic film defects and the quality assessment value of the aluminum-plastic film through the human-computer interaction platform.

[0051] See also Figure 2 , in this embodiment 2: an aluminum-plastic film quality online monitoring system based on comparative analysis is provided, the system comprising a data acquisition module, a database, a defect recognition module, a three-dimensional model analysis module, an intelligent calculation and evaluation module, an interactive display module and an intelligent monitoring module;

[0052] The data acquisition module is used to collect image data and camera parameter data during the quality monitoring process of the aluminum-plastic film; send the collected image data to the defect recognition module; and send the collected camera parameter data to the three-dimensional model analysis module;

[0053] The database is used to store image data of different aluminum-plastic film defect types;

[0054] The defect recognition module is used to establish a CNN model based on the image data of aluminum-plastic film with defects stored in the database and train the CNN model; analyze and process the collected image data, input the processed image data into the trained CNN model, identify the aluminum-plastic film defects in the image data and the corresponding defect types; send the identified aluminum-plastic film defects in the image data to the three-dimensional model analysis module; and send the identified aluminum-plastic film defects in the image data and the corresponding defect types to the intelligent calculation and evaluation module;

[0055] The three-dimensional model analysis module is used to determine the position information of the feature points corresponding to each aluminum-plastic film defect in the three-dimensional space model based on the principle of triangulation according to the aluminum-plastic film defects in the identified image data and the camera parameter data, and obtain the surface functions of different aluminum-plastic film defects in the three-dimensional space model; and send the surface functions of different aluminum-plastic film defects in the three-dimensional space model to the surface functions of different aluminum-plastic film defects in the three-dimensional space model;

[0056] The intelligent calculation and evaluation module is used to calculate the defect degrees corresponding to different aluminum-plastic film defects based on the surface functions of different aluminum-plastic film defects in the three-dimensional space model; perform online evaluation of the quality of the aluminum-plastic film based on the defect types corresponding to the different aluminum-plastic film defects and the influence weights of the different defect types on the quality of the aluminum-plastic film, and calculate the comprehensive quality value of the aluminum-plastic film; and send the calculated defect degrees corresponding to the different aluminum-plastic film defects and the quality evaluation values of the aluminum-plastic film to the interactive display module;

[0057] The interactive display module is used to provide a human-computer interaction platform to digitally display the defect degrees corresponding to different calculated aluminum-plastic film defects and the quality assessment values of the aluminum-plastic film;

[0058] The intelligent monitoring module is used to determine the unqualified threshold values of the defect degree and the unqualified threshold values of the comprehensive quality value of the aluminum-plastic film under different aluminum-plastic film defect types; monitor the defect degree and quality assessment value of the aluminum-plastic film corresponding to different aluminum-plastic film defects displayed in the interactive display module; when the defect degree and the comprehensive quality value reach the unqualified threshold value, the aluminum-plastic film quality is determined to be unqualified, and the unqualified information is sent to the management personnel.

[0059] In this embodiment:

[0060] When the degree of defects under the defect type of the aluminum-plastic film reaches the unqualified threshold, the intelligent monitoring module determines that the quality of the aluminum-plastic film is unqualified and sends the unqualified quality information of the aluminum-plastic film to the interactive display module. The interactive display module records the defect type of the unqualified aluminum-plastic film and displays it digitally for the convenience of management personnel to view.

[0061] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A method for online monitoring of aluminum-plastic film quality based on comparative analysis, characterized in that: The method comprises the following steps: S10, establishing a three-dimensional spatial model about the X-axis, Y-axis, and Z-axis, using an inspection instrument to monitor the quality of the aluminum-plastic film, and collecting image data and camera parameter data during the aluminum-plastic film quality monitoring process; the camera parameters include built-in parameters and external parameters of the CCD camera; S20, establishing a CNN model based on the image data of aluminum-plastic film with defects stored in the database, and training the CNN model; analyzing and processing the collected image data, inputting the processed image data into the trained CNN model, and identifying the aluminum-plastic film defects and corresponding defect types in the image data; S30. Based on the recognized aluminum-plastic film defects and camera parameter data in the image data, position information of feature points corresponding to each aluminum-plastic film defect in the three-dimensional space model is determined based on the principle of triangulation, and surface functions of different aluminum-plastic film defects in the three-dimensional space model are obtained; wherein the feature points correspond to pixels in the image data; and each pixel represents a feature point; by obtaining feature point sets corresponding to different aluminum-plastic film defects, surface functions of different aluminum-plastic film defects in the three-dimensional space model are determined based on the obtained feature point sets; S40, calculating the defect degrees corresponding to the different aluminum-plastic film defects based on the surface functions of the different aluminum-plastic film defects in the three-dimensional space model; performing an online assessment of the quality of the aluminum-plastic film based on the defect types corresponding to the different aluminum-plastic film defects and the weights of the impact of the different defect types on the quality of the aluminum-plastic film, and calculating a comprehensive quality value of the aluminum-plastic film; The method steps of step S40 are: S401, determine the plane function F(X, Y, Z0) of the aluminum-plastic film in the three-dimensional space model, and calculate the surface functions F1(X, Y, Z), Ff2(X, Y, Z), ..., Ff3(X, Y, Z) of different aluminum-plastic film defects in the three-dimensional space model. N (X, Y, Z), respectively determine the defect degree corresponding to different aluminum-plastic film defects, according to the calculation formula: Among them, S i Indicates the defect degree corresponding to the i-th aluminum-plastic film defect; F i (X, Y, ε) represents the surface function of the i-th aluminum-plastic film defect in the three-dimensional space model; ε represents the integral variable of the height in the three-dimensional space model; G i represents the defect area corresponding to the i-th aluminum-plastic film defect; Z0 represents the height of the aluminum-plastic film plane in the three-dimensional space model, and Z0 is a constant value; N represents the number of aluminum-plastic film defects; S402, respectively determine the influence weights w1, w2, ..., w on the quality of the aluminum-plastic film under different types of aluminum-plastic film defects. L , and according to the defect types corresponding to different aluminum-plastic film defects, the quality of the aluminum-plastic film is evaluated online to obtain the comprehensive quality value K of the aluminum-plastic film: in, represents the influence weight of the defect type corresponding to the i-th aluminum-plastic film defect on the quality of the aluminum-plastic film; L represents the number of aluminum-plastic film defect types; The Z axis on the three-dimensional space model is the horizontal height. The inspection instrument is fixed in place. The aluminum-plastic film moves via the conveyor belt on the inspection instrument. The conveyor belt is at the same horizontal height. By determining the horizontal height of the conveyor belt and the thickness of the aluminum-plastic film, the height Z0 of the aluminum-plastic film plane in the three-dimensional space model is determined. S50. Provide a human-computer interaction platform to digitally display the calculated defect degrees corresponding to different aluminum-plastic film defects and the quality assessment values of the aluminum-plastic film; wherein, determine the unqualified threshold values of the defect degrees and the unqualified threshold values of the comprehensive quality values of the aluminum-plastic film under different aluminum-plastic film defect types; when the defect degree and the comprehensive quality value reach the unqualified threshold values, determine that the quality of the aluminum-plastic film is unqualified, and send the unqualified information to the management personnel.

2. The method for online monitoring of aluminum-plastic film quality based on comparative analysis according to claim 1, wherein: The inspection instrument includes two CCD cameras; wherein, the two CCD cameras are respectively mounted above the aluminum-plastic film to collect image data during the quality monitoring process of the aluminum-plastic film, and the two CCD cameras are respectively installed at different positions to form a viewing angle difference; when the inspection instrument is used to monitor the quality of the aluminum-plastic film, the corresponding camera parameters of the two CCD cameras in the three-dimensional space model are determined respectively; wherein, the built-in parameters include the camera focal length and the camera principal point; the external parameters include the camera rotation matrix and the translation vector.

3. The method for online monitoring of aluminum-plastic film quality based on comparative analysis according to claim 1, wherein: The method for training the CNN model is as follows: establishing a database, storing image data of different types of aluminum-plastic film defects in the database, and dividing the image data stored in the database into a training set and a validation set, establishing a CNN model, and using the training set to train the CNN model so that the CNN model can identify aluminum-plastic film defects and corresponding defect types in the image data, and using the validation set to evaluate the performance of the trained CNN model, and adjusting the model parameters according to the recognition accuracy of the CNN model.

4. The method for online monitoring of aluminum-plastic film quality based on comparative analysis according to claim 2, wherein: The method steps of step S30 are: S301. Analyze the collected camera parameter data, determine the intrinsic parameter matrices K1 and K2 of the two CCD cameras based on the built-in parameters of the CCD cameras, and determine the extrinsic parameters [R1|P1] and [R2|P2] of the two CCD cameras based on the extrinsic parameters of the CCD cameras; wherein R1 and R2 represent the rotation matrices of the two CCD cameras, respectively; and P1 and P2 represent the translation vectors of the two CCD cameras, respectively. S302, according to the aluminum-plastic film defects in the identified image data, respectively determine the position information of the feature points corresponding to the aluminum-plastic film defects in the three-dimensional space model, and obtain (X ij ,Y ij ,Z ij ); where X ij 、Y ij 、Z ij The formula to meet the conditions is: Among them, H1 and H2 represent the distance from the principal point of the two CCD cameras to the coordinate point (X ij ,Y ij ,Z ij ) rays; and Respectively represent the inverse of the internal parameter matrix; [X ij ,Y ij ,Z ij ,1] means (X ij ,Y ij ,Z ij ) is a homogeneous coordinate in a three-dimensional space model; i = 1, 2, ... N; j = 1, 2, ... M i ; N represents the number of defects in the aluminum-plastic film; M i represents the number of characteristic points in the i-th aluminum-plastic film defect; S303, according to (X ij ,Y ij ,Y ij ), determine the characteristic point sets corresponding to different aluminum-plastic film defects, and obtain the surface functions F1(X, Y, Z), F2(X, Y, Z), ..., F of different aluminum-plastic film defects in the three-dimensional space model. N (X,Y,Z).

5. An online monitoring system for aluminum-plastic film quality based on comparative analysis, using the online monitoring method for aluminum-plastic film quality based on comparative analysis according to any one of claims 1 to 4, characterized in that: The system includes a data acquisition module, a database, a defect recognition module, a three-dimensional model analysis module, an intelligent calculation and evaluation module, an interactive display module and an intelligent monitoring module; The data acquisition module is used to collect image data and camera parameter data during the quality monitoring process of the aluminum-plastic film; send the collected image data to the defect recognition module; and send the collected camera parameter data to the three-dimensional model analysis module; The database is used to store image data of different aluminum-plastic film defect types; The defect recognition module is used to establish a CNN model based on the image data of aluminum-plastic film with defects stored in the database and train the CNN model; analyze and process the collected image data, input the processed image data into the trained CNN model, identify the aluminum-plastic film defects and corresponding defect types in the image data; send the identified aluminum-plastic film defects in the image data to the three-dimensional model analysis module; and send the identified aluminum-plastic film defects and corresponding defect types in the image data to the intelligent calculation and evaluation module; The three-dimensional model analysis module is used to establish a three-dimensional space model. Based on the aluminum-plastic film defects in the identified image data and the camera parameter data, the module determines the position information of the feature points corresponding to each aluminum-plastic film defect in the three-dimensional space model based on the principle of triangulation, and obtains the surface functions of different aluminum-plastic film defects in the three-dimensional space model; the surface functions of different aluminum-plastic film defects in the three-dimensional space model are sent to the intelligent calculation and evaluation module; The intelligent calculation and evaluation module is used to calculate the defect degree corresponding to different aluminum-plastic film defects according to the surface function of different aluminum-plastic film defects in the three-dimensional space model; Based on the defect types corresponding to different aluminum-plastic film defects and the weights of the impact of different defect types on the quality of the aluminum-plastic film, the quality of the aluminum-plastic film is evaluated online to calculate the comprehensive quality value of the aluminum-plastic film; the calculated defect degrees corresponding to the different aluminum-plastic film defects and the quality evaluation values of the aluminum-plastic film are sent to the interactive display module; The interactive display module is used to provide a human-computer interaction platform to digitally display the defect degrees corresponding to different calculated aluminum-plastic film defects and the quality assessment values of the aluminum-plastic film; The intelligent monitoring module is used to determine the unqualified threshold value of the defect degree under different aluminum-plastic film defect types and the unqualified threshold value of the comprehensive value of the aluminum-plastic film quality; The defect degree and quality assessment value of the aluminum-plastic film corresponding to different aluminum-plastic film defects displayed in the interactive display module are monitored. When the defect degree and the comprehensive quality value reach the unqualified threshold, the aluminum-plastic film is judged to be unqualified and the unqualified information is sent to the management personnel.

Images