PVC film performance detection method and system based on machine vision technology

Abstract

The invention discloses a PVC film performance detection method and system based on a machine vision technology, and relates to the field of PVC film detection. According to the technical scheme, the method comprises the following steps: performing visual inspection on a to-be-detected PVC film sample to obtain defect characteristic parameters of the to-be-detected sample, performing performance test on the to-be-detected PVC film sample to obtain performance index parameters of the to-be-detected sample, and performing gray correlation analysis on the performance evaluation parameters and the defect characteristic parameters to obtain the to-be-detected PVC film sample. The method comprises the following steps: acquiring a performance index of a PVC film and a defect characteristic parameter, acquiring a correlation degree between the performance index of the PVC film and the defect characteristic parameter, constructing an evaluation model according to the acquired parameter and the correlation degree, and calculating a performance evaluation quantized value of the to-be-detected PVC film through the model. According to the invention, real-time performance detection can be carried out on the PVC film in the production and processing process, and a traditional detection mode of sampling a formed finished film is not needed, so that the production and performance detection efficiency of the PVC film is improved, and the detection complexity is reduced.

Description

Technical Field

[0001] The present invention relates to the field of PVC film detection, and in particular to a PVC film performance detection method and system based on machine vision technology. Background Art

[0002] Polyvinyl chloride (PVC) is a high molecular polymer formed by the polymerization of vinyl chloride (VC) monomers. It has the advantages of low price, excellent flame retardancy, strong acid and alkali corrosion resistance, anti-microbial invasion, and a wide range of raw material sources. In particular, the price of polyvinyl chloride is only about one-tenth of the price of polyvinylidene fluoride (PVDF), so polyvinyl chloride materials are widely used in the construction industry, automobile companies and packaging industry. In addition, they are also used in membrane processes such as microfiltration and ultrafiltration. At present, the most widely used type of PVC film is calendered film, which is to feed the mixed PVC material into an extruder, melt-extrude it under high temperature and high pressure, and the melt-extruded PVC material is processed by a calender to make it into a film. Common defects in the processing process include uneven film thickness, black spots on the film surface due to foreign matter sticking to the surface of the calender roller, fish eyes, cold spots, and waxy or powdery precipitation on the film surface.

[0003] In the prior art, different measurement methods are often used for different properties in the performance test of PVC film. For example, for the test of tensile strength, common methods include experimental mechanical method, universal material testing machine method, hanging method, heat sealing method, etc. For the test of impact resistance in mechanical strength, a pendulum film impact tester is often used for test measurement. The above performance tests are all completed by sampling the finished film. Once the test sample fails to meet the standard, the entire batch of film products often needs to be re-melted and re-molded, which not only increases production costs, but also reduces production efficiency. To this end, we propose a PVC film performance detection method and system based on machine vision technology. Summary of the invention

[0004] The main purpose of the present invention is to provide a PVC film performance detection method and system based on machine vision technology, which can effectively solve the problems in the background technology.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is:

[0006] PVC film performance detection method based on machine vision technology includes:

[0007] Performing visual inspection on the PVC film sample to be tested to obtain defect characteristic parameters of the sample to be tested, wherein the defect characteristic parameters include at least one of defect type, defect area per unit area, number of defect areas per unit area, and distribution position of defect areas;

[0008] Performing a performance test on the PVC film sample to be tested to obtain performance index parameters of the sample to be tested, wherein the performance index includes at least one of longitudinal tensile strength, transverse tensile strength, mechanical strength, and transmittance;

[0009] The performance evaluation parameters and the defect characteristic parameters are subjected to grey correlation analysis to obtain the correlation degree R between the i-th performance index and the j-th defect characteristic parameter of the PVC film. ij The specific process includes the following steps:

[0010] S01: Construct data sequences E and F using the obtained performance index parameters and defect characteristic parameters of the sample to be tested. The expression of the sequence is recorded as:

[0011] E={E1,E2,...,E i}

[0012] F={F1,F2,...,F j}

[0013] In the formula, E i Represented as the i-th performance indicator parameter data set; F j Represented as the j-th defect feature parameter data set;

[0014] S02: Standardize the data values ​​in sequences E and F according to the following formula:

[0015]

[0016] In the formula, e ik is the i-th performance indicator parameter data set E in sequence E i The kth data in e' ik is the i-th performance indicator parameter data set E in sequence E i The standardized data of the kth data in ; is the mean value of the performance index parameter of the i-th item in sequence E; f jk is the j-th defect feature parameter data set F in sequence F i The kth data in; f' jk is the j-th defect feature parameter data set F in sequence F i The standardized data of the kth data in ; is the mean value of the j-th defect feature parameter in the sequence F; λ1 and λ2 are the number of performance indicators and the number of defect feature types respectively; Q is the amount of data;

[0017] S03: Calculate the correlation coefficient between all data points in sequence F and all data points in sequence E. The calculation formula is:

[0018]

[0019] In the formula, r jk Represented as the j-th defect feature parameter dataset F in sequence F i The correlation coefficient between the kth standardized data in and the performance index in sequence E; Δ min and Δ max They are the minimum difference between the performance evaluation quantization value and the maximum difference between the two levels of the defect characteristic parameter standardized data, and the calculation formulas are:

[0020] Δ min =min i (min k (|e' ik -f' jk |))

[0021] Δ max =max i (max k (|e' ik -f' jk |))

[0022] Among them, min(min(·)) means taking the minimum value of all factors in (·) first, and then taking the minimum value of all minimum values; max(max(·)) means taking the maximum value of all factors in (·) first, and then taking the maximum value of all maximum values; ρ is a constant coefficient, and its value range is (0,1), usually ρ=0.5;

[0023] S04: Calculate the correlation degree R between the i-th performance index and the j-th defect characteristic parameter according to the formula ij , the calculation formula is:

[0024] An evaluation model is constructed based on the obtained parameters and correlation degree, and the performance evaluation quantitative value of the tested PVC film is calculated through the model. The expression of the evaluation model is:

[0025]

[0026] Where A is the quantitative value of the performance evaluation of the PVC film to be tested; e i Expressed as the quantitative value of the i-th performance indicator evaluation; b i It is expressed as the coefficient of the quantitative value of the i-th performance indicator evaluation; ε is a constant coefficient; f j Expressed as the jth defect characteristic parameter; ω j It is expressed as the weight of the j-th defect feature parameter; n is the type of performance indicator; m is the type of defect feature.

[0027] PVC film performance detection system based on machine vision technology, including visual detection module, performance test module, data analysis module, and evaluation model building module;

[0028] The visual inspection module is used to perform visual inspection on the PVC film sample to be tested, and obtain defect characteristic parameters of the sample to be tested, including at least one of defect type, defect area per unit area, number of defect areas per unit area, and defect area distribution position;

[0029] The performance testing module performs a performance test on the PVC film sample to be tested to obtain performance index parameters of the sample to be tested, and the performance testing module includes a tensile strength testing submodule, a mechanical strength testing submodule, and a transmittance testing submodule;

[0030] The tensile strength test submodule is used to perform a tensile strength test on the PVC film sample to be tested, and obtain the longitudinal tensile strength and transverse tensile strength parameters of the sample to be tested;

[0031] The mechanical strength test submodule is used to perform a mechanical strength test on the PVC film sample to be tested, and obtain the mechanical strength parameters of the sample to be tested;

[0032] The transmittance test submodule is used to perform a transmittance test on the PVC film sample to be tested, and obtain the transmittance parameters of the sample to be tested;

[0033] The data analysis module is used to perform grey correlation analysis on the performance evaluation parameters and the defect characteristic parameters to obtain the correlation degree R between the i-th performance index of the PVC film and the j-th defect characteristic parameter. ij ;

[0034] The evaluation model building module is used to build an evaluation model according to the acquired parameters and correlation degree, wherein the expression of the evaluation model is:

[0035]

[0036] And the performance evaluation quantitative value A used to characterize the performance of the PVC film to be tested is calculated by the constructed evaluation model;

[0037] The system further includes a memory, a processor, and an electronic program stored in the memory and capable of running on the processor.

[0038] The present invention has the following beneficial effects:

[0039] Compared with the prior art, the technical solution of the present invention performs visual inspection on the PVC film sample to be tested to obtain the defect characteristic parameters of the sample to be tested, performs performance test on the PVC film sample to be tested to obtain the performance index parameters of the sample to be tested, performs grey correlation analysis on the performance evaluation parameters and the defect characteristic parameters to obtain the degree of correlation between the performance index and the defect characteristic parameters of the PVC film, constructs an evaluation model according to the obtained parameters and the degree of correlation, calculates the performance evaluation quantitative value of the PVC film to be tested through the model, analyzes the correlation relationship between the parameters affecting the performance index and the defect characteristic parameters, and calculates the performance evaluation quantitative value based on the defect characteristic parameters. The present invention can perform real-time performance detection on the PVC film in the production and processing process without adopting the traditional detection method of sampling the formed finished film, thereby improving the production and performance detection efficiency of the PVC film, reducing the detection complexity, and saving production costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 This is a flow chart of the PVC film performance detection method based on machine vision technology of the present invention;

[0041] Figure 2 The structure diagram of the PVC film performance detection system based on machine vision technology of the present invention. DETAILED DESCRIPTION

[0042] The present invention will be further described below in conjunction with specific implementation methods, wherein the accompanying drawings are only used for exemplary descriptions and represent only schematic diagrams rather than actual drawings, and should not be understood as limiting the present invention. In order to better illustrate the specific implementation methods of the present invention, some parts of the accompanying drawings may be omitted, enlarged or reduced, and do not represent the size of the actual product.

[0043] Example 1

[0044] like Figure 1-2 As shown, the specific implementation process of the technical solution of the present invention includes the following steps:

[0045] Step 1: Obtain a sample of the PVC film to be tested, perform visual inspection on the sample of the PVC film to be tested, and obtain defect characteristic parameters of the sample to be tested;

[0046] It should be noted that the defect characteristic parameters include at least one of defect type, defect area per unit area, number of defect areas per unit area, and distribution position of defect areas. Among them, common defect types of calendered films include uneven film thickness, black spots, fish eyes, cold spots, waxy or powdery precipitates. By acquiring images of defect types as training samples and extracting image features of each defect in the training samples, the defect type identification of the PVC film sample to be tested can be achieved.

[0047] Step 2: Perform a performance test on the PVC film sample to be tested to obtain the performance index parameters of the sample to be tested;

[0048] It should be noted that the performance indicators include at least one of longitudinal tensile strength, transverse tensile strength, mechanical strength, and transmittance. For the test of tensile strength, common methods include experimental mechanical method, universal material testing machine method, hanging method, heat sealing method, etc. Among them, the experimental mechanical method and the universal material testing machine method are more accurate test methods. The experimental mechanical method mainly calculates the tensile strength value by changing the size and shape of the film, fixing the film on the fixture, and stretching the film at a certain rate, while recording parameters such as tensile strength and elongation; the universal material testing machine method is to clamp the film between two fixtures, stretch it at a certain rate, and record the tensile strength when the film breaks through a force sensor. The test results of this method are more accurate. Yes, but the equipment cost is relatively high. For the test of mechanical strength, taking the impact resistance test as an example, GB / T8809-2015 "Test Method for Pendulum Impact Resistance of Plastic Film" stipulates the pendulum impact resistance test of various plastic films. The test is carried out using a pendulum film impact tester. During the experiment, a hemispherical punch impacts and passes through the film sample at a certain speed, and the energy consumed by the hemispherical punch is measured to evaluate the plastic film's ability to resist pendulum impact. For the transmittance measurement of the film, the spectral transmission method can be used. Its principle is to use a spectrometer to divide the light entering the film into spectra of different wavelengths, and then measure the spectrum transmitted through the film. By comparing the intensity of the transmission spectrum and the incident spectrum, the transmittance of the film can be calculated.

[0049] Step 3: After obtaining the performance evaluation parameters and defect characteristic parameters of the PVC film sample to be tested, a grey correlation analysis is performed on the performance evaluation parameters and the defect characteristic parameters to obtain the correlation degree R between the i-th performance index and the j-th defect characteristic parameter of the PVC film. ij The specific process includes the following steps:

[0050] S01: Construct data sequences E and F using the obtained performance index parameters and defect characteristic parameters of the sample to be tested. The expression of the sequence is recorded as:

[0051] E={E1,E2,...,E i}

[0052] F={F1,F2,...,F j}

[0053] In the formula, E i Represented as the i-th performance indicator parameter data set; F j It is represented as the j-th defect characteristic parameter data set, where the specific expressions of data sequences E and F are:

[0054]

[0055] It should be noted that for qualitative parameters in defect characteristic parameters, such as defect type, the assignment method can be used to quantify the qualitative parameters. For example, common defect types including uneven film thickness, black spots, fish eyes, cold spots, waxy or powdery precipitates are assigned values ​​of 1, 2, 3, 4, and 5 respectively, and then the assigned data are used to construct the data sequence;

[0056] S02: Standardize the data values ​​in sequences E and F according to the following formula:

[0057]

[0058] In the formula, e ik is the i-th performance indicator parameter data set E in sequence E i The kth data in e' ik is the i-th performance indicator parameter data set E in sequence E i The standardized data of the kth data in ; is the mean value of the performance index parameter of the i-th item in sequence E; f jk is the j-th defect feature parameter data set F in sequence F i The kth data in; f' jk is the j-th defect feature parameter data set F in sequence F i The standardized data of the kth data in ; is the mean value of the j-th defect characteristic parameter in the sequence F; λ1 and λ2 are the number of performance indicators and the number of defect characteristic types respectively; Q is the amount of data, and the specific expressions of the standardized sequences E' and F' are:

[0059]

[0060] S03: Calculate the correlation coefficient between all data points in sequence F and all data points in sequence E. The calculation formula is:

[0061]

[0062] In the formula, r jk Represented as the j-th defect feature parameter dataset F in sequence F i The correlation coefficient between the kth standardized data in and the performance index in sequence E; Δ min and Δ max They are the minimum difference between the performance evaluation quantization value and the maximum difference between the two levels of the defect characteristic parameter standardized data, and the calculation formulas are:

[0063] Δ min =min i (min k(|e' ik -f' jk |))

[0064] Δ max =max i (max k (|e' ik -f' jk |))

[0065] Among them, min(min(·)) means taking the minimum value of all factors in (·) first, and then taking the minimum value of all minimum values; max(max(·)) means taking the maximum value of all factors in (·) first, and then taking the maximum value of all maximum values; ρ is a constant coefficient, and its value range is (0,1), usually ρ=0.5;

[0066] S04: Calculate the correlation degree R between the i-th performance index and the j-th defect characteristic parameter according to the formula ij , the calculation formula is:

[0067] Step 4: Construct an evaluation model based on the obtained parameters and correlation degree, and calculate the performance evaluation quantitative value of the PVC film to be tested through the model. The expression of the evaluation model is:

[0068]

[0069] Where A is the quantitative value of the performance evaluation of the PVC film to be tested; e i Expressed as the quantitative value of the i-th performance indicator evaluation; b i It is expressed as the coefficient of the quantitative value of the i-th performance indicator evaluation; ε is a constant coefficient; f j Expressed as the jth defect characteristic parameter; ω j It is expressed as the weight of the j-th defect feature parameter; n is the type of performance indicator; m is the type of defect feature.

[0070] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A PVC film performance detection method based on machine vision technology, characterized in that: include: Perform visual inspection on the PVC film sample to be tested to obtain defect characteristic parameters of the sample to be tested; Perform performance tests on the PVC film samples to be tested to obtain performance index parameters of the samples to be tested; The performance evaluation parameters and the defect characteristic parameters are subjected to grey correlation analysis to obtain the correlation degree R between the i-th performance index and the j-th defect characteristic parameter of the PVC film. ij ; An evaluation model is constructed based on the obtained parameters and correlation degree, and the performance evaluation quantitative value of the tested PVC film is calculated through the model. The expression of the evaluation model is: Where A is the quantitative value of the performance evaluation of the PVC film to be tested; e i Expressed as the quantitative value of the i-th performance indicator evaluation; b i It is expressed as the coefficient of the quantitative value of the i-th performance indicator evaluation; ε is a constant coefficient; f j Expressed as the jth defect characteristic parameter; ω j It is expressed as the weight of the j-th defect feature parameter; n is the type of performance indicator; m is the type of defect feature.

2. The PVC film performance detection method based on machine vision technology according to claim 1 is characterized in that: The defect characteristic parameters include at least one of defect type, defect area per unit area, number of defect areas per unit area, and distribution position of defect areas; the performance indicators include at least one of longitudinal tensile strength, transverse tensile strength, mechanical strength, and transmittance.

3. The PVC film performance detection method based on machine vision technology according to claim 1 is characterized in that: The specific process of grey relational analysis includes the following steps: S01: Construct data sequences E and F using the obtained performance index parameters and defect characteristic parameters of the sample to be tested. The expression of the sequence is recorded as: E={E1,E2,...,E i } F={F1,F2,...,F j } In the formula, E i Represented as the i-th performance indicator parameter data set; F j Represented as the j-th defect feature parameter data set; S02: Standardize the data values ​​in sequences E and F according to the following formula: In the formula, e ik is the i-th performance indicator parameter data set E in sequence E i The kth data in e' ik is the i-th performance indicator parameter data set E in sequence E i The standardized data of the kth data in ; is the mean value of the performance index parameter of the i-th item in sequence E; f jk is the j-th defect feature parameter data set F in sequence F i The kth data in; f' jk is the j-th defect feature parameter data set F in sequence F i The standardized data of the kth data in ; is the mean value of the j-th defect feature parameter in the sequence F; λ1 and λ2 are the number of performance indicators and the number of defect feature types respectively; Q is the amount of data; S03: Calculate the correlation coefficient between all data points in sequence F and all data points in sequence E. The calculation formula is: In the formula, r jk Represented as the j-th defect feature parameter dataset F in sequence F i The correlation coefficient between the kth standardized data in and the performance index in sequence E; Δ min and Δ max They are the minimum difference between the performance evaluation quantization value and the maximum difference between the two levels of the defect characteristic parameter standardized data, and the calculation formulas are: D min =min i (min) k (e' ik -f' jk )) D max =max i (max. k (|e' ik -f' jk |)) Among them, min(min(·)) means taking the minimum value of all factors in (·) first, and then taking the minimum value of all minimum values; max(max(·)) means taking the maximum value of all factors in (·) first, and then taking the maximum value of all maximum values; ρ is a constant coefficient, and its value range is (0,1), usually ρ=0.5; S04: Calculate the correlation degree R between the i-th performance index and the j-th defect characteristic parameter according to the formula ij , the calculation formula is:

4. The PVC film performance detection system based on machine vision technology is characterized by: It includes visual inspection module, performance testing module, data analysis module and evaluation model building module; The visual inspection module is used to perform visual inspection on the PVC film sample to be tested, and obtain defect characteristic parameters of the sample to be tested, including at least one of defect type, defect area per unit area, number of defect areas per unit area, and defect area distribution position; The performance testing module performs a performance test on the PVC film sample to be tested to obtain performance index parameters of the sample to be tested, and the performance testing module includes a tensile strength testing submodule, a mechanical strength testing submodule, and a transmittance testing submodule; The tensile strength test submodule is used to perform a tensile strength test on the PVC film sample to be tested, and obtain the longitudinal tensile strength and transverse tensile strength parameters of the sample to be tested; The mechanical strength test submodule is used to perform a mechanical strength test on the PVC film sample to be tested, and obtain the mechanical strength parameters of the sample to be tested; The transmittance test submodule is used to perform a transmittance test on the PVC film sample to be tested, and obtain the transmittance parameters of the sample to be tested; The data analysis module is used to perform grey correlation analysis on the performance evaluation parameters and the defect characteristic parameters to obtain the correlation degree R between the i-th performance index of the PVC film and the j-th defect characteristic parameter. ij ; The evaluation model building module is used to build an evaluation model according to the acquired parameters and correlation degree, wherein the expression of the evaluation model is: The performance evaluation quantitative value A used to characterize the performance of the PVC film to be tested is obtained by calculation through the constructed evaluation model.

5. The PVC film performance detection system based on machine vision technology according to claim 4 is characterized in that: The system also includes a memory, a processor, and an electronic program stored in the memory and capable of running on the processor, wherein the processor can implement the steps of the method described in any one of claims 1 to 3 when running the electronic program.

Images