Light-transmitting film defect detection method, device, equipment, storage medium and product

Abstract

The invention discloses a light-transmitting film defect detection method and device, equipment, a storage medium and a product, and the method comprises the steps: carrying out the defect detection of a target light-transmitting film, and obtaining the actual light field distribution characteristics of the target light-transmitting film; according to the actual light field distribution characteristics, performing defect parameter inversion on the target light-transmitting film by adopting a preset defect inversion theoretical model to obtain an actual defect result of the target light-transmitting film; wherein the preset defect inversion theoretical model is constructed through theoretical light field distribution characteristics of the light-transmitting film. The preset defect inversion theoretical model can be constructed based on the theoretical light field distribution characteristics, and the precision of light-transmitting film defect detection is improved in combination with the theoretical light field distribution characteristics and the actual light field distribution characteristics.

Description

Technical Field

[0001] The present invention relates to the field of defect detection technology, and in particular to a light-transmitting film defect detection method, device, equipment, storage medium and product. Background Art

[0002] Light-transmitting films are widely used in high-end industrial manufacturing industries such as integrated circuits, optoelectronic devices, and micro-nano manufacturing. With the significant improvement in modern processing accuracy, the traditional manual visual light-transmitting film defect detection method can no longer meet the needs of large-scale industrial automated production. At present, the light-transmitting film defect detection method based on machine vision is mainly used, but this method also has some limitations. On the one hand, the detection accuracy of this method depends to a large extent on the quality of the defect image and the accuracy of feature annotation. On the other hand, due to the high difficulty in collecting defect samples inside the light-transmitting film and the insufficient number of defect samples, the detection accuracy of this method is difficult to improve. Therefore, how to improve the accuracy of light-transmitting film defect detection is a problem that needs to be solved urgently. Summary of the invention

[0003] The present invention provides a method, device, equipment, storage medium and product for detecting defects in transparent films. A preset defect inversion theoretical model is constructed based on theoretical light field distribution characteristics, and the accuracy of defect detection in transparent films is improved by combining theoretical light field distribution characteristics with actual light field distribution characteristics.

[0004] In order to achieve the above object, an embodiment of the present invention provides a method for detecting defects of a light-transmitting film, comprising:

[0005] Performing defect detection on a target light-transmitting film to obtain actual light field distribution characteristics of the target light-transmitting film;

[0006] According to the actual light field distribution characteristics, a preset defect inversion theoretical model is used to perform defect parameter inversion on the target light-transmitting film to obtain the actual defect result of the target light-transmitting film; wherein the preset defect inversion theoretical model is constructed based on the theoretical light field distribution characteristics of the light-transmitting film.

[0007] As an improvement of the above solution, the preset defect inversion theoretical model is constructed based on the theoretical light field distribution characteristics of the light-transmitting film, including:

[0008] According to the theoretical light field distribution characteristics of the light-transmitting film, a defect simulation data set of the light-transmitting film is established;

[0009] According to the defect simulation data set, a preset defect inversion theoretical model of the light-transmitting film is constructed.

[0010] As an improvement of the above solution, the defect simulation data set of the light-transmitting film is established according to the theoretical light field distribution characteristics of the light-transmitting film, including:

[0011] Establishing a defect detection simulation model for a light-transmitting film, and setting light source parameters of the defect detection simulation model as well as standard parameters and defect parameters of the light-transmitting film;

[0012] According to the light source parameters, defect simulation is performed on the light-transmitting film under different standard parameter and defect parameter combinations to obtain theoretical scattered light field distribution and theoretical transmitted light field distribution of the light-transmitting film;

[0013] Performing data processing on the theoretical scattered light field distribution and the theoretical transmitted light field distribution to obtain theoretical light field distribution characteristics of the light-transmitting film;

[0014] According to the theoretical light field distribution characteristics, a defect simulation data set corresponding to the light source parameters and the standard parameters is established.

[0015] As an improvement of the above solution, the preset defect inversion theoretical model of the light-transmitting film is constructed according to the defect simulation data set, including:

[0016] Dividing the defect simulation data set into a training set, a validation set and a test set;

[0017] Performing model training on the training set using a deep learning algorithm to obtain a trained model;

[0018] The validation set is used to evaluate the performance of the model, and the evaluated model is optimized through the test set to obtain a preset defect inversion theoretical model of the light-transmitting film.

[0019] As an improvement of the above solution, the defect detection of the target light-transmitting film to obtain the actual light field distribution characteristics of the target light-transmitting film includes:

[0020] Acquire actual standard parameters of the target light-transmitting film; and match actual light source parameters in the optical detection system according to the actual standard parameters;

[0021] According to the actual light source parameters, the target light-transmitting film is subjected to defect detection by the optical detection system to extract the actual scattered light field distribution and the actual transmitted light field distribution of the target light-transmitting film;

[0022] According to the actual scattered light field distribution and the actual transmitted light field distribution, the actual light field distribution characteristics of the target light-transmitting film are obtained.

[0023] As an improvement of the above scheme, according to the actual light field distribution characteristics, a preset defect inversion theoretical model is used to perform defect parameter inversion on the target light-transmitting film to obtain the actual defect result of the target light-transmitting film, including:

[0024] Inputting the actual standard parameters, the actual light source parameters and the actual light field distribution characteristics into a preset defect inversion theoretical model;

[0025] Inverting the defect parameters of the target light-transmitting film, calculating the matching probability between the actual light field distribution feature and the theoretical light field distribution feature, and selecting the theoretical light field distribution feature with the highest matching probability as the target theoretical light field distribution feature;

[0026] The defect parameters corresponding to the target theoretical light field distribution characteristics are used as actual defect results of the target light-transmitting film.

[0027] In order to achieve the above-mentioned object, an embodiment of the present invention provides a light-transmitting film defect detection device, comprising:

[0028] A distribution feature detection module, used to perform defect detection on a target light-transmitting film to obtain actual light field distribution features of the target light-transmitting film;

[0029] The defect result acquisition module is used to perform defect parameter inversion on the target light-transmitting film according to the actual light field distribution characteristics using a preset defect inversion theoretical model to obtain the actual defect result of the target light-transmitting film; wherein the preset defect inversion theoretical model is constructed based on the theoretical light field distribution characteristics of the light-transmitting film.

[0030] In order to achieve the above-mentioned purpose, an embodiment of the present invention provides a light-transmitting film defect detection device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and the processor implements the above-mentioned light-transmitting film defect detection method when executing the computer program.

[0031] In order to achieve the above-mentioned purpose, an embodiment of the present invention further provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the above-mentioned light-transmitting film defect detection method.

[0032] To achieve the above objective, an embodiment of the present invention further provides a computer program product, which is stored in a storage medium and is executed by at least one processor to implement the steps of the above light-transmitting film defect detection method.

[0033] Compared with the prior art, the embodiments of the present invention disclose a method, device, equipment, storage medium and product for detecting defects in a light-transmitting film. By performing defect detection on a target light-transmitting film, the actual light field distribution characteristics of the target light-transmitting film are obtained. According to the actual light field distribution characteristics, a preset defect inversion theoretical model is used to invert the defect parameters of the target light-transmitting film to obtain the actual defect result of the target light-transmitting film. The preset defect inversion theoretical model is constructed based on the theoretical light field distribution characteristics of the light-transmitting film. The preset defect inversion theoretical model can be constructed based on the theoretical light field distribution characteristics, and defect detection can be performed in combination with the theoretical light field distribution characteristics and the actual light field distribution characteristics, thereby improving the accuracy of defect detection of light-transmitting films. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a flow chart of a light-transmitting film defect detection method provided by an embodiment of the present invention;

[0035] Figure 2 A schematic diagram of the structure of an actual optical detection system provided by an embodiment of the present invention;

[0036] Figure 3 It is a structural schematic diagram of a light-transmitting film defect detection device provided by an embodiment of the present invention;

[0037] Figure 4 It is a structural block diagram of a light-transmitting film defect detection device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] It should be noted that the terms "comprises" and "specifically" and any variations of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or inherent to these processes, methods, products or apparatuses.

[0040] See also Figure 1 , Figure 1 1 is a flow chart of a light-transmitting film defect detection method provided by an embodiment of the present invention, the light-transmitting film defect detection method comprising:

[0041] S1, performing defect detection on a target light-transmitting film to obtain actual light field distribution characteristics of the target light-transmitting film;

[0042] S2, according to the actual light field distribution characteristics, use a preset defect inversion theoretical model to invert the defect parameters of the target light-transmitting film to obtain the actual defect results of the target light-transmitting film; wherein the preset defect inversion theoretical model is constructed based on the theoretical light field distribution characteristics of the light-transmitting film.

[0043] Exemplarily, the light-transmitting film defect detection method described in the embodiment of the present invention can be implemented by a defect detection server, and the defect detection server can interact with users and simulation software. When the target light-transmitting film is detected for the first time, it is necessary to first build a preset defect inversion theoretical model. When the target light-transmitting film is detected subsequently, the preset defect inversion theoretical model can be directly used without building it again. For example, a defect detection simulation model of a light-transmitting film is established by FDTD (Finite-Difference Time-Domain) simulation software, and the light source parameters of the defect detection simulation model and the standard parameters and defect parameters of the light-transmitting film are set. The light-transmitting film under different combinations of standard parameters and defect parameters is simulated to obtain the theoretical light field distribution characteristics of the light-transmitting film, so as to establish a defect simulation data set corresponding to various light source parameters and standard parameters (the defect simulation data set contains the theoretical light field distribution characteristics corresponding to various defect parameters under the set light source parameters and standard parameters). After the defect detection server uses deep learning to train, verify and test the simulation data set, the preset defect inversion theoretical model of the light-transmitting film is obtained. After obtaining the preset defect inversion theoretical model, the defect detection server performs defect detection on the target light-transmitting film to obtain the actual light field distribution characteristics of the target light-transmitting film. According to the actual light field distribution characteristics, the preset defect inversion theoretical model is used to invert the defect parameters of the target light-transmitting film, and the actual light field distribution characteristics are compared with various theoretical light field distribution characteristics to find the theoretical light field distribution characteristics that are most similar to the actual light field distribution characteristics, and the defect parameters corresponding to the most similar theoretical light field distribution characteristics are used as the actual defect results of the target light-transmitting film. It is possible to inspect various defects (such as surface defects and internal defects) of the light-transmitting film at the same time to improve the efficiency of light-transmitting film defect detection; by establishing a simulation data set, the theoretical light field distribution characteristics of various defect parameters are obtained, which are used for the inversion of defect parameters in actual detection to solve the problem of lack of internal defect samples of the light-transmitting film; by using the deep learning method to establish a preset defect inversion theoretical model, the problem of inaccurate direct defect inversion caused by the discretization of simulation parameters is solved, and the accuracy of light-transmitting film defect detection is improved.

[0044] Specifically, the preset defect inversion theoretical model is constructed based on the theoretical light field distribution characteristics of the light-transmitting film, including:

[0045] S01, establishing a defect simulation data set of the light-transmitting film according to theoretical light field distribution characteristics of the light-transmitting film;

[0046] S02: constructing a preset defect inversion theoretical model of the light-transmitting film according to the defect simulation data set.

[0047] Exemplarily, a defect detection simulation model for a light-transmitting film is established through simulation software (such as FDTD simulation software), and various light source parameters as well as standard parameters and defect parameters of various light-transmitting films are set in the defect detection simulation model. Under a set light source parameter, light-transmitting films with different combinations of standard parameters and defect parameters are simulated to obtain theoretical light field distribution characteristics corresponding to various defect parameters, and defect simulation data sets corresponding to the light source parameters and standard parameters are obtained. Deep learning is used to train the simulation data sets to obtain a preset defect inversion theoretical model for the light-transmitting film.

[0048] It is worth mentioning that the light source parameters include the wavelength of incident light, the intensity of incident light, the angle of incidence, and the polarization of incident light; the standard parameters include the thickness, refractive index, transmittance, and surface roughness of the light-transmitting film in a defect-free state; the defect parameters include the shape, size, position, number, refractive index, and transmittance of the defects; wherein the threshold ranges for setting defect parameters include: the size of a single defect is usually in the submicron to micron range; the defect shapes usually include regular shapes such as spheres, triangles, rectangles, and irregular shapes customized based on actual production experience; the defect positions can cover all positions on the surface and inside of the light-transmitting film; the number of defects usually includes one defect or multiple different defects; the refractive index and transmittance of the defect are usually related to the defect type. If the defect appears as impurities, bubbles, scratches, etc., its refractive index and transmittance are usually different from the standard parameters of the light-transmitting film. If the defect appears as burrs, wrinkles, etc., its refractive index and transmittance are usually consistent with the standard parameters of the light-transmitting film.

[0049] Specifically, the step S01 includes:

[0050] S011, establishing a defect detection simulation model for a light-transmitting film, and setting light source parameters of the defect detection simulation model as well as standard parameters and defect parameters of the light-transmitting film;

[0051] S012, performing defect simulation on the light-transmitting film under different combinations of standard parameters and defect parameters according to the light source parameters, to obtain theoretical scattered light field distribution and theoretical transmitted light field distribution of the light-transmitting film;

[0052] S013, performing data processing on the theoretical scattered light field distribution and the theoretical transmitted light field distribution to obtain theoretical light field distribution characteristics of the light-transmitting film;

[0053] S014: establishing a defect simulation data set corresponding to the light source parameters and the standard parameters according to the theoretical light field distribution characteristics.

[0054] Exemplarily, a light-transmitting film defect detection simulation model is established through FDTD simulation software. After the light source parameters and standard parameters and defect parameters of the light-transmitting film are set through the FDTD simulation software, the standard parameters and defect parameters are combined to obtain light-transmitting film samples with various defects. Defect simulation is performed on each light-transmitting film sample (e.g., a light-transmitting film with bubble defects), and the theoretical scattered light field distribution and the theoretical transmitted light field distribution corresponding to the light-transmitting film sample are obtained. The theoretical scattered light field distribution and the theoretical transmitted light field distribution are data processed to obtain the theoretical light field distribution characteristics corresponding to the various light-transmitting film samples. The corresponding light source parameters and standard parameters are used as classification indexes to establish a corresponding theoretical light field distribution characteristic simulation data set (defect simulation data set).

[0055] It is worth noting that the theoretical light field distribution characteristics include the maximum transmittance Maximum scattering ratio Asymmetry of scattered light field Transmitted light field asymmetry and light field asymmetry in, is the maximum intensity of the transmitted light, is the maximum intensity of scattered light, I in is the incident light intensity, and are the total intensities of scattered light and transmitted light along the s-polarization direction, and are the total intensities of scattered light and transmitted light along the p-polarization direction, and They represent the total intensity of scattered light and transmitted light along the normal direction of the incident plane, respectively.

[0056] Specifically, the parameter combination method for combining the standard parameters and the defect parameters mainly depends on the following rules:

[0057] Physical feasibility principle: ensure that the set defect parameter combination is physically feasible;

[0058] Comprehensive coverage principle: In the parameter space, try to cover all possible parameter combinations to improve the generalization ability and robustness of the model;

[0059] Actual-driven principle: Based on historical production data or actual measurement results, select common or typical defect parameter combinations for simulation to ensure that the simulation data can reflect the problems that may occur in actual production.

[0060] Specifically, the step S02 includes:

[0061] S021, dividing the defect simulation data set into a training set, a verification set and a test set;

[0062] S022, performing model training on the training set using a deep learning algorithm to obtain a trained model;

[0063] S023, using the verification set to evaluate the performance of the model, and optimizing the evaluated model through the test set to obtain a preset defect inversion theoretical model of the light-transmitting film.

[0064] Exemplarily, the simulation data set is divided into a training set, a validation set and a test set according to a certain ratio, and the training set is trained using a deep learning algorithm. During the model training process, the light source parameters, the standard parameters of the light-transmitting film and the theoretical light field distribution characteristics are used as inputs, and the defect parameters are used as output results; the validation set is used to evaluate and verify the performance of the trained model (such as accuracy, recall rate, precision rate). It can be understood that the validation set is mainly used to adjust the hyperparameters of the model. Hyperparameters are parameters that need to be set manually before the model training begins, such as the number of layers of the neural network, the number of neurons in each layer, the learning rate, etc. By evaluating the performance of the model under different hyperparameter combinations on the validation set (such as accuracy, recall rate, etc.), the optimal hyperparameter settings can be selected. For example, when trying different learning rates to train a neural network, use the validation set to observe which learning rate can make the model obtain the highest accuracy on the validation set, so as to avoid the model from overfitting on the training set and performing poorly on new data; use the test set to optimize and improve the model adjusted by evaluation, so as to adjust the model parameters to improve the accuracy and generalization ability of the model, and obtain the preset defect inversion theoretical model of the light-transmitting film; it can be understood that the test set is a data set used to finally evaluate the performance of the trained model. Its data is never seen by the model during the training and validation process. The evaluation results of the test set are used to measure the performance of the model in practical applications and can truly reflect the generalization ability of the model. In the process of model training, the general process is to first train the model with the training set, adjust the hyperparameters through the validation set during the training process, and finally use the test set to evaluate the performance of the final model, so as to ensure that the model has good generalization ability and can accurately process new data in practical applications.

[0065] It is worth mentioning that the preset defect inversion theoretical model can reflect the accurate mapping relationship between the theoretical light field distribution characteristics and the defect parameters.

[0066] For example, the preset defect inversion theoretical model consists of a deep neural network architecture, which mainly includes the following layers:

[0067] Input layer: receives the light source parameters, the standard parameters and the theoretical light field distribution characteristics as input. For example, suppose the light source parameters are L = (L1, L2, ..., L nL ), the standard parameters of the light-transmitting film are T=(T1,T2,…,T nT ), the theoretical light field distribution characteristic is G=(G1,G2,…,G nG ), where L nL is the nLth light source parameter, T nT is the nTth standard parameter, G nG is the nGth theoretical light field distribution characteristic, and nL, nT and nG represent their respective quantities.

[0068] Feature extraction layer: extract features from the input light source parameters, standard parameters of the light-transmitting film, and theoretical light field distribution features. Here, a fully connected layer (FC Layer) is used, and each type of parameter corresponds to an FC layer. For light source parameters, its feature extraction layer is denoted as FC L ; The feature extraction layer corresponding to the standard parameters of the light-transmitting film is denoted as FC T ; The feature extraction layer corresponding to the theoretical light field distribution feature is recorded as FC G The output of each FC layer is recorded as: h L =σ(W L ·L+b L ), h T =σ(W T ·T+b T ), h G =σ(W G ·G+b G ), where h L 、h T 、h G are the output vectors corresponding to the light source parameters, the standard parameters of the light-transmitting film and the theoretical light field distribution characteristics, respectively, L , W T , W G are the weight matrices corresponding to the light source parameters, the standard parameters of the light-transmitting film and the theoretical light field distribution characteristics, respectively, L , b T , b G are the bias terms corresponding to the light source parameters, the standard parameters of the light-transmitting film and the theoretical light field distribution characteristics, and σ is the activation function.

[0069] Feature fusion layer: The vectors extracted from the above three types of features are fused. A simple splicing method can be used, namely: h fusion =[h L ,h T ,h G ],h fusion is the fusion vector.

[0070] Hidden layer: further nonlinearly transform the fused features (fused vectors) to capture deeper abstract relationships. Multiple layers of fully connected layers can be set, and the output of each layer is obtained through the activation function: i =σ(W i ·h i-1 +b i ), i = 1, 2, ...; where h i is the output of the i-th layer in the hidden layer, W i is the weight matrix of the i-th layer in the hidden layer, b i is the bias term of the i-th layer in the hidden layer.

[0071] Output layer: The last fully connected layer is used to output the defect parameter prediction value D = W O ·h lhl +b O , where W O is the weight matrix of the output layer, b O is the bias term of the output layer, h lhl is the output of the hidden layer.

[0072] Specifically, the step S1 includes:

[0073] S11, obtaining actual standard parameters of the target light-transmitting film; and matching actual light source parameters in the optical detection system according to the actual standard parameters;

[0074] S12, performing defect detection on the target light-transmitting film through the optical detection system according to the actual light source parameters, and extracting the actual scattered light field distribution and the actual transmitted light field distribution of the target light-transmitting film;

[0075] S13, obtaining actual light field distribution characteristics of the target light-transmitting film according to the actual scattered light field distribution and the actual transmitted light field distribution.

[0076] It is worth noting that matching the actual light source parameters in the optical detection system according to the actual standard parameters is to make the actual detection conditions as close to the theoretical simulation conditions as possible, to ensure as much as possible that the actual standard parameters (such as thickness, refractive index, and transmittance) obtained from the actual film measurement can be found in the standard parameters of the simulation data set, and to select the light source parameters corresponding to the standard parameters in the simulation data set to match the actual light source parameters (such as wavelength, intensity, angle, and polarization) in the optical detection system.

[0077] More specifically, the step S12 includes:

[0078] S121, determining a laser light source of the actual light source parameters according to the actual light source parameters;

[0079] S122, performing defect detection on the target light-transmitting film by the optical detection system according to the laser light source to obtain a scattering signal and a transmission signal of the target light-transmitting film;

[0080] S123, performing data preprocessing on the scattering signal and the transmission signal, and performing light field extraction on the processed scattering signal and the transmission signal to obtain actual scattered light field distribution and actual transmitted light field distribution of the target light-transmitting film.

[0081] For example, Figure 2 As shown, Figure 2 A schematic diagram of the structure of an actual optical detection system provided for an embodiment of the present invention, the actual optical detection system comprises an upper focusing lens 3, a lower focusing lens 4, an upper photodetector 5, and a lower photodetector 6; a laser light source 1 matching the actual light source parameters is used to obliquely illuminate the target light-transmitting film 2 from the upper side of the target light-transmitting film 2; focusing lenses 3 and 4 and high-sensitivity photodetectors 5 and 6 are arranged above and below the target light-transmitting film 2, respectively, and the photodetectors 5 and 6 are used to collect scattered light signals and transmitted light signals generated by defects of the target light-transmitting film. Photoelectric detectors 5 and 6 convert scattered light signals and transmitted light signals into scattered electrical signals and transmitted electrical signals, and then convert them into digital signals through analog-to-digital converters for storage; computer terminals 7 and 8 are used to perform preliminary processing on the digital signals received by photoelectric detectors 5 and 6 after analog-to-digital conversion, including steps such as noise filtering, gain calibration and signal smoothing, to ensure the effectiveness and accuracy of subsequent analysis; computer terminal 9 is used to integrate the scattered light field data and transmitted light field data collected by photoelectric detectors 5 and 6, extract spatial light field distribution data, and obtain characteristics such as light field position distribution, light field intensity distribution, and light field propagation direction, which are used for subsequent calculation of actual light field distribution characteristics.

[0082] It is worth noting that the actual light field distribution characteristics include the actual maximum transmittance Actual maximum scattering ratio Asymmetry of actual scattered light field Asymmetry of actual transmitted light field and the actual light field asymmetry in, is the maximum intensity of the actual transmitted light, is the maximum intensity of the actual scattered light, I′ in is the actual incident light intensity, and are the total intensities of the actual scattered light and the actual transmitted light along the s-polarization direction, and are the total intensities of the actual scattered light and the actual transmitted light along the p-polarization direction, and They represent the total intensity of the actual scattered light and the actual transmitted light along the normal direction of the incident plane, respectively.

[0083] Specifically, the step S2 includes:

[0084] S21, inputting the actual standard parameters, the actual light source parameters and the actual light field distribution characteristics into a preset defect inversion theoretical model;

[0085] S22, inverting the defect parameters of the target light-transmitting film, calculating the matching probability between the actual light field distribution feature and the theoretical light field distribution feature, and selecting the theoretical light field distribution feature with the highest matching probability as the target theoretical light field distribution feature;

[0086] S23, taking the defect parameters corresponding to the target theoretical light field distribution characteristics as the actual defect results of the target light-transmitting film.

[0087] Exemplarily, the actual standard parameters of the target light-transmitting film, the actual light source parameters and the actual light field distribution characteristics are input into a preset defect inversion theoretical model; the defect parameters are inverted, the matching probability between the actual light field distribution characteristics and the theoretical light field distribution characteristics is calculated, and the theoretical light field distribution characteristics with the highest matching probability are selected; the defect parameters that generate the theoretical light field distribution characteristics are output as the actual defect results of the target light-transmitting film.

[0088] Specifically, the actual defect result includes: defect type, defect size, defect location, matching probability and confidence.

[0089] Defect type: clearly indicate the type or types of defects detected, such as scratches, bubbles, impurities, wrinkles, etc.

[0090] Defect size: Quantify the size of the defect, such as length, width, area or diameter, so as to evaluate its impact on the optical properties of the film;

[0091] Defect location: record the specific location coordinates of the defect on the film to facilitate later positioning and review;

[0092] Matching probability and confidence: Calculate the matching probability between the actual light field distribution characteristics and the theoretical model, which reflects the reliability of the inversion results. A high matching probability means that the detected defect parameters are closer to the true value.

[0093] The embodiment of the present invention discloses a method for detecting defects of a light-transmitting film. By performing defect detection on a target light-transmitting film, the actual light field distribution characteristics of the target light-transmitting film are obtained. According to the actual light field distribution characteristics, a preset defect inversion theoretical model is used to invert the defect parameters of the target light-transmitting film to obtain the actual defect result of the target light-transmitting film. The preset defect inversion theoretical model can be constructed based on the theoretical light field distribution characteristics, and the theoretical light field distribution characteristics and the actual light field distribution characteristics are combined to improve the accuracy of light-transmitting film defect detection.

[0094] See also Figure 3 , Figure 3 1 is a schematic diagram of a light-transmitting film defect detection device 10 provided in an embodiment of the present invention. The light-transmitting film defect detection device 10 comprises:

[0095] A distribution feature detection module 11 is used to perform defect detection on a target light-transmitting film to obtain actual light field distribution features of the target light-transmitting film;

[0096] The defect result acquisition module 12 is used to perform defect parameter inversion on the target light-transmitting film according to the actual light field distribution characteristics using a preset defect inversion theoretical model to obtain the actual defect result of the target light-transmitting film; wherein the preset defect inversion theoretical model is constructed based on the theoretical light field distribution characteristics of the light-transmitting film.

[0097] Furthermore, the preset defect inversion theoretical model is constructed based on the theoretical light field distribution characteristics of the light-transmitting film, including:

[0098] According to the theoretical light field distribution characteristics of the light-transmitting film, a defect simulation data set of the light-transmitting film is established;

[0099] According to the defect simulation data set, a preset defect inversion theoretical model of the light-transmitting film is constructed.

[0100] A light-transmitting film defect detection device 10 provided in an embodiment of the present invention can implement all processes of the light-transmitting film defect detection method of the above-mentioned embodiment. The functions of each module in the device and the technical effects achieved are respectively the same as the functions and technical effects achieved by the light-transmitting film defect detection method of the above-mentioned embodiment, and will not be repeated here.

[0101] See also Figure 4 , Figure 4 : is a schematic diagram of the structure of a light-transmitting film defect detection device 20 provided in an embodiment of the present invention. The light-transmitting film defect detection device 20 of this embodiment includes: a processor 21, a memory 22, and a computer program stored in the memory 22 and executable on the processor 21. When the processor 21 executes the computer program, the steps in the above-mentioned light-transmitting film defect detection method embodiment are implemented. Alternatively, when the processor 21 executes the computer program, the functions of each module in the above-mentioned light-transmitting film defect detection device embodiment are implemented.

[0102] Exemplarily, the computer program may be divided into one or more modules, which are stored in the memory 22 and executed by the processor 21 to implement the present invention. The one or more modules may be a series of computer program instruction segments capable of implementing specific functions, and the instruction segments are used to describe the execution process of the computer program in the light-transmitting film defect detection device 20.

[0103] The light-transmitting film defect detection device 20 may be a computing device such as a desktop computer, a notebook, a PDA, and a cloud server. The light-transmitting film defect detection device 20 may include, but is not limited to, a processor 21 and a memory 22. Those skilled in the art may understand that the schematic diagram is only an example of the light-transmitting film defect detection device 20 and does not constitute a limitation on the light-transmitting film defect detection device 20. The light-transmitting film defect detection device 20 may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the light-transmitting film defect detection device 20 may also include input and output devices, network access devices, buses, etc.

[0104] The processor 21 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor 21 is the control center of the light-transmitting film defect detection device 20, and uses various interfaces and lines to connect various parts of the entire light-transmitting film defect detection device 20.

[0105] The memory 22 can be used to store the computer program and / or module. The processor 21 realizes various functions of the light-transmitting film defect detection device 20 by running or executing the computer program and / or module stored in the memory 22 and calling the data stored in the memory 22. The memory 22 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.). In addition, the memory 22 can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0106] Wherein, if the module integrated in the light-transmitting film defect detection device 20 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor 21, the steps of the above-mentioned method embodiments can be implemented. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media does not include electrical carrier signals and telecommunication signals.

[0107] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0108] An embodiment of the present invention further provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the light-transmitting film defect detection method as described in the above embodiment.

[0109] In addition, an embodiment of the present invention further provides a computer program product, which is stored in a storage medium and is executed by at least one processor to implement the steps of the light-transmitting film defect detection method of the above embodiment.

[0110] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for detecting defects in a light-transmitting film, characterized in that: include: Performing defect detection on a target light-transmitting film to obtain actual light field distribution characteristics of the target light-transmitting film; According to the actual light field distribution characteristics, a preset defect inversion theoretical model is used to perform defect parameter inversion on the target light-transmitting film to obtain the actual defect result of the target light-transmitting film; wherein the preset defect inversion theoretical model is constructed based on the theoretical light field distribution characteristics of the light-transmitting film.

2. The light-transmitting film defect detection method according to claim 1, characterized in that: The preset defect inversion theoretical model is constructed based on the theoretical light field distribution characteristics of the light-transmitting film, including: According to the theoretical light field distribution characteristics of the light-transmitting film, a defect simulation data set of the light-transmitting film is established; According to the defect simulation data set, a preset defect inversion theoretical model of the light-transmitting film is constructed.

3. The light-transmitting film defect detection method according to claim 2, characterized in that: The step of establishing a defect simulation data set of the light-transmitting film according to the theoretical light field distribution characteristics of the light-transmitting film comprises: Establishing a defect detection simulation model for a light-transmitting film, and setting light source parameters of the defect detection simulation model as well as standard parameters and defect parameters of the light-transmitting film; According to the light source parameters, defect simulation is performed on the light-transmitting film under different standard parameter and defect parameter combinations to obtain theoretical scattered light field distribution and theoretical transmitted light field distribution of the light-transmitting film; Performing data processing on the theoretical scattered light field distribution and the theoretical transmitted light field distribution to obtain theoretical light field distribution characteristics of the light-transmitting film; According to the theoretical light field distribution characteristics, a defect simulation data set corresponding to the light source parameters and the standard parameters is established.

4. The light-transmitting film defect detection method according to claim 2, characterized in that: The step of constructing a preset defect inversion theoretical model of the light-transmitting film according to the defect simulation data set includes: Dividing the defect simulation data set into a training set, a validation set and a test set; Performing model training on the training set using a deep learning algorithm to obtain a trained model; The validation set is used to evaluate the performance of the model, and the evaluated model is optimized through the test set to obtain a preset defect inversion theoretical model of the light-transmitting film.

5. The light-transmitting film defect detection method according to claim 1, characterized in that: The defect detection of the target light-transmitting film to obtain the actual light field distribution characteristics of the target light-transmitting film includes: Acquire actual standard parameters of the target light-transmitting film; and match actual light source parameters in the optical detection system according to the actual standard parameters; According to the actual light source parameters, the target light-transmitting film is subjected to defect detection by the optical detection system to extract the actual scattered light field distribution and the actual transmitted light field distribution of the target light-transmitting film; According to the actual scattered light field distribution and the actual transmitted light field distribution, the actual light field distribution characteristics of the target light-transmitting film are obtained.

6. The light-transmitting film defect detection method according to claim 5, characterized in that: According to the actual light field distribution characteristics, a preset defect inversion theoretical model is used to perform defect parameter inversion on the target light-transmitting film to obtain the actual defect result of the target light-transmitting film, including: Inputting the actual standard parameters, the actual light source parameters and the actual light field distribution characteristics into a preset defect inversion theoretical model; Inverting the defect parameters of the target light-transmitting film, calculating the matching probability between the actual light field distribution feature and the theoretical light field distribution feature, and selecting the theoretical light field distribution feature with the highest matching probability as the target theoretical light field distribution feature; The defect parameters corresponding to the target theoretical light field distribution characteristics are used as actual defect results of the target light-transmitting film.

7. A light-transmitting film defect detection device, characterized in that: include: A distribution feature detection module, used to perform defect detection on a target light-transmitting film to obtain actual light field distribution features of the target light-transmitting film; The defect result acquisition module is used to perform defect parameter inversion on the target light-transmitting film according to the actual light field distribution characteristics using a preset defect inversion theoretical model to obtain the actual defect result of the target light-transmitting film; wherein the preset defect inversion theoretical model is constructed based on the theoretical light field distribution characteristics of the light-transmitting film.

8. A light-transmitting film defect detection device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the light-transmitting film defect detection method according to any one of claims 1 to 6 when executing the computer program.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the light-transmitting film defect detection method according to any one of claims 1 to 6.

10. A computer program product, characterized in that The computer program product is stored in a storage medium, and the program product is executed by at least one processor to implement the steps of the light-transmitting film defect detection method according to any one of claims 1 to 6.

Images