Glasses coating quality detection method and system based on color shift analysis

Abstract

The invention discloses a glasses coating quality detection method and system based on color shift analysis. According to the method, transmission spectrum data of the lens of the coated glasses is obtained, and color shift analysis is carried out to obtain color shift information of the lens. According to the information, collecting precision analysis is carried out, the reliability of detection accuracy is evaluated, and a detection reliability matrix is constructed. And further constructing a detection accuracy prediction model based on the matrix, and performing accuracy prediction on the coating quality by using real-time transmission spectrum data. And finally, optimizing an acquisition strategy of the transmission spectrum data according to a prediction result, and carrying out final quality detection. According to the method, color shift analysis and an accurate prediction model are introduced, so that the accuracy and reliability of quality detection of the coating film of the glasses are remarkably improved, human intervention is reduced, and the method is suitable for automatic quality monitoring and real-time detection in the production process of the glasses.

Description

Technical Field

[0001] The present invention relates to the technical field of quality detection, and in particular to a method and system for detecting the quality of eyeglass coating based on color shift analysis. Background Art

[0002] Glasses coating technology plays a vital role in the production process of glasses. Coating can not only improve the durability and UV resistance of lenses, but also improve the optical properties of lenses, such as anti-reflection and anti-fouling. However, during the production process, the control of coating quality is the key to ensure that the final product meets the standards. Quality problems such as uneven coating, coating thickness deviation, and impurities in the coating layer will directly affect the use effect and performance of eyeglass lenses, so it is necessary to accurately test the coating quality.

[0003] At present, transmission spectrum analysis is one of the commonly used methods for testing the quality of eyeglass coatings. By analyzing the transmission spectrum data of coated eyeglass lenses, the optical properties and color shift data of the lenses can be obtained, thereby judging the quality of the coating. However, in the process of collecting transmission spectrum data, it may be interfered by various factors, resulting in reduced accuracy of the collected data, thereby affecting the accuracy of quality inspection. For example, changes in light sources, such as intensity fluctuations and unstable wavelengths, will directly affect the quality of transmission spectrum data. If the light source is unstable or uneven, the collected transmission spectrum data will deviate from the actual value, resulting in errors in the test results.

[0004] In order to overcome these challenges, the present invention proposes a method for detecting the quality of eyeglass coatings based on color shift analysis, which aims to improve the accuracy of data collection by accurately analyzing the color shift information of coated eyeglass lenses. This method can not only effectively improve the accuracy of data collection, but also optimize the data collection strategy by constructing a detection reliability prediction model, thereby achieving accurate detection of coating quality. By comprehensively considering multiple factors such as acquisition accuracy, detection reliability, and real-time prediction, the deficiencies in the existing technology can be solved, the accuracy and reliability of coating quality detection can be improved, and production efficiency can be improved and the occurrence of quality problems can be reduced. Summary of the invention

[0005] In order to solve at least one of the above technical problems, the present invention proposes a method and system for detecting eyeglass coating quality based on color shift analysis.

[0006] A first aspect of the present invention provides a method for detecting the quality of eyeglass coatings based on color shift analysis, comprising: Acquire transmission spectrum data of a preset number of coated eyeglass lenses according to a preset time period, perform color shift analysis on the lenses according to the transmission spectrum data, and obtain color shift information of the coated eyeglass lenses; Performing an acquisition accuracy analysis on the transmission spectrum data according to the color shift information, performing a reliability analysis on the quality detection accuracy of the coated eyeglass lenses according to the acquisition accuracy, and constructing a detection reliability matrix; Constructing a detection accuracy prediction model according to the detection reliability matrix, obtaining real-time transmission spectrum data of coated eyeglass lenses, importing the real-time transmission spectrum data into the detection accuracy prediction model to predict the accuracy of eyeglass coating quality detection based on the real-time transmission spectrum data, and obtaining a prediction result; Determining a strategy for acquiring transmission spectrum data according to the prediction result; The quality inspection transmission spectrum data of the coated eyeglass lenses is obtained according to the acquisition strategy to perform coating quality inspection.

[0007] In this solution, the transmission spectrum data of a preset number of coated eyeglass lenses are obtained according to a preset time period, and the color shift information of the coated eyeglass lenses is obtained by performing color shift analysis on the lenses according to the transmission spectrum data, specifically: Acquire the transmission spectrum data and light source data of each position of a preset number of coated eyeglass lenses according to a preset time period, wherein the transmission spectrum data includes transmittance, wavelength range, and spectral transmission curve, and the light source data includes light intensity, light wavelength, and light color; Determine the light transmittance data of each position of the lens according to the transmission spectrum data, wherein the light transmittance data includes the transmittance of light of different wavelengths and the light transmittance; Comparing the light transmittance data with the light source data, calculating the reflectivity of the lens to the light intensity of the light source and the light of different wavelengths; Obtain data on the effect of the reflectivity on the illumination color, and determine color shift information at different positions of the lens according to the effect change data, wherein the color shift information includes color saturation change, hue change, and illumination color change.

[0008] In this solution, the transmission spectrum data is analyzed for acquisition accuracy according to the color shift information, and the quality detection accuracy of the coated eyeglass lenses is analyzed for reliability according to the acquisition accuracy, so as to construct a detection reliability matrix, which is specifically: Constructing a color shift distribution map of a preset number of lenses according to the color shift information, acquiring actual transmission spectrum data of a preset number of coated eyeglass lenses, analyzing actual color shift information of the coated eyeglass lenses according to the actual transmission spectrum data, and constructing an actual color shift distribution map according to the actual color shift information; Comparing the color shift distribution map with the actual color shift distribution map, calculating the Manhattan distance of the color shift at each position between the color shift distribution map and the actual color shift distribution map, determining the degree of color shift difference between the color shift distribution map and the actual color distribution map according to the Manhattan distance, and constructing a color shift difference degree data set; Obtaining the number c of quality detection reliability levels, constructing c clusters based on the fuzzy C-means clustering algorithm, setting the membership factor m and the stop threshold ϵ of the fuzzy C-means clustering algorithm, and initializing a random membership matrix for the color shift difference degree data set according to the fuzzy C-means clustering algorithm; Calculate the center vector of each cluster according to the random membership matrix, use the center vector as the cluster center, calculate the Euclidean distance from each data point in the color shift difference degree data set to each cluster center, update the membership of each data point according to the Euclidean distance, and construct a new membership matrix according to the membership; Determine the difference between the new membership matrix and the random membership matrix, if the difference is less than the stopping threshold, output the clustering result, if the difference is not less than the stopping threshold, iteratively calculate the cluster center of the new membership matrix and update the membership of each data point and the updated membership matrix, until the difference between the updated membership matrix and the previously constructed membership matrix is ​​less than the stopping threshold, and output the clustering result; Determine the detection reliability of the color shift distribution diagram in each cluster for the eyeglass coating quality detection according to the clustering result, obtain the detection reliability data of each cluster, and mark it as cluster-detection reliability data; Obtain the transmission spectrum data corresponding to the color shift distribution diagram in each cluster, extract the data features of each transmission spectrum data, obtain the data features of the transmission spectrum data corresponding to the color shift distribution diagram of each cluster, and mark them as cluster-data features; A detection reliability matrix is ​​constructed according to the detection reliability data of each cluster and the cluster-data features.

[0009] In this solution, the detection accuracy prediction model is constructed according to the detection reliability matrix, the real-time transmission spectrum data of the coated eyeglass lenses is obtained, and the real-time transmission spectrum data is imported into the detection accuracy prediction model to predict the accuracy of eyeglass coating quality detection based on the real-time transmission spectrum data to obtain the prediction results, which are specifically: Constructing an initialization detection accuracy prediction model based on a gradient boosting decision tree algorithm, and setting model parameters of the initialization detection accuracy prediction model, wherein the model parameters include a learning rate, a number of iterations, and a tree depth; Calculating the residual and gradient of the initialized detection accuracy prediction model, constructing a regression tree according to the residual and gradient, importing the detection reliability matrix into the initialized detection accuracy prediction model to fit the regression tree, and updating the model parameters until the prediction accuracy of the initialized detection accuracy prediction model is greater than the preset accuracy, thereby obtaining a detection accuracy prediction model; Real-time transmission spectrum data of coated eyeglass lenses for coating quality detection is obtained, the real-time transmission spectrum data is imported into the detection accuracy prediction model to extract real-time spectrum features of the real-time transmission spectrum, the real-time spectrum features are predicted based on the detection accuracy prediction model, the detection accuracy of the coating quality detection of the eyeglass lenses according to the real-time transmission spectrum data is determined, and a prediction result is obtained.

[0010] In this solution, the strategy for acquiring the transmission spectrum data is determined according to the prediction result, and the transmission spectrum data of the coated eyeglass lens is acquired according to the acquisition strategy to perform coating quality detection, specifically: Acquire detection accuracy requirement data of the current detection batch of coated eyeglass lenses, and determine whether the real-time transmission spectrum data meets the detection accuracy requirement of quality inspection according to the detection accuracy requirement data and the prediction result; If the conditions are met, the real-time transmission spectrum data is used as the transmission spectrum data for the eyeglass film quality detection and calibrated as the first transmission spectrum data acquisition strategy; If not, obtain the test result data of the coated eyeglass lenses of the current test batch that have been tested, calculate the frequency of quality problems at each position of the tested lenses according to the test result data, and determine the importance of testing each position of the coated eyeglass lenses of the current batch that have not been tested for quality according to the quality problem frequency; The detection accuracy requirement of each position is updated according to the detection importance to obtain the detection accuracy requirement update data of each position of the coated eyeglass lens, and the transmission spectrum data of the coated eyeglass lens that has not been subjected to quality inspection is determined according to the detection accuracy requirement update data to obtain the focal area; Acquire quantity information and detection efficiency information of coated eyeglass lenses that have not undergone quality inspection, determine the acquisition efficiency of transmission spectrum data of a single coated eyeglass lens according to the quantity information and the detection efficiency information, and determine the acquisition frequency of transmission spectrum data of the coated eyeglass lenses that have not undergone quality inspection according to the acquisition efficiency; Acquire the transmission spectrum data of the transmission spectrum data acquisition focus area according to the transmission spectrum data acquisition frequency as the transmission spectrum data for eyeglass film quality detection, and calibrate it as the second transmission spectrum data acquisition strategy; The first transmission spectrum data acquisition strategy and the strategy calibrated as the second transmission spectrum data acquisition strategy are integrated into a transmission spectrum data acquisition strategy for quality inspection of coated eyeglass lenses.

[0011] In this solution, the quality inspection transmission spectrum data of the coated eyeglass lenses is obtained according to the acquisition strategy to perform coating quality inspection, specifically: Acquire the quality inspection transmission spectrum data of the coated eyeglass lens according to the acquisition strategy, determine the quality inspection color shift information of each position of the coated eyeglass lens according to the quality inspection transmission spectrum data, and construct a quality inspection color shift information distribution map; Obtaining a standard color shift information distribution diagram of a standard quality lens, comparing the quality inspection color shift information distribution diagram with the standard color shift information distribution diagram, determining a quality inspection color shift difference between the quality inspection color shift information distribution diagram and the standard color shift information, and performing a coating quality inspection on the coated eyeglass lens according to the quality inspection color shift difference to obtain a quality inspection result; The quality inspection results include the type of quality problem and the location of the quality problem. The type of quality problem includes uneven coating, deviation in coating thickness, and impurities in the coating layer.

[0012] The second aspect of the present invention further provides a system for detecting the quality of eyeglass coating based on color shift analysis, the system comprising: a memory and a processor, wherein the memory comprises a program for detecting the quality of eyeglass coating based on color shift analysis, and when the program for detecting the quality of eyeglass coating based on color shift analysis is executed by the processor, the following steps are implemented: Acquire transmission spectrum data of a preset number of coated eyeglass lenses according to a preset time period, perform color shift analysis on the lenses according to the transmission spectrum data, and obtain color shift information of the coated eyeglass lenses; Performing an acquisition accuracy analysis on the transmission spectrum data according to the color shift information, performing a reliability analysis on the quality detection accuracy of the coated eyeglass lenses according to the acquisition accuracy, and constructing a detection reliability matrix; Constructing a detection accuracy prediction model according to the detection reliability matrix, obtaining real-time transmission spectrum data of coated eyeglass lenses, importing the real-time transmission spectrum data into the detection accuracy prediction model to predict the accuracy of eyeglass coating quality detection based on the real-time transmission spectrum data, and obtaining a prediction result; Determining a strategy for acquiring transmission spectrum data according to the prediction result; The quality inspection transmission spectrum data of the coated eyeglass lenses is obtained according to the acquisition strategy to perform coating quality inspection.

[0013] The present invention discloses a method and system for detecting the quality of eyeglass coating based on color shift analysis. The method acquires the transmission spectrum data of the coated eyeglass lens and performs color shift analysis to obtain the color shift information of the lens. Based on the information, an acquisition accuracy analysis is performed and the reliability of the detection accuracy is evaluated to construct a detection reliability matrix. Further, a detection accuracy prediction model is constructed based on this matrix, and the accuracy of the coating quality is predicted using real-time transmission spectrum data. Finally, the acquisition strategy of the transmission spectrum data is optimized according to the prediction results, and the final quality inspection is performed. By introducing color shift analysis and an accurate prediction model, the present invention significantly improves the accuracy and reliability of eyeglass coating quality inspection, reduces human intervention, and is suitable for automated quality monitoring and real-time inspection in the eyeglass production process. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 A flow chart of a method for detecting the quality of eyeglass coating based on color shift analysis according to the present invention is shown; Figure 2 A flow chart of obtaining color shift information of coated eyeglass lenses according to the present invention is shown; Figure 3 A flow chart showing the prediction results obtained by the present invention is shown; Figure 4 A block diagram of a system for detecting the quality of eyeglass coatings based on color shift analysis according to the present invention is shown. DETAILED DESCRIPTION

[0015] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0016] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0017] Figure 1 A flow chart of a method for detecting eyeglass coating quality based on color shift analysis of the present invention is shown.

[0018] like Figure 1 As shown, the first aspect of the present invention provides a method for detecting the quality of eyeglass coating based on color shift analysis, comprising: S102, acquiring transmission spectrum data of a preset number of coated eyeglass lenses according to a preset time period, performing color shift analysis on the lenses according to the transmission spectrum data, and obtaining color shift information of the coated eyeglass lenses; S104, performing an acquisition accuracy analysis on the transmission spectrum data according to the color shift information, performing a reliability analysis on the quality detection accuracy of the coated eyeglass lenses according to the acquisition accuracy, and constructing a detection reliability matrix; S106, constructing a detection accuracy prediction model according to the detection reliability matrix, obtaining real-time transmission spectrum data of coated eyeglass lenses, importing the real-time transmission spectrum data into the detection accuracy prediction model to predict the accuracy of eyeglass coating quality detection based on the real-time transmission spectrum data, and obtaining a prediction result; S108, determining a strategy for acquiring transmission spectrum data according to the prediction result; S110, acquiring the quality inspection transmission spectrum data of the coated eyeglass lenses according to the acquisition strategy to perform coating quality inspection.

[0019] It should be noted that by acquiring the transmittance spectrum data of the lens and performing color shift analysis, the uniformity of the coating layer, the thickness of the film layer, whether there are defects and other quality issues can be intuitively evaluated. As an important representation of the characteristics of the coating layer, color shift can help detect lenses with uneven coating or abnormal quality. By analyzing the color shift information, the accuracy of data collected under different environmental conditions (such as light source changes, equipment deviations, etc.) can be evaluated. This analysis helps identify potential errors in the collection process, thereby reducing the impact of external factors on data quality. Constructing a detection reliability matrix can quantify the impact of different collection accuracies on the accuracy of the test results; the detection accuracy prediction model constructed based on the reliability matrix can analyze the real-time transmittance spectrum data and predict the detection accuracy of the coating quality detection of the lens based on the real-time transmittance spectrum data; based on the prediction results, it can be well judged whether the detection accuracy of the currently acquired lens transmittance spectrum data can meet the accuracy requirements of the coating quality detection. According to the acquisition strategy, the transmittance spectrum data with high detection accuracy can be obtained to perform coating quality detection on the lens, thereby improving the accuracy of the lens coating quality detection.

[0020] Figure 2 A flow chart of obtaining color shift information of coated eyeglass lenses according to the present invention is shown.

[0021] According to an embodiment of the present invention, the transmission spectrum data of a preset number of coated eyeglass lenses are obtained according to a preset time period, and the color shift information of the coated eyeglass lenses is obtained by performing color shift analysis on the lenses according to the transmission spectrum data, specifically: S202, obtaining transmission spectrum data and light source data of each position of a preset number of coated eyeglass lenses according to a preset time period, wherein the transmission spectrum data includes transmittance, wavelength range, and spectral transmission curve, and the light source data includes light intensity, light wavelength, and light color; S204, determining light transmittance data of each position of the lens according to the transmission spectrum data, wherein the light transmittance data includes transmittance of light of different wavelengths and light transmittance; S206, calculating the reflectivity of the lens to the light intensity of the light source and light of different wavelengths according to the light transmittance data and the light source data; S208, obtaining the change data of the influence of the reflectivity on the illumination color, and determining the color shift information of different positions of the lens according to the change data, wherein the color shift information includes the change of color saturation, the change of hue, and the change of illumination color.

[0022] It should be noted that by acquiring the transmission spectrum data (transmittance, wavelength range, spectral transmission curve) and light source data (light intensity, wavelength, color), the optical characteristics of the lens under different lighting conditions can be fully captured. This comprehensive data collection makes the subsequent analysis more accurate and avoids the limitations of single spectrum data or light source conditions. The combination of light source data and transmission spectrum data helps to analyze the coating quality of the lens from different dimensions, overcomes the interference of light source changes on the lens transmission spectrum data, and ensures the multi-angle and multi-level analysis capabilities of the collected data; by calculating the transmittance data of each position of the lens based on the transmission spectrum data (such as the transmittance and light transmittance of different wavelengths of light), the light transmission performance of the lens under different lighting conditions can be understood in detail; by comparing the transmittance data and light source data and calculating the reflectivity, the reflective performance of the lens under lighting can be deeply understood. This analysis helps to identify whether there are problems such as uneven reflection or abnormal reflectivity in the coating layer.

[0023] According to an embodiment of the present invention, the transmission spectrum data is collected with accuracy analysis according to the color shift information, and the quality detection accuracy of the coated eyeglass lenses is analyzed with reliability according to the collection accuracy to construct a detection reliability matrix, specifically as follows: Constructing a color shift distribution map of a preset number of lenses according to the color shift information, acquiring actual transmission spectrum data of a preset number of coated eyeglass lenses, analyzing actual color shift information of the coated eyeglass lenses according to the actual transmission spectrum data, and constructing an actual color shift distribution map according to the actual color shift information; Comparing the color shift distribution map with the actual color shift distribution map, calculating the Manhattan distance of the color shift at each position between the color shift distribution map and the actual color shift distribution map, determining the degree of color shift difference between the color shift distribution map and the actual color distribution map according to the Manhattan distance, and constructing a color shift difference degree data set; Obtaining the number c of quality detection reliability levels, constructing c clusters based on the fuzzy C-means clustering algorithm, setting the membership factor m and the stop threshold ϵ of the fuzzy C-means clustering algorithm, and initializing a random membership matrix for the color shift difference degree data set according to the fuzzy C-means clustering algorithm; Calculate the center vector of each cluster according to the random membership matrix, use the center vector as the cluster center, calculate the Euclidean distance from each data point in the color shift difference degree data set to each cluster center, update the membership of each data point according to the Euclidean distance, and construct a new membership matrix according to the membership; Determine the difference between the new membership matrix and the random membership matrix, if the difference is less than the stopping threshold, output the clustering result, if the difference is not less than the stopping threshold, iteratively calculate the cluster center of the new membership matrix and update the membership of each data point and the updated membership matrix, until the difference between the updated membership matrix and the previously constructed membership matrix is ​​less than the stopping threshold, and output the clustering result; Determine the detection reliability of the color shift distribution diagram in each cluster for the eyeglass coating quality detection according to the clustering result, obtain the detection reliability data of each cluster, and mark it as cluster-detection reliability data; Obtain the transmission spectrum data corresponding to the color shift distribution diagram in each cluster, extract the data features of each transmission spectrum data, obtain the data features of the transmission spectrum data corresponding to the color shift distribution diagram of each cluster, and mark them as cluster-data features; A detection reliability matrix is ​​constructed according to the detection reliability data of each cluster and the cluster-data features.

[0024] It should be noted that, by obtaining the actual color shift information of a preset number of lenses, the actual color shift information is the actual color shift situation of the coated lenses, and the color shift information is the color shift situation analyzed by the transmission spectrum data; by comparing the color shift distribution map with the actual color shift distribution map, the difference between the color shift information of the lens coating and the actual color shift information can be analyzed. If there is a difference between the color shift information and the actual color shift information, it can be considered that the acquisition accuracy of the obtained transmission spectrum data is low, resulting in a difference between the color shift information and the actual color shift information. Therefore, the color shift distribution map with a similar color shift difference degree to the actual color distribution map is clustered into one category according to the fuzzy C-means clustering algorithm, and the color shift distribution maps in each cluster in the clustering results are evaluated for the reliability of coating quality detection, and each type of color shift distribution map is identified for further analysis. The reliability of the coating quality detection is measured, and then the data features of the transmission spectrum data corresponding to each type of color shift distribution diagram are obtained, and finally a detection reliability matrix is ​​constructed. The detection reliability matrix is ​​a reliability level matrix of the data features of the transmission spectrum data of a preset number of coated glasses obtained during the quality detection process. For example, if the reliability of the transmission spectrum data of lens A for quality detection is high, then the corresponding reliability of the data features of the transmission spectrum data of lens A is high. According to the reliability matrix, it is possible to well identify what kind of data features of the transmission spectrum data for coating quality detection have high detection reliability; the clustering result divides the color shift distribution diagrams with similar color shift differences into one category; the cluster-detection reliability data is the detection reliability data corresponding to which cluster, for example, if the detection reliability of cluster A is high, then the cluster-detection reliability data is cluster A-high detection reliability; the greater the color shift difference, the lower the detection reliability. By clustering the color shift difference degree data using the fuzzy C-means clustering algorithm, the potential patterns and laws in the data can be identified, and then different types of color shift distributions can be analyzed. The central vector of each cluster reflects the typical characteristics of this type of color shift; the data characteristics of the transmission spectrum data include transmittance characteristics, color difference characteristics, spectrum projection curve change characteristics, and wavelength range characteristics.

[0025] Figure 3 A flow chart of obtaining prediction results of the present invention is shown.

[0026] According to an embodiment of the present invention, the detection accuracy prediction model is constructed according to the detection reliability matrix, the real-time transmission spectrum data of the coated eyeglass lenses is obtained, and the real-time transmission spectrum data is imported into the detection accuracy prediction model to predict the accuracy of eyeglass coating quality detection based on the real-time transmission spectrum data to obtain the prediction result, which is specifically: S302, constructing an initialization detection accuracy prediction model based on a gradient boosting decision tree algorithm, and setting model parameters of the initialization detection accuracy prediction model, wherein the model parameters include a learning rate, a number of iterations, and a tree depth; S304, calculating the residual and gradient of the initialization detection accuracy prediction model, constructing a regression tree according to the residual and gradient, importing the detection reliability matrix into the initialization detection accuracy prediction model to fit the regression tree, and updating the model parameters until the prediction accuracy of the initialization detection accuracy prediction model is greater than the preset accuracy, thereby obtaining the detection accuracy prediction model; S306, obtaining real-time transmission spectrum data of coated eyeglass lenses for coating quality detection, importing the real-time transmission spectrum data into the detection accuracy prediction model to extract real-time spectrum features of the real-time transmission spectrum, predicting the real-time spectrum features based on the detection accuracy prediction model, determining the detection accuracy of coating quality detection of eyeglass lenses based on the real-time transmission spectrum data, and obtaining a prediction result.

[0027] It should be noted that, in the process of testing the coating quality of coated glasses, the stability of the light source is crucial to the quality of the transmission spectrum data. If the intensity or wavelength of the light source fluctuates, the transmission spectrum data may be inconsistent, resulting in a decrease in the acquisition accuracy. Temperature changes will affect the response of the photoelectric sensor or detector of the spectrometer, thereby affecting the accuracy of the spectral data. Temperature changes may also cause the expansion or contraction of the lens, affecting the light transmission characteristics, thereby affecting the stability of the transmission spectrum. Therefore, by constructing a detection accuracy prediction model to predict the real-time transmission spectrum data collected in real time, the prediction results are obtained. The gradient boosting decision tree algorithm can capture complex nonlinear relationships from the detection reliability matrix by gradually optimizing the model to minimize the prediction error. By analyzing the real-time transmission spectrum data, the model can accurately predict the quality detection accuracy of each data point; by predicting the accuracy of the real-time transmission spectrum data, it can be judged in real time whether the data meets the detection requirements. If the model predicts that the current data quality does not meet the standard, it can reduce repeated tests, avoid unnecessary waste of time and material consumption, improve overall production efficiency, and improve the accuracy of eyeglass coating quality detection.

[0028] According to an embodiment of the present invention, the strategy for acquiring the transmission spectrum data is determined according to the prediction result, and the transmission spectrum data of the coated eyeglass lens is acquired according to the acquisition strategy to perform coating quality detection, specifically: Acquire detection accuracy requirement data of the current detection batch of coated eyeglass lenses, and determine whether the real-time transmission spectrum data meets the detection accuracy requirement of quality inspection according to the detection accuracy requirement data and the prediction result; If the conditions are met, the real-time transmission spectrum data is used as the transmission spectrum data for the eyeglass film quality detection and calibrated as the first transmission spectrum data acquisition strategy; If not, obtain the test result data of the coated eyeglass lenses of the current test batch that have been tested, calculate the frequency of quality problems at each position of the tested lenses according to the test result data, and determine the importance of testing each position of the coated eyeglass lenses of the current batch that have not been tested for quality according to the quality problem frequency; The detection accuracy requirement of each position is updated according to the detection importance to obtain the detection accuracy requirement update data of each position of the coated eyeglass lens, and the transmission spectrum data of the coated eyeglass lens that has not been subjected to quality inspection is determined according to the detection accuracy requirement update data to obtain the focal area; Acquire quantity information and detection efficiency information of coated eyeglass lenses that have not undergone quality inspection, determine the acquisition efficiency of transmission spectrum data of a single coated eyeglass lens according to the quantity information and the detection efficiency information, and determine the acquisition frequency of transmission spectrum data of the coated eyeglass lenses that have not undergone quality inspection according to the acquisition efficiency; Acquire the transmission spectrum data of the transmission spectrum data acquisition focus area according to the transmission spectrum data acquisition frequency as the transmission spectrum data for eyeglass film quality detection, and calibrate it as the second transmission spectrum data acquisition strategy; The first transmission spectrum data acquisition strategy and the strategy calibrated as the second transmission spectrum data acquisition strategy are integrated into a transmission spectrum data acquisition strategy for quality inspection of coated eyeglass lenses.

[0029] It should be noted that, according to the detection accuracy prediction result of the real-time transmission spectrum data, if the real-time transmission spectrum data meets the detection accuracy requirement of the quality inspection, the real-time transmission spectrum data of the lens can be directly used for coating quality inspection; if not, by analyzing the lenses of the current batch that have been inspected, the frequency of coating quality problems at each position of the lenses of the current batch is determined, and the positions with high frequency of problems are given high detection importance, and then the detection accuracy requirement of each position is updated, and the positions with high frequency of problems are given high detection accuracy requirements, and the areas with high detection accuracy requirements are used as the focus areas for the spectrometer to obtain transmission spectrum data. By setting the focus area, the spectrometer can concentrate resources to obtain the transmission spectrum data of the focus area, thereby improving the transmission spectrum data acquisition accuracy of the focus area, and the areas with low frequency of quality problems can set lower detection accuracy requirements, because the probability of problems at these positions is low; finally, the transmission spectrum data acquisition strategy for quality inspection of coated lenses is obtained, and the transmission spectrum data acquired according to the acquisition strategy is used for quality inspection, which can greatly improve the accuracy of coating quality inspection and reduce the impact of low coating quality inspection accuracy caused by low transmission spectrum data acquisition quality; the detection result data includes quality problem type and quality problem location.

[0030] According to an embodiment of the present invention, the quality inspection transmission spectrum data of the coated eyeglass lenses is obtained according to the acquisition strategy to perform coating quality inspection, specifically: Acquire the quality inspection transmission spectrum data of the coated eyeglass lens according to the acquisition strategy, determine the quality inspection color shift information of each position of the coated eyeglass lens according to the quality inspection transmission spectrum data, and construct a quality inspection color shift information distribution map; Obtaining a standard color shift information distribution diagram of a standard quality lens, comparing the quality inspection color shift information distribution diagram with the standard color shift information distribution diagram, determining a quality inspection color shift difference between the quality inspection color shift information distribution diagram and the standard color shift information, and performing a coating quality inspection on the coated eyeglass lens according to the quality inspection color shift difference to obtain a quality inspection result; The quality inspection results include the type of quality problem and the location of the quality problem. The type of quality problem includes uneven coating, deviation in coating thickness, and impurities in the coating layer.

[0031] It should be noted that by obtaining the quality inspection transmission spectrum data according to the acquisition strategy, determining the quality inspection color shift information of each position, and constructing the quality inspection color shift information distribution map, it is possible to accurately identify whether there is a quality problem at the position of each lens. This high-precision analysis method can ensure the accuracy of quality inspection and avoid missed inspections or misjudgments. By comparing the quality inspection color shift information distribution map with the standard color shift information distribution map, the system can accurately determine the abnormalities in the coating process, accurately identify quality problems such as uneven coating, coating thickness deviation, and impurities in the coating layer, and mark the specific location, which enhances the detail of quality problem analysis. The color shift information distribution map of the standard coated eyeglass lens represents an ideal quality state, which is usually a lens manufactured under strictly controlled and optimized production conditions. Ideally, the color shift information distribution map of the standard lens shows uniform and defect-free coating characteristics. By comparing the quality inspection color shift information distribution map with the standard color shift distribution map, these deviations can be quantitatively identified, and then the specific type of quality problem can be inferred. For example, large color shifts may mean that the coating is uneven, large differences in reflectivity in certain areas may indicate thickness deviations in the coating, or localized areas of color difference may indicate impurities or contaminants.

[0032] Figure 4 A block diagram of a system for detecting the quality of eyeglass coatings based on color shift analysis according to the present invention is shown.

[0033] The second aspect of the present invention further provides a system 4 for detecting the quality of eyeglass coating based on color shift analysis, the system comprising: a memory 41 and a processor 42, wherein the memory comprises a program for detecting the quality of eyeglass coating based on color shift analysis, and when the program for detecting the quality of eyeglass coating based on color shift analysis is executed by the processor, the following steps are implemented: Acquire transmission spectrum data of a preset number of coated eyeglass lenses according to a preset time period, perform color shift analysis on the lenses according to the transmission spectrum data, and obtain color shift information of the coated eyeglass lenses; Performing an acquisition accuracy analysis on the transmission spectrum data according to the color shift information, performing a reliability analysis on the quality detection accuracy of the coated eyeglass lenses according to the acquisition accuracy, and constructing a detection reliability matrix; Constructing a detection accuracy prediction model according to the detection reliability matrix, obtaining real-time transmission spectrum data of coated eyeglass lenses, importing the real-time transmission spectrum data into the detection accuracy prediction model to predict the accuracy of eyeglass coating quality detection based on the real-time transmission spectrum data, and obtaining a prediction result; Determining a strategy for acquiring transmission spectrum data according to the prediction result; The quality inspection transmission spectrum data of the coated eyeglass lenses is obtained according to the acquisition strategy to perform coating quality inspection.

[0034] The present invention discloses a method and system for detecting the quality of eyeglass coating based on color shift analysis. The method acquires the transmission spectrum data of the coated eyeglass lens and performs color shift analysis to obtain the color shift information of the lens. Based on the information, an acquisition accuracy analysis is performed and the reliability of the detection accuracy is evaluated to construct a detection reliability matrix. Further, a detection accuracy prediction model is constructed based on this matrix, and the accuracy of the coating quality is predicted using real-time transmission spectrum data. Finally, the acquisition strategy of the transmission spectrum data is optimized according to the prediction results, and the final quality inspection is performed. By introducing color shift analysis and an accurate prediction model, the present invention significantly improves the accuracy and reliability of eyeglass coating quality inspection, reduces human intervention, and is suitable for automated quality monitoring and real-time inspection in the eyeglass production process.

[0035] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0036] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0037] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0038] A person of ordinary skill in the art can understand that: all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: a mobile storage device, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0039] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0040] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for detecting the quality of eyeglass coating based on color shift analysis, characterized in that: The following steps are involved: Acquire transmission spectrum data of a preset number of coated eyeglass lenses according to a preset time period, perform color shift analysis on the lenses according to the transmission spectrum data, and obtain color shift information of the coated eyeglass lenses; Performing an acquisition accuracy analysis on the transmission spectrum data according to the color shift information, performing a reliability analysis on the quality detection accuracy of the coated eyeglass lenses according to the acquisition accuracy, and constructing a detection reliability matrix; Constructing a detection accuracy prediction model according to the detection reliability matrix, obtaining real-time transmission spectrum data of coated eyeglass lenses, importing the real-time transmission spectrum data into the detection accuracy prediction model to predict the accuracy of eyeglass coating quality detection based on the real-time transmission spectrum data, and obtaining a prediction result; Determining a strategy for acquiring transmission spectrum data according to the prediction result; The quality inspection transmission spectrum data of the coated eyeglass lenses is obtained according to the acquisition strategy to perform coating quality inspection.

2. The method for detecting eyeglass coating quality based on color shift analysis according to claim 1, characterized in that: The method of acquiring the transmission spectrum data of a preset number of coated eyeglass lenses according to a preset time period, and performing color shift analysis on the lenses according to the transmission spectrum data to obtain the color shift information of the coated eyeglass lenses is specifically as follows: Acquire the transmission spectrum data and light source data of each position of a preset number of coated eyeglass lenses according to a preset time period, wherein the transmission spectrum data includes transmittance, wavelength range, and spectral transmission curve, and the light source data includes light intensity, light wavelength, and light color; Determine the light transmittance data of each position of the lens according to the transmission spectrum data, wherein the light transmittance data includes the transmittance of light of different wavelengths and the light transmittance; Comparing the light transmittance data with the light source data, calculating the reflectivity of the lens to the light intensity of the light source and the light of different wavelengths; Obtain data on the effect of the reflectivity on the illumination color, and determine color shift information at different positions of the lens according to the effect change data, wherein the color shift information includes color saturation change, hue change, and illumination color change.

3. The method for detecting eyeglass coating quality based on color shift analysis according to claim 1, characterized in that: The analysis of the acquisition accuracy of the transmission spectrum data is performed according to the color shift information, and the reliability analysis of the quality detection accuracy of the coated eyeglass lenses is performed according to the acquisition accuracy to construct a detection reliability matrix, specifically: Constructing a color shift distribution map of a preset number of lenses according to the color shift information, acquiring actual transmission spectrum data of a preset number of coated eyeglass lenses, analyzing actual color shift information of the coated eyeglass lenses according to the actual transmission spectrum data, and constructing an actual color shift distribution map according to the actual color shift information; Comparing the color shift distribution map with the actual color shift distribution map, calculating the Manhattan distance of the color shift at each position between the color shift distribution map and the actual color shift distribution map, determining the degree of color shift difference between the color shift distribution map and the actual color distribution map according to the Manhattan distance, and constructing a color shift difference degree data set; Obtaining the number c of quality detection reliability levels, constructing c clusters based on the fuzzy C-means clustering algorithm, setting the membership factor m and the stop threshold ϵ of the fuzzy C-means clustering algorithm, and initializing a random membership matrix for the color shift difference degree data set according to the fuzzy C-means clustering algorithm; Calculate the center vector of each cluster according to the random membership matrix, use the center vector as the cluster center, calculate the Euclidean distance from each data point in the color shift difference degree data set to each cluster center, update the membership of each data point according to the Euclidean distance, and construct a new membership matrix according to the membership; Determine the difference between the new membership matrix and the random membership matrix, if the difference is less than the stopping threshold, output the clustering result, if the difference is not less than the stopping threshold, iteratively calculate the cluster center of the new membership matrix and update the membership of each data point and the updated membership matrix, until the difference between the updated membership matrix and the previously constructed membership matrix is ​​less than the stopping threshold, and output the clustering result; Determine the detection reliability of the color shift distribution diagram in each cluster for the eyeglass coating quality detection according to the clustering result, obtain the detection reliability data of each cluster, and mark it as cluster-detection reliability data; Obtain the transmission spectrum data corresponding to the color shift distribution diagram in each cluster, extract the data features of each transmission spectrum data, obtain the data features of the transmission spectrum data corresponding to the color shift distribution diagram of each cluster, and mark them as cluster-data features; A detection reliability matrix is ​​constructed according to the detection reliability data of each cluster and the cluster-data features.

4. The method for detecting eyeglass coating quality based on color shift analysis according to claim 1, characterized in that: The detection accuracy prediction model is constructed according to the detection reliability matrix, and the real-time transmission spectrum data of the coated eyeglass lenses is obtained. The real-time transmission spectrum data is imported into the detection accuracy prediction model to predict the accuracy of eyeglass coating quality detection based on the real-time transmission spectrum data, and the prediction result is obtained, specifically: Constructing an initialization detection accuracy prediction model based on a gradient boosting decision tree algorithm, and setting model parameters of the initialization detection accuracy prediction model, wherein the model parameters include a learning rate, a number of iterations, and a tree depth; Calculating the residual and gradient of the initialized detection accuracy prediction model, constructing a regression tree according to the residual and gradient, importing the detection reliability matrix into the initialized detection accuracy prediction model to fit the regression tree, and updating the model parameters until the prediction accuracy of the initialized detection accuracy prediction model is greater than the preset accuracy, thereby obtaining a detection accuracy prediction model; Real-time transmission spectrum data of coated eyeglass lenses for coating quality detection is obtained, the real-time transmission spectrum data is imported into the detection accuracy prediction model to extract real-time spectrum features of the real-time transmission spectrum, the real-time spectrum features are predicted based on the detection accuracy prediction model, the detection accuracy of the coating quality detection of the eyeglass lenses according to the real-time transmission spectrum data is determined, and a prediction result is obtained.

5. The method for detecting eyeglass coating quality based on color shift analysis according to claim 1, characterized in that: The method of determining a transmission spectrum data acquisition strategy according to the prediction result, and acquiring the transmission spectrum data of the coated eyeglass lens according to the acquisition strategy to perform coating quality detection, is specifically as follows: Acquire detection accuracy requirement data of the current detection batch of coated eyeglass lenses, and determine whether the real-time transmission spectrum data meets the detection accuracy requirement of quality inspection according to the detection accuracy requirement data and the prediction result; If the conditions are met, the real-time transmission spectrum data is used as the transmission spectrum data for the eyeglass film quality detection and calibrated as the first transmission spectrum data acquisition strategy; If not, obtain the test result data of the coated eyeglass lenses of the current test batch that have been tested, calculate the frequency of quality problems at each position of the tested lenses according to the test result data, and determine the importance of testing each position of the coated eyeglass lenses of the current batch that have not been tested for quality according to the quality problem frequency; The detection accuracy requirement of each position is updated according to the detection importance to obtain the detection accuracy requirement update data of each position of the coated eyeglass lens, and the transmission spectrum data of the coated eyeglass lens that has not been subjected to quality inspection is determined according to the detection accuracy requirement update data to obtain the focal area; Acquire quantity information and detection efficiency information of coated eyeglass lenses that have not undergone quality inspection, determine the acquisition efficiency of transmission spectrum data of a single coated eyeglass lens according to the quantity information and the detection efficiency information, and determine the acquisition frequency of transmission spectrum data of the coated eyeglass lenses that have not undergone quality inspection according to the acquisition efficiency; Acquire the transmission spectrum data of the transmission spectrum data acquisition focus area according to the transmission spectrum data acquisition frequency as the transmission spectrum data for eyeglass film quality detection, and calibrate it as the second transmission spectrum data acquisition strategy; The first transmission spectrum data acquisition strategy and the strategy calibrated as the second transmission spectrum data acquisition strategy are integrated into a transmission spectrum data acquisition strategy for quality inspection of coated eyeglass lenses.

6. The method for detecting eyeglass coating quality based on color shift analysis according to claim 1, characterized in that: The method of obtaining the quality inspection transmission spectrum data of the coated eyeglass lenses according to the acquisition strategy to perform coating quality inspection is specifically as follows: Acquire the quality inspection transmission spectrum data of the coated eyeglass lens according to the acquisition strategy, determine the quality inspection color shift information of each position of the coated eyeglass lens according to the quality inspection transmission spectrum data, and construct a quality inspection color shift information distribution map; Obtaining a standard color shift information distribution diagram of a standard quality lens, comparing the quality inspection color shift information distribution diagram with the standard color shift information distribution diagram, determining a quality inspection color shift difference between the quality inspection color shift information distribution diagram and the standard color shift information, and performing a coating quality inspection on the coated eyeglass lens according to the quality inspection color shift difference to obtain a quality inspection result; The quality inspection results include the type of quality problem and the location of the quality problem. The type of quality problem includes uneven coating, deviation in coating thickness, and impurities in the coating layer.

7. A system for detecting the quality of eyeglass coating based on color shift analysis, characterized in that: The eyeglass coating quality detection system based on color shift analysis includes a storage device and a processor. The storage device includes an eyeglass coating quality detection method program based on color shift analysis. When the eyeglass coating quality detection method program based on color shift analysis is executed by the processor, the following steps are implemented: Acquire transmission spectrum data of a preset number of coated eyeglass lenses according to a preset time period, perform color shift analysis on the lenses according to the transmission spectrum data, and obtain color shift information of the coated eyeglass lenses; Performing an acquisition accuracy analysis on the transmission spectrum data according to the color shift information, performing a reliability analysis on the quality detection accuracy of the coated eyeglass lenses according to the acquisition accuracy, and constructing a detection reliability matrix; Constructing a detection accuracy prediction model according to the detection reliability matrix, obtaining real-time transmission spectrum data of coated eyeglass lenses, importing the real-time transmission spectrum data into the detection accuracy prediction model to predict the accuracy of eyeglass coating quality detection based on the real-time transmission spectrum data, and obtaining a prediction result; Determining a strategy for acquiring transmission spectrum data according to the prediction result; The quality inspection transmission spectrum data of the coated eyeglass lenses is obtained according to the acquisition strategy to perform coating quality inspection.

Images