Lens production defect detection method and equipment thereof

Abstract

The invention discloses a lens production defect detection method and equipment thereof, and relates to the technical field of visual inspection.The equipment comprises an information acquisition module, a central processing unit, a process regulation and control module and an interactive communication module.The information acquisition module acquires multi-source sensing information at regular time, so that both comprehensiveness and accuracy are improved; high accuracy of visual detection is guaranteed; through an information processing model, quantification is accurate and traceable, defect degree visualization and defect accurate classification are realized, then process defects are positioned, the interference degree of the environment on a detection result is determined, and misjudgment of the defects due to environmental factors is avoided; through linkage of defect data, process precision and environmental interference, a process optimization scheme is generated in a targeted manner, closed-loop response from defect generation to process adjustment is realized, and the production line response speed is improved; and finally, the lens production yield is remarkably improved, and the production cost of long-term detection is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of visual inspection, in particular to a lens production defect detection method and device. BACKGROUND

[0002] Defects generated in the lens production process can lead to decreased optical performance, discomfort and eye health risks such as eye fatigue, among which defects such as bubbles can destroy the optical uniformity of the lens, leading to reduced light transmittance, image distortion, and visual problems such as blurred vision and dizziness, while rough surface lenses can increase the burden on the eyes, causing eye acid, dryness, headache and other visual fatigue symptoms, and long-term wear can further deepen myopia and astigmatism, therefore, lens production defect detection is very important. However, the existing lens production defect detection technology has the problems of single information collection dimension, lack of quantitative and accurate model, and defects such as process control lag and lack of dynamics, weak data collaboration and traceability ability. Since traditional detection only collects lens geometric shape or single optical parameter, it does not cover production process data and external environment data, which is easy to cause data distortion due to environmental light and equipment vibration, affecting defect judgment and accuracy, and the existing detection mostly relies on simple threshold comparison, without establishing a comprehensive scoring model for geometric shape defects and material optical defects, which cannot fully quantify the defect degree, and without correlating the influence of process parameters and environmental interference on the detection results, leading to rough defect classification and missing intermediate states such as repairable minor defects. Therefore, traditional detection only outputs defect results, without linking defect data with production process, and cannot generate real-time control schemes by analyzing process precision and environmental interference, mostly relying on manual post-adjustment, leading to repeated defects; in addition, the existing system data is scattered, such as process data stored in device logs and defect data stored in detection terminals, lacking real-time communication interfaces such as industrial Ethernet, and cannot realize data interaction with other devices on the production line; at the same time, there is no integration and storage of historical data, making it difficult to trace the root cause of defects and providing data support for process optimization. In view of the above technical defects, a solution is proposed. SUMMARY

[0003] The present application aims to solve the problems of single information collection dimension, lack of quantitative and accurate model in the existing lens production defect detection technology, and defects such as process control lag and lack of dynamics, weak data collaboration and traceability ability.

[0004] In order to achieve the above purpose, the present application adopts the following technical scheme: A lens production defect detection method, comprising the following steps: S1, the information collection module collects multi-source sensing information in a timely manner: the multi-source sensing information includes lens attribute data, production process data and external environment data; S2, the central processor constructs an information processing model to analyze the multi-source sensing information: the information processing model includes a lens defect detection sub-model and a lens detection efficiency sub-model; The lens defect detection sub-model analyzes the lens defect degree through the lens attribute data and performs dynamic defect classification marking; the lens detection efficiency sub-model analyzes the process precision and risk degree of lens production defect detection through the production process data and the external environment data, and then comprehensively obtains a lens detection efficiency change function, and generates a lens production process optimization scheme; S3, the process control module controls and executes the lens production process data: by receiving the lens production process optimization scheme, a process control instruction set is generated, and a lens production process control operation is executed, so as to flexibly adjust the lens production defect detection process; S4, the interactive communication module stores, interacts and communicates the data: the analysis data of the information processing model and the lens production process optimization scheme are integrated and stored, and are communicated and fed back to the system terminal, so as to be visually interacted and displayed.

[0005] Further, the specific process of collecting multi-source sensing information is as follows: S1-1, the lens attribute data includes geometric shape parameters and material optical parameters; S1-101, the geometric shape parameters of the lens are collected, including the radius of curvature , the center thickness , the surface roughness , the scratch level , and the bright spot bubble level ; S1-102, the material optical parameters of the lens are collected, including the refractive index , the total transmittance , the diffuse light transmittance , the haze value , and the dispersion coefficient ; S1-2, the production process data includes control execution parameters, production line hardware parameters and system detection parameters; The control execution parameters are collected, including the machining speed Ve, the machining pressure Pe and the machining temperature Te; The production line hardware parameters are collected, including the equipment vibration frequency Fe and the motor speed Ne; The system detection parameters are collected, including the camera resolution Re, the light source intensity Le and the detection accuracy Ae; S1-3, the external environment data includes environmental illumination parameters, mechanical vibration parameters and temperature and humidity parameters; collecting light interference parameters, including light intensity values; collecting mechanical interference parameters, including vibration amplitude and vibration frequency; collecting temperature and humidity fluctuation parameters, including temperature change rate and humidity deviation value.

[0006] Further, the specific process of constructing the information processing model is as follows: S2-1, the specific process of the lens defect detection sub-model is as follows: Through lens attribute data analysis of geometric morphology defects and material optical defects, lens production defect scores are obtained, the lens defect degree is evaluated, and dynamic defect classification marking is performed; S2-2, the specific process of the lens detection efficiency sub-model is as follows: Through production process data analysis of control execution efficiency, software recognition accuracy and hardware imaging quality, process precision index is obtained, and the process precision of lens production defect detection is evaluated; Through external environment data analysis of imaging light interference, mechanical vibration interference and temperature and humidity interference, influence risk index is obtained, and the risk degree of lens production defect detection is analyzed; Then, through the combination of process precision index and influence risk index, lens detection efficiency change function is obtained, and lens production process optimization scheme is generated.

[0007] Further, the specific process of the lens defect detection sub-model is as follows: Through lens attribute data analysis of geometric morphology defects and material optical defects, lens production defect scores are obtained, the lens defect degree is evaluated, and dynamic defect classification marking is performed; S2-101, geometric morphology parameters include curvature radius , center thickness , surface roughness , scratch grade and bright spot bubble grade ; Geometric morphology parameters are weighted and calculated to obtain geometric morphology defect score ; S2-102, material optical parameters include total transmittance , diffuse light transmittance , haze value , refractive index and dispersion coefficient ; Material optical parameters are weighted and calculated to obtain material optical defect score ; S2-103, further through geometric morphology defect score and material optical defect score combined, to calculate the lens defect score Sc; By setting the evaluation interval of the lens defect score Sc as [ , ], the lens defect degree is quantitatively evaluated and classified.

[0008] Further, the specific process of evaluating the process precision of lens production defect detection is: S2-201, by analyzing the production process data, the control execution efficiency, the software recognition accuracy and the hardware imaging quality are obtained, and then the process precision index is obtained, and the process precision of lens production defect detection is evaluated; S2-201-1, the machining speed Ve, the machining pressure Pe, the machining temperature Te, the equipment vibration frequency Fe, the motor speed Ne, the camera resolution Re, the light source intensity Le and the detection accuracy Ae are integrated as the indexes of the production process data; Set the upper limit, lower limit, sample mean and sample standard deviation of each index of the production process data; Obtain the Cpk of each index of the production process data through the process capability index Cpk calculation of single index; S2-201-2, by the process capability index of machining speed Ve, pressure Pe and temperature Te, the control execution efficiency score is obtained by weighted summation calculation and analysis; The hardware imaging quality score is obtained by weighted summation calculation and analysis through the process capability index of equipment vibration frequency Fe and motor speed Ne; The software recognition accuracy score is obtained by weighted summation calculation and analysis through the process capability index of camera resolution Re, light source intensity Le and detection accuracy Ae; S2-201-3, by combining the control execution efficiency score , the hardware imaging quality score and the software recognition accuracy score , the process precision index Zu is obtained by weighted summation calculation; Further, the evaluation interval of the process precision index Zu is set, and the process precision of lens production defect detection is evaluated by interval comparison.

[0009] Further, the specific process of evaluating the risk degree of lens production defect detection is: S2-202, by analyzing the imaging light interference, mechanical vibration interference and temperature and humidity interference through external environment data, the influence risk index is obtained, and the risk degree of lens production defect detection is analyzed; The external environment data includes environmental light parameters, mechanical vibration parameters, and temperature and humidity parameters: S2-202-1, mark the light intensity value as , mark the standard light intensity as , mark the maximum allowable fluctuation deviation as ; Further, the light intensity fluctuation interference degree is obtained, and the light intensity fluctuation interference degree is obtained according to the calculation method of the light intensity fluctuation interference degree , , ; Mark the vibration frequency as , mark the standard interval of the vibration frequency as , , mark the standard value of the vibration frequency as , and further obtain the vibration frequency interference degree ; S2-202-2, analyze the light intensity fluctuation interference degree through the light interference parameter, and assign a preset conversion coefficient to calculate the imaging light interference score , and evaluate the imaging light interference degree; Analyze the vibration amplitude interference degree and the vibration frequency interference degree through the mechanical interference parameter, and calculate the mechanical vibration interference score by weighted calculation, and evaluate the mechanical vibration interference degree; Analyze the temperature change rate interference degree and the humidity deviation value interference degree through the temperature and humidity fluctuation parameter, and calculate the temperature and humidity interference score by weighted calculation, and evaluate the temperature and humidity interference degree; S2-202-3, further combine the imaging light interference score , the mechanical vibration interference score and the temperature and humidity interference score , and calculate the comprehensive influence risk index Zw by weighted summation; Further, set the evaluation interval of the influence risk index Zw, and analyze and evaluate the risk degree of lens production defect detection through interval comparison.

[0010] Further, the specific process of generating the lens production process optimization scheme is as follows: S2-203-1, obtain the lens detection efficiency change function F by combining the process precision index Zu and the influence risk index Zw; S2-203-2, re-introducing the lens defect score Sc, establishing and inputting a vector To a multivariate regression model for training and optimization, thereby generating a lens production process optimization scheme.

[0011] A lens production defect detection device, comprising an information acquisition module, a central processing unit, a process control module and an interactive communication module, the information acquisition module, the central processing unit, the process control module and the interactive communication module are in communication connection, the device applies the lens production defect detection method.

[0012] In summary, due to the adoption of the technical scheme, the beneficial effects of the present application are: The information acquisition module collects multi-source sensing information at regular intervals, achieving comprehensive and accurate improvement, synchronously collecting lens attribute data, production process data and external environment data, achieving full-chain data coverage of lens-process-environment, avoiding detection blind spots caused by single data, and ensuring high accuracy of visual detection; The information processing model makes the quantification accurate and traceable, wherein the lens defect detection sub-model realizes defect degree visualization and accurate defect classification, reduces the waste of qualified product misjudgment and repairable product direct scrapping, and through the lens detection efficiency sub-model for deep analysis, the execution efficiency, hardware imaging quality and software recognition accuracy are quantitatively controlled, and then the process short board is located, the interference degree of environment on the detection result is determined, and the defect misjudgment caused by environmental factors is avoided; The present application generates a process optimization scheme by linking defect data, process precision and environmental interference, realizes closed-loop response from defect generation to process adjustment, avoids mass production of defects through defect detection anomaly early warning, improves the response speed of the production line; finally, through accurate classification of defects, dynamic control of process and real-time early warning, the lens production yield is significantly improved, and the historical data accumulation can continuously optimize the algorithm model parameters, further reducing the long-term detection production cost. BRIEF DESCRIPTION OF DRAWINGS

[0013] Fig. 1 The step schematic diagram of the working process of the present application is shown; Fig. 2 The connection schematic diagram of the device module of the present application is shown. DETAILED DESCRIPTION

[0014] The technical solutions in the embodiments of the present application will be described clearly and completely in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0015] Embodiment 1: As Figs. 1-2 shown, a lens production defect detection device includes an information acquisition module, a central processor, a process control module, and an interactive communication module, wherein the information acquisition module, the central processor, the process control module, and the interactive communication module are communicatively connected; The information acquisition module is configured to collect multi-source sensing information at regular intervals; The central processor is configured to construct an information processing model to analyze the multi-source sensing information and generate a lens production process optimization scheme; The process control module is configured to control and execute lens production process data; The interactive communication module is configured to store, interact, and communicate data: it stores historical production data, algorithm model parameters, etc., and supports real-time data interaction with other devices on the production line through a communication interface, which transmits data using an industrial Ethernet.

[0016] The device works as follows: S1, the information acquisition module collects multi-source sensing information at regular intervals: the multi-source sensing information includes lens attribute data, production process data, and external environment data; The specific process of collecting multi-source sensing information is as follows: S1-1, the lens attribute data includes geometric shape parameters and material optical parameters; S1-101, the geometric shape parameters of the lens are collected, including the radius of curvature , the center thickness , the surface roughness , the scratch level , and the bright spot bubble level ; A geometric shape parameter acquisition unit is set up: it includes an industrial camera, a three-coordinate measuring machine, a roundness tester, etc., which are used to obtain the geometric shape data of the lens; it also includes a darkroom cabinet and an imaging light screen, which are used to eliminate environmental light interference and ensure the accuracy of image acquisition; An industrial camera is used to collect lens images, edge detection and region growing processing are performed through OpenCV to extract the geometric shape parameters of the target region, and a three-coordinate measuring machine or a roundness tester is used to accurately measure the radius of curvature of the lens; the levels of scratch and bright spot bubble are divided according to international standards, and the level range is set to 1-5, with no scratch at level 1 and severe scratch at level 5; S1-102, the material optical parameters of the lens are collected, including the refractive index , the total transmittance , the diffuse light transmittance , the haze value , and the dispersion coefficient ; Setting material optical parameter acquisition unit: including HR-100 haze, reflectometer, spectrometer, etc., for obtaining the material optical parameters of the lens; Using HR-100 haze & reflectometer to measure the total transmittance, diffuse light transmittance and haze value of the lens, in line with international standards; Through the spectrometer to measure the refractive index and dispersion characteristics of the lens, to establish the database of material optical parameters; S1-2, production process data includes control execution parameters, production line hardware parameters and system detection parameters; Acquire control execution parameters, including processing speed Ve, processing pressure Pe, processing temperature Te; Acquire production line hardware parameters, including equipment vibration frequency Fe, motor speed Ne; Acquire system detection parameters, including camera resolution Re, light source intensity Le, detection accuracy Ae; Set production process parameter acquisition unit: including sensor, data acquisition card, system log, etc., for real-time acquisition of control execution parameters, production line hardware parameters and system detection parameters; S1-3, external environment data includes environmental light parameters, mechanical vibration parameters and temperature and humidity parameters; Acquire light interference parameters, including light intensity; Acquire mechanical interference parameters, including vibration amplitude and vibration frequency; Acquire temperature and humidity fluctuation parameters, including temperature change rate and humidity deviation value; Set external environment parameter acquisition unit: including light sensor, vibration sensor, temperature and humidity sensor, for monitoring external environmental interference.

[0017] S2, central processor constructs information processing model to analyze multi-source sensing information: the information processing model includes lens defect detection sub-model and lens detection efficiency sub-model; The specific process of constructing the information processing model is as follows: S2-1, the specific process of lens defect detection sub-model is as follows: Through lens attribute data analysis geometric morphological defects and material optical defects, and then obtain lens production defect score, evaluate lens defect degree, and perform dynamic defect classification marking; S2-101, geometric morphological parameters include curvature radius , center thickness , surface roughness , scratch grade and bright spot bubble grade ; Through weighted calculation of geometric morphological parameters, analyze and obtain geometric morphological defect score :

[0018] wherein, is the standard value of the curvature radius; is the standard value of the center thickness; is the standard value of the surface roughness; the standard values of the curvature radius, the center thickness and the surface roughness are obtained by presetting; , , , and are the weight factors of the curvature radius , the center thickness , the surface roughness , the scratch grade and the bright spot bubble grade , and , , , and sum up to 1, and the weight factors are obtained by presetting after a large amount of experimental data calculation; The curvature radius directly determines the focal length and aberration of the lens, and is the core parameter of optical imaging; the center thickness affects the optical path length and determines the assembly space; the higher the surface roughness , the stronger the surface scattering, the increased stray light and the decreased light transmission uniformity; The curvature radius and the center thickness adopt bidirectional deviation: taking the curvature radius as an example, no matter > (the curvature is too large, and the focal length is shortened) or < (the curvature is too small, and the focal length is lengthened), the bidirectional deviation will destroy the design aberration balance of the optical system, and therefore the absolute value is used to calculate the influence of the bidirectional deviation; The surface roughness adopts unidirectional deviation: the surface roughness is a defect only when > (the lower the roughness, the smoother the surface, and the better the light transmission), and if < is a "more optimal state" and should not be counted as a defect, and therefore only the exceeding deviation is retained by ; S2-102, the material optical parameters include total transmittance , diffuse light transmittance , haze value , refractive index and dispersion coefficient ; The material optical defect score is obtained by weighted calculation through material optical parameters :

[0019] wherein, is the standard value of total transmittance; is the standard value of diffuse light transmittance; is the standard value of haze value; is the standard value of refractive index; is the standard value of dispersion coefficient; , , , and are the weight factors of total transmittance , diffuse light transmittance , haze value , refractive index and dispersion coefficient respectively, and , , , and sum up to 1; S2-103, and then the lens defect score Sc is obtained by combining the geometric defect score and the material optical defect score : ; wherein, and are the weight coefficients of the geometric defect score and the material optical defect score respectively, and and are both greater than 0; the higher the geometric defect score and the material optical defect score , the higher the lens defect score Sc, indicating the higher the lens defect degree; By setting the evaluation interval of the lens defect score Sc as , , the lens defect degree is quantitatively evaluated and classified; When the lens defect score Sc is lower than the evaluation interval , , the lens defect degree is evaluated as qualified, and directly enters the next process; When the lens defect score Sc is in the evaluation interval , , the lens defect degree is evaluated as slight defect, and then is detected after repair; When the lens defect score Sc is higher than the evaluation interval [0, 1], the lens defect degree is evaluated as a serious defect, and the lens is determined as unqualified. , ] is higher than the evaluation interval [0, 1], the lens defect degree is evaluated as a serious defect, and the lens is determined as unqualified.

[0020] S2-2, the specific process of the lens detection performance sub-model is as follows: S2-201, the process precision index is obtained by analyzing and controlling the execution efficiency, software recognition accuracy and hardware imaging quality through production process data, so as to evaluate the process precision of lens production defect detection; S2-201-1, the machining speed Ve, the machining pressure Pe, the machining temperature Te, the equipment vibration frequency Fe, the motor speed Ne, the camera resolution Re, the light source intensity Le and the detection accuracy Ae are integrated as the indexes of the production process data; The upper limit, lower limit, sample mean and sample standard deviation of each index of the production process data are set; The Cpk of each index of the production process data is obtained by single-index process capability index Cpk calculation: ; Among them, the upper limit, lower limit, sample mean and sample standard deviation of any index i of the production process data are respectively marked as 、 、 、 ; S2-201-2, the control execution efficiency score Zu is obtained by weighted summation calculation of the process capability indexes of the machining speed Ve, the pressure Pe and the temperature Te: ; The hardware imaging quality score Zu is obtained by weighted summation calculation of the process capability indexes of the equipment vibration frequency Fe and the motor speed Ne: ; The software recognition accuracy score Zu is obtained by weighted summation calculation of the process capability indexes of the camera resolution Re, the light source intensity Le and the detection accuracy Ae: ; S2-201-3, the process precision index Zu is further obtained by weighted summation calculation of the control execution efficiency score Zu, the hardware imaging quality score Zu and the software recognition accuracy score Zu: , , ; Further, the evaluation interval of the process precision index Zu is set, and the process precision of lens production defect detection is evaluated by interval comparison.

[0021] S2-202, by analyzing the external environment data, the imaging light interference, mechanical vibration interference and temperature and humidity interference are analyzed, and then the influence risk index is obtained, and the risk degree of lens production defect detection is analyzed; The external environment data includes environmental light parameters, mechanical vibration parameters and temperature and humidity parameters: S2-202-1, the light intensity value is marked as , the standard light intensity is marked as , the maximum allowable fluctuation deviation is marked as ; that is, the light intensity value needs to meet ; , and then the light intensity fluctuation interference degree is obtained: ; According to the calculation method of the light intensity fluctuation interference degree , the vibration amplitude, temperature change rate and humidity deviation value interference degrees are obtained respectively, and are marked as , , respectively; The vibration frequency is marked as , the standard interval of the vibration frequency is marked as , , the standard value of the vibration frequency is marked as , and then the vibration frequency interference degree is obtained: ; S2-202-2, by analyzing the light intensity fluctuation interference degree , and assigning a preset conversion coefficient, the imaging light interference score is calculated, and the imaging light interference degree is evaluated; By analyzing the vibration amplitude interference degree and the vibration frequency interference degree , the mechanical vibration interference score is calculated, and the mechanical vibration interference degree is evaluated; By analyzing the temperature change rate interference degree and the humidity deviation value interference degree , the temperature and humidity interference score is calculated, and the temperature and humidity interference degree is evaluated; S2-202-3, then the imaging light interference score , the mechanical vibration interference score and the temperature and humidity interference score are combined, and the influence risk index Zw is obtained by weighted summation calculation; Further, the evaluation interval of the influence risk index Zw is set, and the risk degree of lens production defect detection is evaluated through interval comparison analysis.

[0022] S2-203, further, the lens detection performance change function is comprehensively obtained by combining the process precision index and the influence risk index, and the lens production process optimization scheme is generated; S2-203-1, the lens detection performance change function F is comprehensively obtained by combining the process precision index Zu and the influence risk index Zw; The influence risk index Zw is substituted into the lens detection performance change function F, and the corresponding process precision index Zu is obtained. S2-203-2, the lens defect score Sc is introduced again, and the vector is established and input into the multivariate regression model for training and optimization, thereby generating the lens production process optimization scheme.

[0023] S3, the process control module controls and executes the lens production process data: by receiving the lens production process optimization scheme, generating a process control instruction set, and executing the lens production process control operation, the lens production defect detection process is flexibly adjusted; Among them, through the risk threshold of the lens detection performance change function F, the sample set of the lens defect score Sc, the process precision index Zu and the influence risk index Zw is refined and constructed, and the control execution efficiency score , the hardware imaging quality score and the software recognition accuracy score , and the Cpk value of each index of the production process data are deeply analyzed, and the corresponding parameter threshold is set, thereby presetting the corresponding lens production process optimization operation; For example, when the lens detection performance change function F exceeds the risk threshold , the operator needs to perform control operation through sound and light alarm or short message notification, therefore, the threshold comparison of the control execution efficiency score , the hardware imaging quality score and the software recognition accuracy score is required. When the control execution efficiency score is lower than the corresponding threshold, the control execution parameter is processed, the processing speed is adjusted through the PID controller, the dynamic defect classification and fine control are realized, the control execution parameter adjustment unit is set: including motor driver, regulating valve, etc., which is used to adjust the processing speed, processing pressure and other parameters in real time.

[0024] S4, the interactive communication module stores and interacts and communicates the data: the analysis data of the information processing model and the lens production process optimization scheme are integrated and stored, and are communicated and fed back to the system terminal, so that visual interactive display is carried out.

[0025] In summary, the application effect of the present application is as follows: The present application realizes comprehensive and accurate double promotion by the information acquisition module collecting multi-source sensing information at regular time, synchronously collects lens attribute data, production process data and external environment data, realizes full-chain data coverage of lens-process-environment, avoids the detection blind area caused by single data, and guarantees the high accuracy of visual detection; The present application realizes quantitative accuracy and traceability through the information processing model, wherein the lens defect detection sub-model realizes defect degree visualization and defect accurate classification, reduces the waste of qualified product misjudgment and repairable product direct scrapping, and the lens detection efficiency sub-model is used for deep analysis, so that the execution efficiency, hardware imaging quality and software identification accuracy are quantitatively controlled, and then the process short board is located, the interference degree of the environment on the detection result is determined, and the defect misjudgment caused by the environment factor is avoided; The present application realizes the closed-loop response from defect generation to process adjustment by linking the defect data, process precision and environmental interference, generates a process optimization scheme, avoids the generation of defect batches through defect detection anomaly early warning, improves the production line response speed, finally, the lens production yield is significantly improved through accurate classification of defects, dynamic control of process and real-time early warning, and the historical data accumulation can continuously optimize the algorithm model parameters, and further reduce the long-term detection production cost.

[0026] The above embodiments can be realized by software, hardware, firmware or any combination thereof, in whole or in part. When realized by software, the above embodiments can be realized in the form of a computer program product in whole or in part. Those skilled in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed in the present text can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solutions.

[0027] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, i.e. they can be located in one place or distributed on multiple network units. Part or all of the units can be selected to realize the purpose of the present embodiment scheme according to actual needs.

[0028] The above data processing is to calculate the value without dimension, and the interval and threshold size is set for comparison. The threshold size depends on the number of sample data and the base number set by the person skilled in the art for each group of sample data, as long as it does not affect the proportional relationship between the parameter and the quantized value. The preset parameter is set by the person skilled in the art according to the actual situation.

[0029] The above is only a preferred specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can make equivalent replacement or change within the technical range disclosed by the present application according to the technical solution and the inventive concept of the present application, which should be covered within the protection scope of the present application.

Claims

1. A method for detecting defects in lens manufacturing, characterized in that: Includes the following steps: S1, the information acquisition module periodically collects multi-source sensor information: multi-source sensor information includes lens attribute data, production process data and external environment data; S2, The central processing unit constructs an information processing model to analyze multi-source sensor information: The information processing model includes a lens defect detection sub-model and a lens detection performance sub-model; Among them, the lens defect detection sub-model analyzes the degree of lens defects through lens attribute data and performs dynamic defect classification and labeling; the lens detection efficiency sub-model analyzes the process accuracy and risk level of lens production defect detection through production process data and external environment data, and then comprehensively obtains the lens detection efficiency change function and generates lens production process optimization scheme. S3, the process control module controls and executes lens production process data: by receiving lens production process optimization schemes, it generates a set of process control instructions and executes lens production process control operations, thereby flexibly adjusting the lens production defect detection process; S4, the interactive communication module stores, interacts with, and transmits data: it integrates and stores the analysis data of the information processing model and the optimization scheme of lens production process, and transmits the feedback to the system terminal for visualization and interactive display.

2. The method for detecting defects in lens manufacturing according to claim 1, characterized in that: The specific process of collecting multi-source sensor information is as follows: S1-1, Lens attribute data includes geometric morphology parameters and material optical parameters; Collect the geometric parameters of the lens, including the radius of curvature. Center thickness Surface roughness Scratch level And highlight bubble level ; Collect the optical parameters of the lens material, including refractive index. Total transmittance Diffuse light transmittance Haze value and dispersion coefficient ; S1-2, Production process data includes control execution parameters, production line hardware parameters, and system detection parameters; The control parameters include processing speed Ve, processing pressure Pe, and processing temperature Te; Production line hardware parameters include equipment vibration frequency Fe and motor speed Ne; The system detection parameters include camera resolution Re, light source intensity Le, and detection accuracy Ae; S1-3, External environmental data includes ambient light parameters, mechanical vibration parameters, and temperature and humidity parameters; Light interference parameters include light intensity; mechanical interference parameters include vibration amplitude and vibration frequency; temperature and humidity fluctuation parameters include temperature change rate and humidity deviation.

3. The method for detecting defects in lens manufacturing according to claim 2, characterized in that: The specific process of constructing an information processing model is as follows: S2-1, the specific process of the lens defect detection sub-model is as follows: By analyzing the geometric and material optical defects through lens attribute data, a lens production defect score is obtained to assess the degree of lens defects and perform dynamic defect classification and labeling. S2-2, the specific process of the lens detection performance sub-model is as follows: By analyzing production process data, we can control execution efficiency, software recognition accuracy, and hardware imaging quality, and then comprehensively obtain the process accuracy index to evaluate the process accuracy of lens production defect detection. By analyzing external environmental data, we can detect interference from imaging light, mechanical vibration, and temperature and humidity, and then comprehensively obtain the impact risk index to analyze the risk level of defect detection in lens production. Furthermore, by combining the process precision index and the impact risk index, a comprehensive function for the change in lens testing efficiency is obtained, and an optimization scheme for lens production process is generated.

4. The method for detecting defects in lens manufacturing according to claim 3, characterized in that: The specific process of the lens defect detection sub-model is as follows: By analyzing the geometric and material optical defects through lens attribute data, a lens production defect score is obtained to assess the degree of lens defects and perform dynamic defect classification and labeling. S2-101, geometric parameters include radius of curvature. Center thickness Surface roughness Scratch level And highlight bubble level ; Geometric morphological defect scores are obtained through weighted calculations based on geometric morphological parameters. ; S2-102, material optical parameters include total transmittance. Diffuse light transmittance Haze value Refractive index and dispersion coefficient ; Material optical defect scores are obtained by weighted calculation using material optical parameters. ; S2-103, and then through geometric morphology defect scoring Material optical defect rating Combined, the lens defect score Sc is calculated and obtained; By setting the evaluation range of lens defect score Sc to [ , This allows for the quantitative assessment and classification of lens defects.

5. The method for detecting defects in lens manufacturing according to claim 4, characterized in that: The specific process for evaluating the precision of lens manufacturing defect detection is as follows: S2-201 analyzes production process data to control execution efficiency, software recognition accuracy, and hardware imaging quality, thereby comprehensively obtaining a process accuracy index to evaluate the process accuracy of lens production defect detection. S2-201-1 integrates processing speed Ve, processing pressure Pe, processing temperature Te, equipment vibration frequency Fe, motor speed Ne, camera resolution Re, light source intensity Le, and detection accuracy Ae as indicators of production process data. Set the upper limit, lower limit, sample mean, and sample standard deviation for each indicator of the production process data; The Cpk of each indicator of the production process data is obtained by calculating the process capability index Cpk of a single indicator. S2-201-2 calculates and analyzes the control execution efficiency score by weighted summation of process capability indices (processing speed Ve, pressure Pe, and temperature Te). ; The hardware imaging quality score is obtained by weighted summation of the process capability indices of equipment vibration frequency Fe and motor speed Ne. ; The software's recognition accuracy score is obtained by weighted summation of the process capability index, which includes camera resolution Re, light source intensity Le, and detection accuracy Ae. ; S2-201-3, then score the execution efficiency through control. Hardware imaging quality score Software recognition accuracy score The process accuracy index Zu is obtained by combining these factors and performing a weighted summation calculation. Then, an evaluation range for the process accuracy index Zu is set, and the process accuracy of lens production defect detection is evaluated by comparing the ranges.

6. The method for detecting defects in lens manufacturing according to claim 5, characterized in that: The specific process for assessing the risk level of lens manufacturing defect detection is as follows: S2-202 analyzes external environmental data to identify interference from imaging light, mechanical vibration, and temperature and humidity, thereby comprehensively obtaining an impact risk index and analyzing the risk level of defect detection in lens production. External environmental data includes ambient light parameters, mechanical vibration parameters, and temperature and humidity parameters: S2-202-1, the light intensity value is marked as The standard light intensity is marked as The maximum allowable fluctuation deviation will be marked as ; In turn, obtain the light intensity fluctuation interference degree And according to the light intensity fluctuation interference degree The calculation method is similar; the interference levels of vibration amplitude, temperature change rate, and humidity deviation are obtained respectively, and then marked as follows. , , ; The vibration frequency is labeled as The standard interval of vibration frequency is marked as [ , The standard value of the vibration frequency is marked as... Thus, the vibration frequency interference degree is obtained. ; S2-202-2, Analysis of Light Intensity Fluctuation Interference Degree through Light Interference Parameters It assigns a preset conversion coefficient and calculates the imaging illumination interference score. Assess the degree of illumination interference in the imaging process; Analysis of vibration amplitude interference degree through mechanical interference parameters and vibration frequency interference Weighted calculation of mechanical vibration interference score Assess the degree of mechanical vibration interference; Analysis of temperature change rate disturbance through temperature and humidity fluctuation parameters and humidity deviation value interference Weighted calculation of temperature and humidity interference score Assess the degree of interference from temperature and humidity; S2-202-3, and then scored based on imaging illumination interference. Mechanical vibration interference score Temperature and humidity interference score By combining these factors and using a weighted calculation, the overall impact risk index Zw is obtained. Then, an assessment range for the risk index Zw is set, and the risk level of lens production defect detection is assessed through range comparison analysis.

7. The method for detecting defects in lens manufacturing according to claim 6, characterized in that: The specific process for generating an optimized lens manufacturing process is as follows: S2-203-1, by combining the process precision index Zu and the impact risk index Zw, comprehensively obtains the lens detection efficiency change function F; S2-203-2, further introduce the lens defect score Sc, establish and input vector The multivariate regression model is trained and optimized to generate an optimized solution for lens manufacturing process.

8. A lens manufacturing defect detection device, characterized in that: The device includes an information acquisition module, a central processing unit, a process control module, and an interactive communication module. The information acquisition module, the central processing unit, the process control module, and the interactive communication module are interconnected. The device applies the lens manufacturing defect detection method described in any one of claims 1-7.

Images