Molded lens edge thickness detection method based on vision and related device

Abstract

The invention discloses a vision-based molded lens edge thickness detection method and a related device. The method comprises the following steps: introducing visible gas into an edge area of a molded lens to form a gas film disturbance layer; the gas film disturbance layer is excited by illumination to generate refractive index change; collecting an interferometric phase image corresponding to the scattered light intensity; carrying out Fourier transform analysis on the interference fringes and carrying out temperature correction; and finally, calculating edge thickness distribution through feature space mapping. A gas film disturbance layer is introduced to the edge of the lens to serve as a physical signal conversion bridge, and a multi-light-path interference imaging and spectrum analysis technology is matched, so that the interference problem of a transparent material in direct optical measurement can be effectively avoided, high-precision and non-contact detection on the edge thickness of the molded lens is realized, and the detection precision is improved. And meanwhile, the device has the characteristics of high environmental interference resistance and good measurement stability.

Description

Technical Field

[0001] The present invention relates to the technical field of molded lens edge thickness detection, and in particular to a molded lens edge thickness detection method based on vision and a related device. Background Art

[0002] Molded lenses are optical components that are formed by placing optical materials into specific molds under high temperature and high pressure conditions. They are widely used in glasses, cameras, precision instruments and other fields. Compared with traditional grinding processes, molding technology can efficiently mass-produce optical lenses of complex shapes, including high-end optical components such as aspheric surfaces and free-form surfaces, significantly improving production efficiency and reducing costs. During the manufacturing process, the edge thickness of the lens is one of the key parameters that affect the optical performance of the final product. Uneven edge thickness can lead to imaging distortion, optical axis deviation and other problems, so accurate edge thickness detection is required to ensure product quality.

[0003] The existing molded lens edge thickness detection mainly relies on two methods: one is to use contact measurement tools such as micrometers to directly measure the physical size, and the other is to use an optical imaging system to shoot the lens edge from different angles and calculate the edge thickness through image processing algorithms. However, these methods have obvious technical limitations in practical applications. Although contact measurement has high accuracy, it is easy to damage the lens surface, and the detection efficiency is low because it needs to measure point by point; and although the existing optical imaging detection has the advantages of non-contact and high efficiency, when the lens rotates during the detection process, the edge area will produce complex light reflection and refraction phenomena due to the transparent characteristics of the lens material. These optical phenomena lead to blurred edge contours, uneven illumination, and even false boundaries, making it difficult for image processing algorithms to accurately identify the edge position. Especially for special optical lenses with non-uniform thickness design, small thickness changes in the edge area are more difficult to be accurately captured. In addition, these optical interference phenomena will change with the rotation angle of the lens, resulting in systematic deviations in the measurement results of the same lens at different rotation positions, seriously affecting the accuracy and consistency of the detection. Summary of the invention

[0004] The main purpose of the present invention is to solve the technical problem of inaccurate edge thickness detection caused by the optical properties of transparent materials during the rotation detection of existing molded lenses, such as blurred edge contours, uneven lighting and false boundaries.

[0005] The first aspect of the present invention provides a method for detecting the edge thickness of a molded lens based on vision, and the method for detecting the edge thickness of a molded lens based on vision comprises: Introduce a visible gas into the edge region of the molded lens, control the flow rate, thickness, density, and pressure of the visible gas, form a gas film disturbance layer on the circumferential surface of the edge of the molded lens, and obtain the gas film thickness change amount and disturbance deformation amount in the gas film disturbance layer; According to the gas film thickness change amount in the gas film disturbance layer, select a corresponding light incident angle, stimulate the refractive index change generated by the disturbance deformation amount in the gas film disturbance layer, and obtain the refractive index distribution change amount and scattered light intensity change amount in the gas film disturbance layer; Based on the refractive index distribution change amount of the gas film disturbance layer, configure the main light source optical path and the reference optical path, collect the interference phase image corresponding to the scattered light intensity change amount, and obtain the fringe offset sequence and phase jump sequence of the interference phase image; Perform Fourier transform analysis according to the fringe offset sequence and phase jump sequence, extract the corresponding amplitude feature sequence and phase feature sequence, and obtain the steady-state spectrum feature through ambient temperature correction; Map the amplitude feature sequence and phase feature sequence in the steady-state spectrum feature to a preset feature space, and calculate the edge thickness distribution of the molded lens.

[0006] Preferably, the step of introducing a visible gas into the edge region of the molded lens, controlling the flow rate, thickness, density, and pressure of the visible gas, forming a gas film disturbance layer on the circumferential surface of the edge of the molded lens, and obtaining the gas film thickness change amount and disturbance deformation amount in the gas film disturbance layer includes: Detect the linear velocity of the rotating molded lens, determine the initial flow rate of the visible gas according to the linear velocity, and calculate the velocity ratio of the flow rate of the visible gas to the linear velocity; Adjust the pressure coefficient of the visible gas according to the velocity ratio, and determine the density distribution parameter of the visible gas in combination with the ambient temperature; Calculate the gas film adhesion parameter based on the product of the density distribution parameter and the pressure coefficient, and control the introduction amount of the visible gas on the circumferential surface of the edge of the molded lens according to the gas film adhesion parameter; Perform uniformity analysis on the introduction amount of the visible gas to obtain the radial distribution coefficient and circumferential distribution coefficient in the gas film disturbance layer; Calculate the gas film thickness change amount in the gas film disturbance layer according to the radial distribution coefficient and the circumferential distribution coefficient; Perform gradient analysis on the gas film thickness change amount to obtain the disturbance deformation amount in the gas film disturbance layer.

[0007] Preferably, the step of performing uniformity analysis on the introduction amount of the visible gas to obtain the radial distribution coefficient and circumferential distribution coefficient in the gas film disturbance layer includes: Perform a flow velocity gradient analysis on the inflow rate of the visible gas along the radial direction to obtain the radial flow velocity change rate, and obtain the radial gas flow modulation coefficient according to the corresponding relationship between the radial flow velocity change rate and the surface curvature of the molded lens; Perform a circumferential gas flow distribution analysis according to the radial gas flow modulation coefficient to obtain the circumferential flow rate compensation amount, and perform a Coriolis force correction on the circumferential flow rate compensation amount in combination with the rotation speed of the molded lens to obtain the circumferential distribution coefficient; Perform a gas flow stability calculation on the circumferential distribution coefficient along the edge contour of the molded lens to obtain the stability criterion value, and correct the radial gas flow modulation coefficient according to the stability criterion value to obtain the radial distribution coefficient.

[0008] Preferably, based on the change amount of the gas film thickness in the gas film disturbance layer, select the corresponding light incident angle to excite the refractive index change caused by the deformation amount in the gas film disturbance layer, and obtain the change amount of the refractive index distribution and the change amount of the scattered light intensity in the gas film disturbance layer, including: Perform an incident angle correlation analysis according to the corresponding relationship between the gas film thickness change amount and the local curvature of the gas film disturbance layer to obtain the angle sensitivity coefficient, and perform a local stress field correction on the angle sensitivity coefficient to obtain the angle mapping coefficient; Perform a hierarchical processing on the angle mapping coefficient in combination with the shear strain distribution of the gas film disturbance layer, calculate the shear response factor of the light incident angle for each layer, and obtain the optical coupling parameter according to the interaction between the shear response factor and the deformation amount; Perform a light excitation on the molecular motion degrees of freedom in the gas film disturbance layer according to the optical coupling parameter and the deformation amount to obtain the molecular motion state quantity inside the gas film, and combine the spatial distribution of the intermolecular force to obtain the molecular arrangement state quantity; Perform a transient response analysis based on the mapping relationship between the molecular arrangement state quantity and the molecular orientation distribution, calculate the local refractive index gradient coefficient, and obtain the refractive index distribution gradient coefficient according to the spatial distribution characteristics of the gradient coefficient; Perform a zonal scan on the gas film disturbance layer according to the refractive index distribution gradient coefficient to obtain the regional refractive index change amount, and perform a spatial superposition process on the regional refractive index change amount to obtain the refractive index distribution change amount; Perform a light intensity modulation analysis on the refractive index distribution change amount in combination with the spatial distribution of the light energy to obtain the light intensity distribution coefficient, and perform an energy conservation correction according to the light intensity distribution coefficient to obtain the scattered light intensity change amount.

[0009] Preferably, irradiate and excite the degrees of freedom of molecular motion in the gas film perturbation layer according to the optical coupling parameter and the deformation amount of the perturbation, obtain the molecular motion state quantity inside the gas film, and combine with the spatial distribution of the intermolecular force to obtain the molecular arrangement state quantity, including: Calculate the molecular excitation energy according to the product relationship between the optical coupling parameter and the deformation amount of the perturbation, obtain the molecular energy level transition coefficient, and analyze the molecular orientation distribution of the gas film molecules based on the molecular energy level transition coefficient to obtain the molecular polarization intensity; Perform gradient analysis on the molecular polarization intensity in the direction of the gas film thickness, obtain the polarization field distribution quantity, and perform interaction calculation according to the polarization field distribution quantity combined with the intermolecular van der Waals force to obtain the intermolecular interaction potential energy distribution; Perform statistical mechanics analysis on the intermolecular interaction potential energy distribution, obtain the molecular motion state quantity inside the gas film, and obtain the molecular arrangement state quantity according to the coupling relationship between the molecular motion state quantity and the local temperature field.

[0010] Preferably, configure the main light source optical path and the reference optical path based on the change amount of the refractive index distribution of the gas film perturbation layer, collect the interference phase image corresponding to the change amount of the scattered light intensity, and obtain the fringe offset sequence and the phase jump sequence of the interference phase image, including: Perform optical path difference analysis according to the spatial frequency distribution of the change amount of the refractive index distribution, obtain the optical path spectrum coefficient, and perform phase modulation response calculation on the optical path spectrum coefficient to obtain the modulation parameter of the main light source optical path; Perform light intensity compensation processing on the modulation parameter of the main light source optical path combined with the change amount of the scattered light intensity, calculate the balance factor of the reference optical path, and perform coherence optimization based on the balance factor to obtain the double optical path coupling coefficient; Construct an interference field according to the double optical path coupling coefficient and the change amount of the scattered light intensity, obtain the interference fringe intensity distribution quantity, and combine with the spatial gradient characteristics of the intensity distribution quantity to obtain the interference phase image; Extract the transient characteristics of the interference phase image, obtain the fringe sequence displacement quantity, and obtain the fringe offset sequence according to the cumulative change law of the fringe sequence displacement quantity; Perform phase continuity analysis according to the fringe offset sequence, obtain the coordinates of the phase mutation point, and perform periodic identification on the coordinates of the phase mutation point to obtain the phase jump period; Perform sequence reconstruction on the phase jump period combined with the phase distribution, obtain the phase jump amplitude, and obtain the phase jump sequence according to the time sequence change of the phase jump amplitude.

[0011] Preferably, the Fourier transform analysis is performed according to the stripe offset sequence and the phase jump sequence to extract the corresponding amplitude feature sequence and phase feature sequence, and the steady-state spectrum feature is obtained through ambient temperature correction, including: Perform cross-correlation analysis on the stripe offset sequence and the phase jump sequence to obtain the sequence correlation coefficient, and perform timing synchronization processing according to the correlation coefficient to obtain a synchronized sequence group; Perform short-time Fourier transform according to the synchronized sequence group to obtain a spectrum distribution matrix, and perform non-linear decomposition on the spectrum distribution matrix to obtain an amplitude feature sequence and a phase feature sequence; Perform temperature response analysis on the amplitude feature sequence and the phase feature sequence in combination with ambient temperature parameters to obtain a temperature correction factor, and perform spectrum compensation according to the temperature correction factor to obtain a temperature compensation sequence; Perform multi-scale filtering processing on the temperature compensation sequence to obtain steady-state spectrum components, and obtain steady-state spectrum features according to the distribution characteristics of the steady-state spectrum components.

[0012] Preferably, mapping the amplitude feature sequence and the phase feature sequence in the steady-state spectrum feature to a preset feature space, and calculating the edge thickness distribution of the molded lens, including: Perform feature decoupling on the amplitude feature sequence and the phase feature sequence in the steady-state spectrum feature based on the preset feature space to obtain the air film perturbation amount and the edge response amount, and obtain a physical space mapping factor according to the coupling relationship between the air film perturbation amount and the edge response amount; Perform dynamic response analysis on the physical space mapping factor and the lens rotation angle sequence to obtain an edge rotation compensation coefficient, and perform non-linear correction on the edge rotation compensation coefficient in combination with the air film interface stress distribution to obtain a dynamic mapping basis vector; Perform air film thickness distribution correlation analysis on the dynamic mapping basis vector to obtain interface tension distribution parameters, and perform morphological reconstruction on the air film microstructure according to the interface tension distribution parameters to obtain an edge topography vector; Perform air film flow field analysis according to the contour distribution of the edge topography vector to obtain an edge curvature correction amount, and perform local optimization on the edge curvature correction amount in combination with the air film shear stress to obtain a thickness correction coefficient; Perform multi-physical field coupling calculation on the thickness correction coefficient and the edge topography vector to obtain an edge thickness distribution sequence, and obtain the edge thickness distribution according to the edge thickness distribution sequence in combination with the steady-state response characteristics of the air film.

[0013] The second aspect of the present invention provides a vision-based molded lens edge thickness detection device, and the vision-based molded lens edge thickness detection device includes: The air film control module is used to introduce a visible gas into the edge area of the molded lens, control the flow rate, thickness, density and pressure of the visible gas, form an air film disturbance layer on the circumferential surface of the edge of the molded lens, and obtain the air film thickness change amount and disturbance deformation amount in the air film disturbance layer; The optical excitation module is used to select a corresponding light incident angle according to the air film thickness change amount in the air film disturbance layer, excite the disturbance deformation amount in the air film disturbance layer to generate a refractive index change, and obtain the refractive index distribution change amount and scattered light intensity change amount in the air film disturbance layer; The interference acquisition module is used to configure a main light source optical path and a reference optical path based on the refractive index distribution change amount of the air film disturbance layer, collect the interference phase image corresponding to the scattered light intensity change amount, and obtain the fringe shift sequence and phase jump sequence of the interference phase image; The spectrum analysis module is used to perform Fourier transform analysis according to the fringe shift sequence and phase jump sequence, extract the corresponding amplitude feature sequence and phase feature sequence, and obtain the steady-state spectrum feature through ambient temperature correction; The edge thickness calculation module is used to map the amplitude feature sequence and phase feature sequence in the steady-state spectrum feature to a preset feature space and calculate the edge thickness distribution of the molded lens.

[0014] In a third aspect of the present invention, a vision-based molded lens edge thickness detection device is provided, including: a memory and at least one processor, instructions are stored in the memory, and the memory and the at least one processor are interconnected by a line; the at least one processor calls the instructions in the memory so that the vision-based molded lens edge thickness detection device executes the steps of the above-mentioned vision-based molded lens edge thickness detection method.

[0015] In a fourth aspect of the present invention, a computer-readable storage medium is provided, instructions are stored in the computer-readable storage medium, and when it runs on a computer, it causes the computer to execute the steps of the above-mentioned vision-based molded lens edge thickness detection method.

[0016] The present invention proposes a vision-based molded lens edge thickness detection method, which cleverly uses an air film disturbance layer as a physical signal conversion bridge to solve the optical interference problem in the rotation detection of transparent lenses.

[0017] Specifically, a visible gas is introduced into the edge area of the molded lens and its flow field parameters are controlled to form an air film disturbance layer coupled with the edge thickness characteristics. This air film will produce local deformation and thickness change due to the different edge thicknesses of the lens, converting geometric characteristics into controllable flow field disturbance characteristics. When the lens rotates, the air film disturbance layer will always maintain dynamic coupling with the edge surface, avoiding the reflection and refraction interference encountered in direct optical imaging of transparent materials.

[0018] By applying angularly selective illumination to the air film perturbation layer, the deformation amount of the perturbation in the air film causes a local refractive index change, generating a stable optical response signal. This response does not depend on the transparency characteristics of the lens itself, but comes from the physical deformation of the air film perturbation layer, thus avoiding the interference of the lens material on the optical path. At the same time, the optical response of the air film shows regular changes with the variation of the edge thickness, providing a reliable basis for subsequent precise measurement.

[0019] After obtaining the optical response of the air film, a dual - optical - path interference system is used to collect phase information. The configuration of the main light source optical path and the reference optical path enables the interference fringes to accurately reflect the change in optical path difference caused by the air film perturbation, without being affected by light reflection and refraction during the rotation of the lens. By analyzing the fringe shift and phase jump in the interference phase image, the characteristic information directly related to the edge thickness can be extracted.

[0020] By performing Fourier transform analysis on the extracted characteristic sequence and combining with environmental temperature correction, stable spectral characteristics can be obtained. This process eliminates the interference caused by environmental fluctuations and ensures the stability of the measurement. Finally, through feature space mapping, a quantitative relationship between the spectral characteristics and the edge thickness is established, realizing the accurate calculation of the edge thickness.

[0021] This indirect measurement method using the air film perturbation as an intermediary converts the edge thickness of the transparent material, which is difficult to measure directly, into a reliable physical signal, and then conducts quantitative analysis through optical interference and signal processing means. The entire scheme forms a complete technical chain of air film perturbation, optical response, and signal processing, effectively solving the problem of inaccurate measurement caused by the transparency characteristics during the rotation detection of the molded lens. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following - described drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on the structures shown in these drawings.

[0023] Figure 1 It is a schematic diagram of an embodiment of the method for detecting the edge thickness of a molded lens based on vision in an embodiment of the present invention; Figure 2 It is a schematic diagram of an embodiment of the device for detecting the edge thickness of a molded lens based on vision in an embodiment of the present invention; Figure 3 It is a schematic diagram of an embodiment of the equipment for detecting the edge thickness of a molded lens based on vision in an embodiment of the present invention; Figure 4This is a schematic structural diagram of a partial structure of a vision-based edge thickness detection device for molded lenses in an embodiment of the present invention.

[0024] The realization of the purpose of the present invention, its functional characteristics and advantages will be further described with reference to the accompanying drawings in combination with the embodiments. Specific embodiments

[0025] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts fall within the scope of protection of the present invention.

[0026] It should be noted that if there are directional indications (such as up, down, left, right, front, back,...) involved in the embodiments of the present invention, the directional indications are only used to explain the relative positional relationship and movement conditions between components in a specific posture. If the specific posture changes, the directional indications will also change accordingly.

[0027] In addition, the descriptions involving "first", "second", etc. in the present invention are only for descriptive purposes, and cannot be understood as indicating or implying their relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of such features. In addition, "and / or" throughout the text includes three scenarios. Taking A and / or B as an example, it includes the technical solution of A, the technical solution of B, and the technical solution that both A and B are satisfied. In addition, the technical solutions between the various embodiments can be combined with each other, which must be based on the ability of those of ordinary skill in the art to implement. When the combination of technical solutions results in contradictions or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0028] An embodiment of the present application provides a vision-based method for detecting the edge thickness of molded lenses. This method is applied to a vision-based edge thickness detection device for molded lenses. As an example, Figure 4As shown in the figure, the vision-based edge thickness detection device for molded lenses at least includes a suction component 1, a light source component 2, and a camera component 3. The suction component 1 includes at least one suction head and a motor shaft drivingly connected to the suction head. The motor shaft is used to drive the suction head to rotate. A hollow inner cavity is formed inside the motor shaft, and the hollow inner cavity communicates with the suction head. The central control inner cavity also communicates with a negative pressure generating component. The negative pressure generating component operates to enable the suction head to adsorb the lens. The light source component 2 is used to generate a main light source optical path and a reference optical path, and the camera component 3 is used to acquire image data of the lens during rotation. The above various components are currently relatively mature product components, and the functional principles and specific structures of each component will not be described in detail one by one here.

[0029] Figure 1 FIG. is a flowchart of a vision-based edge thickness detection method for molded lenses provided by an embodiment of the present application. In this embodiment, the method includes: Please refer to Figure 1 , introduce a visible gas into the edge region of the molded lens, control the flow rate, thickness, density, and pressure of the visible gas, form a gas film disturbance layer on the circumferential surface of the edge of the molded lens, and obtain the change amount of the gas film thickness and the deformation amount of the disturbance in the gas film disturbance layer; In an embodiment of the present invention, the step of introducing a visible gas into the edge region of the molded lens, controlling the flow rate, thickness, density, and pressure of the visible gas, forming a gas film disturbance layer on the circumferential surface of the edge of the molded lens, and obtaining the change amount of the gas film thickness and the deformation amount of the disturbance in the gas film disturbance layer includes: Detect the linear velocity of the rotating molded lens, determine the initial flow rate of the visible gas according to the linear velocity, and calculate the velocity ratio of the flow rate of the visible gas to the linear velocity; Adjust the pressure coefficient of the visible gas according to the velocity ratio, and determine the density distribution parameter of the visible gas in combination with the ambient temperature; Calculate the gas film adhesion parameter based on the product of the density distribution parameter and the pressure coefficient, and control the introduction amount of the visible gas on the circumferential surface of the edge of the molded lens according to the gas film adhesion parameter; Conduct a uniformity analysis on the introduction amount of the visible gas to obtain the radial distribution coefficient and the circumferential distribution coefficient in the gas film disturbance layer; Calculate the change amount of the gas film thickness in the gas film disturbance layer according to the radial distribution coefficient and the circumferential distribution coefficient; Conduct a gradient analysis on the change amount of the gas film thickness to obtain the deformation amount of the disturbance in the gas film disturbance layer.

[0030] The following specifically describes the steps involved in the above embodiments: The linear velocity of the rotating molded lens can be detected. The rotational linear velocity of the lens edge can be monitored in real time by a laser Doppler velocimeter. The velocimeter emits a laser beam to illuminate the edge of the rotating lens, receives the reflected light signal, and calculates the linear velocity value by analyzing the frequency change. After obtaining the linear velocity, the initial flow rate of the visible gas (such as trace smoke or gas containing tracer particles) is determined according to the gas dynamic boundary layer theory. In specific operations, the initial flow rate is usually set to 70%-90% of the lens linear velocity to ensure that the gas can adhere to the lens surface without being completely taken away. Then the speed ratio of the visible gas flow rate to the lens linear velocity is calculated. For example, when the lens linear velocity is 2m / s, the initial gas flow rate can be set to 1.6m / s, and the speed ratio is 0.8. This step creates a dynamically balanced air film layer by precisely controlling the proportional relationship between the gas flow rate and the linear velocity of the lens, so that the gas can both rotate with the lens to form a stable air film and maintain a certain relative motion to produce a sensitive disturbance response, thereby transforming the edge characteristics of the transparent lens that were originally difficult to directly observe into measurable aerodynamic characteristics.

[0031] When adjusting the pressure coefficient of the visible gas according to the speed ratio, a gas supply system consisting of a pressure regulating valve and a precision flow controller is used. The pressure coefficient refers to the ratio of the current working pressure to the standard reference pressure, which is monitored in real time by a pressure sensor. The specific adjustment method is: when the speed ratio is low (such as around 0.7), increase the pressure coefficient to 1.3-1.5; when the speed ratio is in the middle (such as around 0.8), set the pressure coefficient to 1.1-1.3; when the speed ratio is high (such as around 0.9), reduce the pressure coefficient to 0.9-1.1. At the same time, a temperature sensor is used to measure the ambient temperature, and the density distribution parameters are calculated according to the ideal gas state equation. For example, when the ambient temperature is 25°C, if the speed ratio is 0.8 and the pressure coefficient is set to 1.2, the density distribution parameters are approximately 1.15kg / , showing a distribution characteristic that gradually decreases from the lens surface to the outside. This step achieves precise control of the characteristics of the air film disturbance layer by establishing a correlation control mechanism between the velocity ratio, pressure coefficient and ambient temperature, maximizing the response sensitivity of the air film to small changes in the edge of the lens, while compensating for the impact of ambient temperature changes on gas density, ensuring the stability and repeatability of the measurement results.

[0032] The air film attachment parameter is calculated based on the product of the density distribution parameter and the pressure coefficient. The air film attachment parameter is a dimensionless parameter that characterizes the interaction strength between the gas and the solid surface. The relationship curve between the air film attachment parameter and the required gas flow rate is established through the fluid mechanics numerical analysis method. The precise gas distribution array system is used to control the gas flow rate according to the calculated air film attachment parameter. The specific implementation method is to control the solenoid valve switching time through a pulse width modulation signal. For example, when the density distribution parameter is 1.15kg / When the pressure coefficient is 1.2, the air film attachment parameter is about 1.38. Accordingly, the gas flow rate is controlled within the range of 5-10 ml / s, and the gas is evenly introduced into the peripheral surface of the lens edge through multiple micro nozzles distributed in a ring array. This step realizes the accurate mapping from theoretical model to actual operation by converting fluid mechanics parameters (density distribution, pressure coefficient) into actual control parameters (gas flow rate), ensuring the controllability and uniformity of the air film disturbance layer formation process, and creating an ideal physical medium for capturing tiny thickness changes at the edge of the lens.

[0033] When analyzing the uniformity of the visible gas intake, a high-speed camera combined with scattered light imaging technology is used to capture real-time images of the gas film distribution. The gas film is partitioned and analyzed using a digital image processing algorithm, and the gas velocity gradient data is extracted radially to calculate the radial velocity change rate. The radial airflow modulation coefficient is calculated in combination with the surface curvature data of the molded lens (usually obtained from design data or pre-measurement). The radial airflow modulation coefficient refers to the degree of influence of the lens surface geometry on the airflow distribution, and the value is usually between 0.7-1.0. For example, for a lens with an edge curvature radius of 5mm, when the radial velocity change rate is 0.2 / mm, the radial airflow modulation coefficient is approximately 0.85. The circumferential airflow distribution is then analyzed based on the radial airflow modulation coefficient, and the circumferential distribution coefficient is obtained in combination with the Coriolis force correction (the influence of the inertial force generated by the rotation of the lens on the airflow). This step establishes the spatial mapping relationship between the air film disturbance layer and the geometric characteristics of the lens by introducing air flow distribution analysis in radial and circumferential dimensions, so that the air film distribution can accurately reflect the geometric characteristics of the lens edge. At the same time, the influence of the Coriolis force in the rotating system is taken into account, thereby improving the adaptability and characterization accuracy of the air film to the lens edge shape.

[0034] When calculating the change in the thickness of the air film in the air film disturbance layer according to the radial distribution coefficient and the circumferential distribution coefficient, the computational fluid dynamics method is used for numerical calculation. The specific implementation process is: first, the radial distribution coefficient and the circumferential distribution coefficient are input as the boundary conditions of the air flow distribution, and then the finite difference method is used to numerically solve the fluid continuity equation to obtain the value of the air film thickness at discrete points in space. The discrete points are connected into a continuous air film thickness distribution map using the cubic spline interpolation algorithm. The calculation results are experimentally verified by a high-precision laser interferometer to ensure that the calculation accuracy reaches the micron level. For example, when the radial distribution coefficient is 0.85 and the circumferential distribution coefficient is 0.92, the change in the thickness of the air film is in the range of 5-20 microns, and it presents a distribution feature corresponding to the change in the thickness of the lens edge. This step converts the edge characteristics of the transparent lens that were originally difficult to directly observe into an air film distribution with obvious optical characteristics by accurately calculating the air film thickness distribution, realizing the conversion from invisible to visible, and providing an innovative idea for solving the problem of edge detection of transparent materials.

[0035] When performing gradient analysis on the change in film thickness, the numerical differentiation method is used to calculate the first-order and second-order derivative distribution of the film thickness in the spatial coordinate system. By analyzing the amplitude and direction changes of the gradient distribution, the disturbance deformation in the film disturbance layer is extracted. The disturbance deformation is a physical quantity that describes the local deformation degree and distribution characteristics of the film, and the unit is inverse length ( / mm). For example, when the film thickness changes from 10 microns to 15 microns in a certain area, and the change distance is 0.5mm, the disturbance deformation in this area is approximately / mm. The spatial distribution of the disturbance deformation variable usually shows a characteristic pattern corresponding to the change in lens edge thickness. The technical consideration of this step is: by converting the absolute change in air film thickness into a relative change rate through gradient analysis, the system's detection sensitivity to tiny changes in lens edge thickness is significantly enhanced, and sub-micron thickness changes are effectively captured. At the same time, physical characteristic variables suitable for subsequent optical interferometry are provided, laying the foundation for optical detection.

[0036] In one embodiment of the present invention, the uniformity analysis of the amount of visible gas introduced to obtain the radial distribution coefficient and the circumferential distribution coefficient in the air film disturbance layer includes: Performing a flow velocity gradient analysis on the amount of the visible gas introduced in the radial direction to obtain a radial flow velocity change rate, and obtaining a radial airflow modulation coefficient according to a corresponding relationship between the radial flow velocity change rate and a surface curvature of the molded lens; Performing circumferential airflow distribution analysis according to the radial airflow modulation coefficient to obtain a circumferential flow compensation amount, and performing Coriolis force correction on the circumferential flow compensation amount in combination with the rotation speed of the molded lens to obtain a circumferential distribution coefficient; Perform airflow stability calculations on the circumferential distribution coefficient along the edge profile of the molded lens to obtain a stability criterion value, and correct the radial airflow modulation coefficient according to the stability criterion value to obtain a radial distribution coefficient.

[0037] The following specifically describes the steps involved in the above embodiments: When performing a flow velocity gradient analysis on the inflow rate of the visible gas along the radial direction, a hot-wire anemometer array is used to perform multi-point measurements of the flow velocities at different radial positions within the gas film disturbance layer. The hot-wire anemometer reflects the airflow velocity by measuring the cooling rate of the thermosensitive element and can achieve precise measurement of tiny flow velocities. The measured radial flow velocity data is subjected to numerical differentiation processing to calculate the flow velocity change rate at each radial point, that is, the first derivative of the flow velocity along the radial direction. Then, surface curvature distribution information is extracted from the design data of the molded lens, and a corresponding relationship diagram between the radial flow velocity change rate and the surface curvature is established. By fitting the curve using the least squares method, a radial airflow modulation coefficient is obtained, which reflects the modulation degree of the lens surface geometry on the airflow distribution. For example, for a typical molded lens with an edge curvature radius of 5 mm, when the radial flow velocity change rate is 0.2 / mm, the radial airflow modulation coefficient is approximately 0.85. This step enables the airflow distribution to be an indirect representation of the lens surface shape by establishing a corresponding relationship between the airflow characteristics and the lens geometric characteristics, converting the geometric characteristics of the transparent material that are difficult to directly measure into measurable aerodynamic characteristics, and significantly improving the reliability and accuracy of the edge detection of the transparent lens.

[0038] When performing circumferential airflow distribution analysis based on the radial airflow modulation coefficient, 16 - 24 pressure sensors arranged in a ring are used to synchronously monitor the airflow pressures at each circumferential point on the lens edge. The circumferential flow rate distribution is calculated through the pressure distribution, and its non-uniformity is analyzed to obtain a circumferential flow rate compensation amount. The circumferential flow rate compensation amount refers to the gas flow rate that needs to be increased or decreased at each circumferential point to make the gas film distribution uniform. Subsequently, the Coriolis force influence coefficient is calculated based on the rotation speed of the lens. The Coriolis force is an inertial force generated due to rotational motion in a rotating system. The circumferential flow rate compensation amount and the Coriolis force influence coefficient are subjected to a weighted fusion operation to obtain a circumferential distribution coefficient. For example, when the lens rotation speed is 300 rpm, the Coriolis force influence coefficient is approximately 1.2, and when the circumferential flow rate compensation amount is ±5%, the calculated circumferential distribution coefficient is approximately 0.88 - 0.96. This step solves the problem of uneven gas film distribution in the rotating state by introducing a Coriolis force correction mechanism, enabling the gas film to maintain a stable distribution characteristic in a dynamic environment and improving the adaptability and measurement accuracy of the detection system for rotating lenses.

[0039] When calculating the airflow stability along the edge contour of the molded lens for the circumferential distribution coefficient, a computational fluid dynamics software is used to perform a time-domain simulation of the air film flow state. Based on the Navier-Stokes equations and turbulence models, the fluctuation characteristics of the airflow over time are analyzed, and dimensionless parameters such as the Reynolds number and Weber number are calculated. Through these parameters, the airflow stability criterion value is determined. The stability criterion value is a dimensionless parameter characterizing the anti-disturbance ability of the air film, and a higher value indicates better air film stability. Subsequently, the stability criterion value is compared with a preset threshold, and the radial airflow modulation coefficient is corrected according to the comparison result to obtain the final radial distribution coefficient. For example, when the original radial airflow modulation coefficient is 0.85 and the stability criterion value is 0.92 (higher than the threshold of 0.9), the corrected radial distribution coefficient is approximately 0.88. This step ensures the stability and reliability of the air film disturbance layer in a dynamic environment by introducing an airflow stability analysis and feedback correction mechanism, enabling the air film characteristics to accurately and persistently reflect the edge thickness distribution of the lens, providing a stable physical medium for subsequent optical interference measurements, and improving the robustness and repeatability of the entire measurement system.

[0040] Please continue to refer to Figure 1 , select a corresponding light incident angle according to the change amount of the air film thickness in the air film disturbance layer, excite the refractive index change generated by the deformation amount of the disturbance in the air film disturbance layer, and obtain the change amount of the refractive index distribution and the change amount of the scattered light intensity of the air film disturbance layer; In an embodiment of the present invention, the step of selecting a corresponding light incident angle according to the change amount of the air film thickness in the air film disturbance layer, exciting the refractive index change generated by the deformation amount of the disturbance in the air film disturbance layer, and obtaining the change amount of the refractive index distribution and the change amount of the scattered light intensity of the air film disturbance layer includes: Conduct an incident angle correlation analysis according to the corresponding relationship between the air film thickness change amount and the local curvature of the air film disturbance layer to obtain an angle sensitivity coefficient, and perform a local stress field correction on the angle sensitivity coefficient to obtain an angle mapping coefficient; Perform a hierarchical processing on the angle mapping coefficient in combination with the shear strain distribution of the air film disturbance layer, calculate the shear response factor of the light incident angle for each layer, and obtain an optical coupling parameter according to the interaction between the shear response factor and the deformation amount of the disturbance; Use light to excite the degree of freedom of molecular motion in the air film disturbance layer according to the optical coupling parameter and the deformation amount of the disturbance, obtain the molecular motion state quantity inside the air film, and combine the spatial distribution of intermolecular forces to obtain the molecular arrangement state quantity; Perform a transient response analysis based on the mapping relationship between the molecular arrangement state quantity and the molecular orientation distribution, calculate the local refractive index gradient coefficient, and obtain the refractive index distribution gradient coefficient according to the spatial distribution characteristics of the gradient coefficient; Perform zonal scanning on the gas film perturbation layer according to the refractive index distribution gradient coefficient, obtain the zonal refractive index change amount, and perform spatial superposition processing on the zonal refractive index change amount to obtain the refractive index distribution change amount; Perform light intensity modulation analysis on the refractive index distribution change amount in combination with the spatial distribution of light energy, obtain the light intensity distribution coefficient, and perform energy conservation correction according to the light intensity distribution coefficient to obtain the scattered light intensity change amount.

[0041] The following specifically describes the steps involved in the above embodiments: When performing incident angle correlation analysis according to the correspondence between the gas film thickness change amount and the local curvature of the gas film perturbation layer, first obtain the local curvature distribution map of the gas film perturbation layer through a confocal microscopy measurement system. This system obtains the three-dimensional morphology of the gas film surface by laser scanning different focal plane positions. Perform correlation analysis on the gas film thickness change amount and the local curvature data, and use the least squares method to fit the corresponding relationship curve between them. Based on this relationship curve, calculate the optimal light incident angle in different curvature regions to obtain the angle sensitivity coefficient, which characterizes the influence degree of the change in the illumination angle on the refractive index sensitivity. For example, when the local curvature of the gas film is 0.5 and the thickness change amount is 8 μm, the corresponding angle sensitivity coefficient is approximately 0.72, indicating that the optimal incident angle here is approximately 43 degrees. Subsequently, use a photoelastic analysis system to measure the stress distribution inside the gas film, perform joint analysis on the stress data and the angle sensitivity coefficient, and obtain the angle mapping coefficient through a weighted correction algorithm. This step provides an optimization method for the key parameters of the optical measurement system by establishing an accurate correspondence between the optical incident angle and the gas film characteristics, enabling the system to adaptively adjust the measurement parameters according to the gas film characteristics in different regions, and greatly improving the measurement sensitivity.

[0042] When performing hierarchical processing on the angle mapping coefficient in combination with the shear strain distribution of the gas film perturbation layer, perform multi-layer scanning on the gas film using a high-precision laser light sectioning method. This method cuts the gas film at different heights by a laser sheet light source, and cooperates with a high-speed camera to capture scattered images to obtain the air flow velocity distribution map of each layer, and calculate the shear strain distribution. Shear strain refers to the deformation rate caused by the relative velocity difference between gas layers, with the unit of The gas film is equally divided into 8 - 12 layers along the thickness direction. The angular mapping coefficient obtained in the previous step is applied to each layer to calculate the shear response factor at the optimal light incident angle for each layer. The shear response factor characterizes the sensitivity of the influence of shear strain on optical properties, and its numerical range is usually between 0.6 - 1.2. The shear response factors of each layer are multiplied by the disturbance deformation amount of the corresponding layer to obtain the optical coupling parameter reflecting the overall optical response characteristics of the gas film. This step accurately captures the three-dimensional structural characteristics inside the gas film through the layered processing method, breaking through the limitation that traditional optical measurements can only obtain surface information and achieving the accurate characterization of the internal structure of the gas film.

[0043] When optically exciting the molecular motion degrees of freedom in the disturbed layer of the gas film according to the optical coupling parameter and the disturbance deformation amount, an experimental operation is carried out using a polarization spectroscopy analysis system. This system directionally excites the gas film molecules with polarized light to change the vibration and rotation states of the molecules. According to the optical coupling parameter obtained in the previous step, the optimal light source wavelength and polarization angle are calculated to maximize the excitation efficiency. Light with a specific wavelength and polarization direction is applied to the gas film, and the absorption spectrum and emission spectrum of the molecules are measured to obtain the molecular motion state quantity inside the gas film, which characterizes the excitation degree of the vibration and rotation degrees of freedom of the molecules. Combining with the molecular dynamics theory, the spatial distribution of the van der Waals force between molecules in the gas film is analyzed, and the intermolecular interaction potential field is calculated. The molecular motion state quantity is combined with the interaction potential field to calculate the molecular arrangement state quantity, which reflects the orientation and arrangement law of the molecules inside the gas film. This step realizes the accurate mapping from macroscopic air flow characteristics to microscopic molecular arrangement by precisely controlling the interaction between the light field and the gas film molecules, providing a microscopic physical mechanism explanation for the subsequent refractive index change.

[0044] When performing transient response analysis based on the mapping relationship between the molecular arrangement state quantity and the molecular orientation distribution, a time-resolved spectroscopy technique is used to measure the change in the optical properties of the gas film before and after illumination. This technique uses femtosecond laser pulses for excitation and is combined with a time-resolved detection system to capture the dynamic process of molecular orientation change. Through Fourier transform spectroscopy analysis, the change in the polarizability of molecules at different spatial positions is calculated to obtain the local refractive index gradient coefficient. This coefficient characterizes the sensitivity of the local refractive index to the change in molecular arrangement, and its value is usually between and . The spatial distribution of the gradient coefficient over the entire gas film region is analyzed, and wavelet transform is used to extract the characteristic frequency to obtain the refractive index distribution gradient coefficient. This step realizes the accurate mapping from the molecular scale to the macroscopic optical properties through the transient response analysis method, revealing the dynamic mechanism of the refractive index change in the disturbed layer of the gas film and laying a theoretical foundation for achieving high-sensitivity optical measurement.

[0045] When performing zonal scanning on the gas film perturbation layer according to the refractive index distribution gradient coefficient, a tunable laser interferometer is used to scan and measure different regions of the gas film. This instrument optimizes the sensitivity to refractive index changes in each region by adjusting the laser wavelength and incident angle. The gas film perturbation layer is evenly divided into 12 - 18 regions along the circumferential direction, and the corresponding refractive index distribution gradient coefficient is applied to each region. The phase-sensitive detection technique is adopted to measure the refractive index change amount in each region. The measurement results of each region are connected through a spatial interpolation algorithm, and the spatial superposition process is carried out using the Bessel surface fitting method to obtain the refractive index distribution change amount covering the entire gas film. For example, in the region where the edge thickness of the lens changes by 5μm, the corresponding refractive index change amount is approximately 3× . This step realizes the high-precision full-coverage measurement of the refractive index distribution at the edge of the complex-curved lens through zonal scanning and spatial splicing techniques, and solves the technical problem that traditional measurement methods are difficult to adapt to complex curves.

[0046] When performing intensity modulation analysis on the refractive index distribution change amount in combination with the spatial distribution of light energy, a light intensity distribution measurement system is used to measure the intensity distributions of incident light and scattered light. This system consists of a high-precision CCD camera and light intensity analysis software, and can capture tiny light intensity changes. The light scattering efficiency at each point is calculated according to the refractive index distribution change amount. Combining with the energy distribution diagram of the incident light, the theoretical distribution of the scattered light is calculated through the light transmission equation. The theoretical calculation result is compared with the actual measurement result, and the light intensity distribution coefficient is calculated through an iterative optimization algorithm. This coefficient reflects the actual light scattering ability of each region. Based on the principle of energy conservation, the light intensity distribution coefficient is corrected to eliminate system errors and environmental interference, and finally the accurate scattered light intensity change amount is obtained. This step establishes an accurate correspondence relationship between the refractive index change and the scattered light intensity by combining optical theory with actual measurement results, provides a directly measurable physical quantity for quantitatively analyzing the edge thickness change of the lens, and effectively solves the low-contrast problem in the edge detection of transparent materials.

[0047] In an embodiment of the present invention, the illumination excitation of the molecular motion degrees of freedom in the gas film perturbation layer according to the optical coupling parameter and the perturbation deformation amount, obtaining the internal molecular motion state quantity of the gas film, and combining with the spatial distribution of intermolecular forces to obtain the molecular arrangement state quantity includes: Calculating the molecular excitation energy according to the product relationship between the optical coupling parameter and the perturbation deformation amount, obtaining the molecular energy level transition coefficient, and analyzing the molecular orientation distribution of the gas film based on the molecular energy level transition coefficient to obtain the molecular polarization intensity; Performing gradient analysis on the molecular polarization intensity in the direction of the gas film thickness to obtain the polarization field distribution quantity, and performing interaction calculation according to the polarization field distribution quantity in combination with the intermolecular van der Waals force to obtain the intermolecular interaction potential energy distribution; Perform a statistical mechanics analysis on the intermolecular interaction potential energy distribution to obtain the molecular motion state quantities inside the gas film, and based on the coupling relationship between the molecular motion state quantities and the local temperature field, obtain the molecular arrangement state quantities.

[0048] The following specifically describes the steps involved in the above embodiments: When calculating the molecular excitation energy according to the product relationship between the optical coupling parameter and the perturbation deformation quantity, a high-precision spectroscopic analyzer is used to measure the absorption spectrum of the molecules in the gas film. The specific operation is to multiply the optical coupling parameter obtained in the previous step by the perturbation deformation quantity to obtain the molecular excitation energy distribution at different positions. The molecular excitation energy is the energy required to transition a molecule from the ground state to the excited state, with the unit of electron volt (eV). By comparing the spectral absorption databases of different gas molecules, the molecular energy level transition coefficient is calculated, which characterizes the response degree of the molecule to the energy of light with a specific wavelength. For example, when the optical coupling parameter is 0.85 and the perturbation deformation quantity is 0.02 / mm, the calculated molecular excitation energy is approximately 0.15 eV, and the corresponding energy level transition coefficient is 0.72. Subsequently, a polarized light scattering measurement system is used to analyze the orientation distribution of the gas film molecules in the excited state and calculate the molecular polarization intensity. When the energy level transition coefficient is in the range of 0.7 - 0.8, the molecular polarization intensity increases significantly, indicating that the sensitivity of the gas film to optical detection is the highest within this range. This step realizes the precise mapping from macroscopic airflow parameters to microscopic molecular characteristics by introducing the principles of quantum optics to analyze the interaction between gas molecules and the light field, enabling the transparent gas to produce a measurable optical response under specific illumination conditions.

[0049] When performing a gradient analysis of the molecular polarization intensity along the thickness direction of the gas film, a tomography technique is used to detect the optical response at different depths of the gas film. This technique scans the gas film using multi-angle light paths to obtain imaging data of different depth planes. The gas film thickness direction is evenly divided into 8 - 10 layers, and numerical differentiation is performed on the molecular polarization intensity data of each layer to calculate the spatial gradient of the polarization intensity and obtain the polarization field distribution quantity. The polarization field distribution quantity characterizes the spatial distribution characteristics of the molecular polarization state inside the gas film, and the numerical range is usually between 0.5 - 2.0. Subsequently, combined with the molecular dynamics theory, the action characteristics of the van der Waals force between the molecules in the gas film are analyzed. The van der Waals force is a weak intermolecular force that decays in an inverse sixth-power relationship with distance. The interaction energy between molecules at different positions is calculated through the position correlation function to obtain the intermolecular interaction potential energy distribution. Under typical measurement conditions, when the polarization field distribution quantity is 1.2 and the average intermolecular distance is approximately 5 nanometers, the calculated interaction potential energy is approximately 0.05 eV. This step reveals the spatial organization law of gas molecules under the action of the light field by precisely analyzing the intermolecular interaction mechanism inside the gas film, providing a microscopic physical basis for explaining the modulation mechanism of the gas film on optical signals.

[0050] When performing a statistical mechanics analysis of the intermolecular interaction potential energy distribution, the Boltzmann distribution principle is applied to calculate the energy distribution probability of the molecular system. The Monte Carlo algorithm is used to perform numerical simulations on the molecular system to calculate the position and velocity distributions of the molecules under given temperature and interaction potential energy conditions. The molecular motion state quantity inside the gas film is calculated from these distribution data, and this physical quantity describes the dynamic characteristics of the molecular ensemble in the thermodynamic equilibrium state. The molecular motion state quantity is closely related to the system temperature. When the temperature increases, the molecular motion intensifies and the value of the state quantity increases. A thermal imaging system is used to measure the temperature distribution at different positions inside the gas film to analyze the correlation between the molecular motion state quantity and the local temperature field. For example, when the local temperature of the gas film changes from 22 °C to 25 °C, the molecular motion state quantity increases by approximately 15%. Based on this coupling relationship and combined with the intermolecular interaction potential energy distribution obtained previously, the molecular arrangement state quantity is calculated, and this quantity reflects the spatial arrangement law of the molecules in the gas film. In the region with uneven edge thickness of the molded lens, the molecular arrangement state quantity shows an obvious change, forming a corresponding relationship with the lens thickness distribution. This step converts the microscopic molecular interactions into macroscopic optical properties through statistical mechanics methods, establishes an internal connection between the optical properties of the gas film and the edge thickness of the lens, and realizes the indirect accurate measurement of the edge thickness of the transparent lens.

[0051] Please continue to refer to Figure 1 , based on the refractive index distribution change amount of the gas film perturbation layer, configure the main light source optical path and the reference optical path, collect the interference phase image corresponding to the scattered light intensity change amount, and obtain the fringe offset sequence and the phase jump sequence of the interference phase image; In an embodiment of the present invention, the configuring the main light source optical path and the reference optical path based on the refractive index distribution change amount of the gas film perturbation layer, collecting the interference phase image corresponding to the scattered light intensity change amount, and obtaining the fringe offset sequence and the phase jump sequence of the interference phase image includes: Perform an optical path difference analysis according to the spatial frequency distribution of the refractive index distribution change amount to obtain the optical path spectrum coefficient, and perform a phase modulation response calculation on the optical path spectrum coefficient to obtain the modulation parameter of the main light source optical path; Perform an optical intensity compensation process on the modulation parameter of the main light source optical path combined with the scattered light intensity change amount, calculate the balance factor of the reference optical path, and perform coherence optimization based on the balance factor to obtain the double optical path coupling coefficient; Construct an interference field according to the double optical path coupling coefficient and the scattered light intensity change amount to obtain the interference fringe intensity distribution amount, and combine the spatial gradient characteristics of the intensity distribution amount to obtain the interference phase image; Extract the transient characteristics of the interference phase image to obtain the fringe sequence displacement amount, and obtain the fringe offset sequence according to the cumulative change law of the fringe sequence displacement amount; Performing phase continuity analysis according to the fringe offset sequence to obtain the coordinates of the phase mutation point, and performing periodic identification on the coordinates of the phase mutation point to obtain the phase jump period; The phase jump period is combined with the phase distribution to perform sequence reconstruction to obtain the phase jump amplitude, and a phase jump sequence is obtained according to the time sequence change of the phase jump amplitude.

[0052] The following is a detailed description of the steps involved in the above embodiment: When performing optical path difference analysis based on the spatial frequency distribution of the refractive index distribution variation, a spectral analysis system is used to perform spatial Fourier transform on the refractive index distribution of the air film disturbance layer. The system consists of a high-precision optical detector and a digital signal processing unit, which can accurately obtain the spatial frequency characteristics of the refractive index distribution. The refractive index distribution variation data is subjected to a two-dimensional fast Fourier transform to obtain the spectrum distribution in the spatial frequency domain. By statistically analyzing the spectrum distribution, the amplitude and phase of the main frequency components are calculated to obtain the optical path spectrum coefficient. The optical path spectrum coefficient is a parameter that characterizes the law of optical path difference variation when light passes through the air film at different spatial positions, and the numerical range is usually between 0.3-0.8. The optical path spectrum coefficient is analyzed using phase modulation theory, and the phase response characteristics corresponding to different optical path differences are calculated to obtain the modulation parameters of the main light source optical path. For example, when the main spatial frequency of the refractive index distribution variation is 0.5 / mm and the corresponding optical path spectrum coefficient is 0.65, the calculated modulation parameter of the main light source optical path is 0.78. This step determines the light source parameters suitable for the current state of the gas film by converting the refractive index distribution into optical path spectrum characteristics, so that the light source can respond most sensitively to tiny refractive index changes in the gas film.

[0053] When the modulation parameters of the main light source optical path are combined with the change in scattered light intensity for light intensity compensation, a light intensity measuring instrument is used to monitor the light intensity distribution of the main light path in real time. According to the modulation parameters of the main light source optical path, the wavelength, power and wavefront shape of the light source are adjusted to match the current air film characteristics. At the same time, the distribution characteristics of the change in scattered light intensity in space are measured, and the intensity ratio of the main light path and the scattered light is calculated. The balance factor of the reference light path is calculated through the light energy balance algorithm. The balance factor is a parameter used to adjust the light intensity and phase of the reference light path to match the main light path, and the value is usually in the range of 0.6-1.4. For example, when the modulation parameters of the main light source optical path are 0.78 and the average value of the change in scattered light intensity is 0.25, the calculated reference light path balance factor is about 1.15. On this basis, by adjusting the coherence length and polarization state of the reference light path, the coherence of the dual light paths is optimized, and the dual light path coupling coefficient is obtained, which characterizes the light field superposition efficiency of the two light paths in the interference area. This step achieves high-sensitivity detection of tiny thickness changes at the edge of the transparent lens by precisely controlling the parameter matching of the main light path and the reference light path, solving the problem of low contrast in transparent material detection in traditional methods.

[0054] When constructing the interference field according to the dual-path coupling coefficient and the change in scattered light intensity, a Michelson interferometer or a Mach-Zehnder interferometer is used to set up the optical path system. According to the dual-path coupling coefficient obtained in the previous step, the optical path difference and the splitting ratio of the interferometer are adjusted so that the main optical path and the reference optical path form a stable interference field in the interference area. The interference fringe image is captured with the help of a high-resolution CCD camera, and the fringe contrast is improved by image enhancement processing. The interference fringes are modulated by the change in the intensity of the scattered light, and the intensity distribution of the interference fringes is calculated, which describes the spatial characteristics of the light and dark distribution of the interference fringes. The intensity distribution of the interference fringes is numerically gradient processed, the spatial gradient features are extracted, and the phase recovery algorithm is applied to convert it into an interference phase image. For example, when the dual-path coupling coefficient is 0.92 and the change in the scattered light intensity is 0.3, the contrast of the interference fringes formed is about 0.85, and the clarity of the corresponding interference phase image reaches more than 95%. This step converts the change in the refractive index of the air film into directly observable interference fringes by constructing a high-quality interference field, realizing the conversion from invisible physical quantities to visible images, greatly improving the intuitiveness and accuracy of the detection.

[0055] When extracting transient features from the interference phase image, a high-speed digital image processing system is used to analyze the continuously collected interference image sequence. The interference phase images at different times are aligned using an image registration algorithm to ensure the temporal continuity of the data. The fringe tracking algorithm is used to identify the position and morphology of the fringes, calculate the change in the position of the fringes at adjacent times, and obtain the displacement of the fringe sequence. The displacement of the fringe sequence characterizes the motion characteristics of the interference fringes over time, and the unit is pixel or actual physical distance. By analyzing the cumulative change law of the fringe displacement during the rotation of the lens, the systematic change component is extracted to obtain the fringe offset sequence. For example, when the lens rotates at a speed of 300rpm, in the area where the edge thickness changes by 5μm, the periodic change amplitude of the fringe offset sequence is about 10-15 pixels. This step realizes the continuous measurement of the edge thickness distribution of the rotating lens by dynamically tracking the change in the position of the interference fringes, overcoming the problem of difficult measurement of the rotation process in the traditional method.

[0056] When performing phase continuity analysis based on the fringe offset sequence, the phase unwrapping algorithm is used to process the fringe displacement data. The phase continuity criterion is used to detect the discontinuous points in the fringe offset sequence, calculate the position where the phase gradient is greater than the threshold, and determine the coordinates of the phase mutation point. The phase mutation point refers to the position where the phase value jumps, and these points usually correspond to the mutation area of ​​the edge thickness of the lens. Cluster analysis is performed on the coordinates of the phase mutation point to extract its spatial and temporal distribution laws. The periodic analysis algorithm is applied to identify the periodic characteristics of the mutation points and obtain the phase jump period. The phase jump period reflects the periodicity of the edge thickness change during the rotation of the lens and is closely related to the structural characteristics of the lens. For example, for some non-uniformly designed optical lenses, the phase jump period may correspond to the symmetry of the lens, such as 4, 6 or 8 periods. This step accurately captures the slight changes in the edge thickness of the lens by analyzing the phase mutation characteristics, and achieves high-sensitivity detection of thickness non-uniformity.

[0057] When performing sequence reconstruction on the phase jump period combined with the phase distribution, the phase synthesis technique is used to process the phase data. According to the phase jump period obtained in the previous step, the phase data of the entire rotation period is segmented and analyzed to calculate the characteristic parameters of the phase change within each period. The phase reconstruction algorithm is used to connect the segmented phase data to eliminate the \(2\pi\) ambiguity caused by phase wrapping and obtain the phase jump amplitude. The phase jump amplitude is a physical quantity that describes the magnitude of the phase change at the jump point and is usually measured in radians. Combining the angle information of the lens rotation, the variation law of the phase jump amplitude with the rotation angle is analyzed to construct a complete phase jump sequence. For example, when the edge thickness of the lens changes from 3 to 8 \(\mu m\), the corresponding phase jump amplitude is approximately 0.5 to 1.2 radians, forming a specific sequence pattern in the complete rotation period. This step establishes the mapping relationship from the interference phase to the lens edge thickness by accurately extracting the phase jump characteristics and performing time series reconstruction, providing an accurate data basis for quantitatively analyzing the lens edge thickness distribution.

[0058] Please continue to refer to Figure 1 , perform Fourier transform analysis according to the fringe offset sequence and the phase jump sequence, extract the corresponding amplitude characteristic sequence and phase characteristic sequence, and obtain the steady-state spectrum characteristics through ambient temperature correction; In an embodiment of the present invention, the performing Fourier transform analysis according to the fringe offset sequence and the phase jump sequence, extracting the corresponding amplitude characteristic sequence and phase characteristic sequence, and obtaining the steady-state spectrum characteristics through ambient temperature correction includes: Perform cross-correlation analysis on the fringe offset sequence and the phase jump sequence to obtain the sequence correlation coefficient, and perform time series synchronization processing according to the correlation coefficient to obtain a synchronized sequence group; Perform short-time Fourier transform according to the synchronized sequence group to obtain a spectrum distribution matrix, and perform non-linear decomposition on the spectrum distribution matrix to obtain an amplitude characteristic sequence and a phase characteristic sequence; Perform temperature response analysis on the amplitude characteristic sequence and the phase characteristic sequence combined with the ambient temperature parameter to obtain a temperature correction factor, and perform spectrum compensation according to the temperature correction factor to obtain a temperature compensation sequence; Perform multi-scale filtering processing on the temperature compensation sequence to obtain steady-state spectrum components, and obtain the steady-state spectrum characteristics according to the distribution characteristics of the steady-state spectrum components.

[0059] The following specifically describes the steps involved in the above embodiments: When performing cross - correlation analysis on the fringe offset sequence and the phase jump sequence, a digital signal processing system is used to calculate the correlation between the two sequences. The specific implementation process is as follows: First, data pre - processing is performed on the fringe offset sequence and the phase jump sequence, including denoising, normalization, and resampling, so that the two sequences have the same number of data points and sampling rate. Then, the cross - correlation function is used to calculate the correlation coefficients of the two sequences at different time delays, and the time delay corresponding to the maximum correlation is found. The cross - correlation function is realized by sliding one sequence and calculating the sum of the point products with the other sequence. The obtained sequence correlation coefficient is a value between 0 and 1, which characterizes the similarity degree and time synchronization of the two sequences. For example, when the fringe offset sequence and the phase jump sequence come from the same lens edge region, the correlation coefficient is usually between 0.75 and 0.90; while for uncorrelated regions, the correlation coefficient is less than 0.3. The two sequences are calibrated on the time axis according to the sequence correlation coefficient to eliminate the time difference caused by acquisition delay and processing delay, and form a sequence group with time - series synchronization. This step solves the problem of time - series inconsistency in the multi - signal - source acquisition process by establishing the time correspondence relationship between the fringe offset feature and the phase jump feature, ensuring the accurate alignment of different optical features in the time domain, and providing high - quality synchronous data for subsequent spectral analysis.

[0060] When performing short - time Fourier transform according to the synchronous sequence group, the sliding window technique is used for time - frequency analysis of the signal. A Hamming window or a Blackman window is set as the sliding window, the window length is usually set to 1 / 8 to 1 / 4 of the sequence length, and the window overlap rate is set to 50% - 75%. The fast Fourier transform (FFT) is applied to the data segment within each window to calculate the spectral components, and the results of all windows are combined into a spectral distribution matrix. The spectral distribution matrix is a two - dimensional array, the horizontal axis represents time, the vertical axis represents frequency, and the matrix element value represents the energy intensity of the corresponding time - frequency point. Non - linear decomposition algorithms such as singular value decomposition (SVD) are applied to this matrix to extract the main spectral components, which are separated into an amplitude - feature sequence and a phase - feature sequence. For example, for a molded lens with an edge thickness change of 3 - 8 μm, after short - time Fourier transform, the main spectral components are concentrated in the range of 2 - 5 Hz. The corresponding amplitude - feature sequence reflects the amplitude of the edge thickness change, and the phase - feature sequence reflects the spatial position information of the edge thickness distribution. This step realizes the conversion from time - domain signal to frequency - domain feature through time - frequency analysis methods, reveals the periodic characteristics and amplitude change rules of the edge thickness change of the lens, and provides a frequency - domain characterization for quantitative analysis of the edge thickness.

[0061] When performing temperature response analysis on the amplitude feature sequence and phase feature sequence in combination with the environmental temperature parameter, a temperature sensor array is used to measure the temperature distribution in the detection environment, and the temperature change curve over time is recorded. According to the thermo-optic coefficient of the optical material and the temperature dependence of the gas refractive index, the sensitivities of the amplitude feature and phase feature to temperature changes are analyzed. The multivariate regression analysis method is adopted to calculate the response function between the feature sequence and the temperature change, and the temperature correction factor is obtained. The temperature correction factor is a matrix that characterizes the degree to which different frequency components are affected by temperature. For example, when the environmental temperature changes from 22°C to 25°C, the amplitude correction factor of the low-frequency component (1 - 2Hz) is about 0.95, and the correction factor of the high-frequency component (4 - 5Hz) is about 0.88, indicating that the high-frequency component is more sensitive to temperature changes. The spectral features are compensated according to the temperature correction factor to eliminate the systematic error caused by temperature changes, and the temperature compensation sequence is obtained. This step solves the problem of interference of environmental temperature fluctuations on optical measurement by introducing a temperature correction mechanism, improves the measurement stability and consistency of the system under different environmental conditions, and ensures the reliability and repeatability of the measurement results.

[0062] When performing multi-scale filtering processing on the temperature compensation sequence, wavelet transform is used to perform joint time-frequency domain analysis on the signal. A suitable wavelet basis function (such as Daubechies or Symlet wavelet) is selected to decompose the signal into sub-signals of different frequency scales. According to the time-frequency characteristics of the signal, a suitable threshold is selected to perform threshold processing on the coefficients of each scale to remove high-frequency noise and low-frequency drift. Especially for the periodic interference caused by the rotation of the lens, a notch filter is used for directional suppression. The processed wavelet coefficients are reconstructed to obtain the steady-state spectral components. The steady-state spectral components refer to the stable spectral components that only reflect the change of the lens edge thickness after excluding environmental interference and random fluctuations. Feature extraction is performed on the steady-state spectral components, including the main frequency component, harmonic structure, phase distribution, etc., to form the steady-state spectral features. For example, for a standard lens with a uniform edge thickness, its steady-state spectral features are manifested as a single main frequency component with an obvious spectral peak; while for a lens with a non-uniform edge thickness, it is manifested as a mixture of multiple frequency components, and the phase distribution has an obvious spatial dependence. This step realizes the effective separation of useful signals and interference noises through multi-scale filtering technology, extracts the steady-state spectral components that truly reflect the characteristics of the lens edge thickness, and provides a reliable feature basis for the final thickness calculation.

[0063] Please continue to refer to Figure 1 , map the amplitude feature sequence and phase feature sequence in the steady-state spectral features to a preset feature space, and calculate the edge thickness distribution of the molded lens.

[0064] In an embodiment of the present invention, the mapping of the amplitude feature sequence and the phase feature sequence in the steady-state spectrum feature to a preset feature space and calculating the edge thickness distribution of the molded lens includes: Perform feature decoupling on the amplitude feature sequence and the phase feature sequence in the steady-state spectrum feature based on the preset feature space to obtain the air film perturbation momentum and the edge response amount, and obtain the physical space mapping factor according to the coupling relationship between the air film perturbation momentum and the edge response amount; Perform dynamic response analysis on the physical space mapping factor and the lens rotation angle sequence to obtain the edge rotation compensation coefficient, and perform non-linear correction on the edge rotation compensation coefficient in combination with the air film interface stress distribution to obtain the dynamic mapping basis vector; Perform air film thickness distribution correlation analysis on the dynamic mapping basis vector to obtain the interface tension distribution parameter, and perform morphological reconstruction on the air film microstructure according to the interface tension distribution parameter to obtain the edge topography vector; Perform air film flow field analysis according to the contour distribution of the edge topography vector to obtain the edge curvature correction amount, and perform local optimization on the edge curvature correction amount in combination with the air film shear stress to obtain the thickness correction coefficient; Perform multi-physical field coupling calculation on the thickness correction coefficient and the edge topography vector to obtain the edge thickness distribution sequence, and obtain the edge thickness distribution according to the edge thickness distribution sequence in combination with the steady-state response characteristics of the air film.

[0065] The following specifically describes the steps involved in the above embodiment: When performing feature decoupling on the amplitude feature sequence and the phase feature sequence in the steady-state frequency spectrum characteristics based on a preset feature space, the principal component analysis (PCA) or independent component analysis (ICA) algorithm is used to decompose the feature sequence. First, a feature vector matrix is constructed by combining the amplitude feature sequence and the phase feature sequence into a feature vector. Then, a projection transformation is performed in the preset feature space (a feature basis space established in advance through standard samples) to achieve feature separation. Through orthogonal transformation, the mixed features are decomposed into independent physical quantities: the air film perturbation momentum and the edge response quantity. The air film perturbation momentum describes the random fluctuation characteristics of the air film itself, while the edge response quantity reflects the effect of the change in the lens edge thickness on the air film. For example, when processing the measurement data of a certain typical molded lens, the air film perturbation momentum usually accounts for 15%-20% of the total change, and the edge response quantity accounts for 80%-85%. Subsequently, the interaction relationship between the two physical quantities is analyzed, and their coupling matrix is calculated to obtain the physical space mapping factor. This mapping factor reflects the conversion relationship between the frequency domain characteristics and the physical space parameters and is the key to mapping the measurement signal to the actual physical parameters. This step effectively separates the effective information and interference components in the measurement signal through the feature space decoupling technology, realizes the accurate conversion from the mixed signal to the physical parameters, and solves the problem of weak signals and strong interference in the optical measurement of transparent materials.

[0066] When performing dynamic response analysis based on the physical space mapping factor and the lens rotation angle sequence, a high-precision angle encoder is used to record the angle change sequence during the lens rotation. The cross-correlation analysis method is adopted to study the variation law of the physical space mapping factor with the rotation angle and identify the systematic errors caused by the lens rotation. The edge rotation compensation coefficient is calculated through least squares fitting, and this coefficient reflects the correction amount required for the measured values at different rotation angles. For example, when the lens rotation speed is 300 rpm, at specific angle positions such as 0°, 90°, 180°, and 270°, the edge rotation compensation coefficients are 1.05, 0.98, 1.03, and 0.95 respectively, indicating that the measurement results at different angle positions need to be compensated to different degrees. At the same time, a micro pressure sensor array is used to measure the stress distribution at the air film interface to obtain the interface pressure gradient data. Combining the edge rotation compensation coefficient and the interface stress distribution, a nonlinear correction is performed through a polynomial correction function to obtain the dynamic mapping basis vector. This basis vector contains the edge response characteristics of the lens in the rotating state, and the dimension is usually 12-16. This step solves the problem of unstable measurement under rotating conditions in the traditional method by introducing a rotation dynamic compensation mechanism, significantly improving the measurement accuracy and consistency.

[0067] When performing correlation analysis on the gas film thickness distribution of dynamic mapping basis vectors, a multivariate regression method is used to calculate the corresponding relationship between the basis vectors and the gas film thickness. The laser interferometry system is used to directly measure the gas film thickness to obtain the reference value of the thickness distribution. The dynamic mapping basis vectors are correlated with the gas film thickness distribution to calculate the interfacial tension distribution parameters. The interfacial tension distribution parameters describe the force state on the gas film surface and directly affect the shape and thickness distribution of the gas film. According to the interfacial tension distribution parameters, the numerical reconstruction method is used to reconstruct the microstructure of the gas film to obtain the edge topography vector. The edge topography vector is a set of parameters describing the geometric shape of the lens edge, including information such as edge curvature, slope, and height. For example, for the region where the edge thickness changes from 5μm to 8μm, the reconstructed edge topography vector can accurately express the spatial distribution characteristics of the thickness change with an error less than 0.2μm. This step realizes the mapping from the apparent characteristics of the gas film to the actual geometric parameters of the lens by establishing an accurate correspondence between the gas film characteristics and the lens edge topography, providing a geometric basis for the final measurement of the edge thickness.

[0068] When performing gas film flow field analysis according to the contour distribution of the edge topography vector, the computational fluid dynamics method is used to simulate the gas flow state at the lens edge. Based on the Navier-Stokes equations and boundary layer theory, the air velocity field and pressure field distributions under different edge topographies are calculated. Through the curvature analysis of the flow field, the curvature change characteristics of the air flow along the lens edge are extracted to obtain the edge curvature correction amount. The edge curvature correction amount reflects the difference between the actual flow field and the ideal flow field and is used to compensate for the air flow changes caused by the edge geometry. At the same time, the shear stress distribution in the gas film is calculated, and the correlation between the shear stress and the edge shape is analyzed. Combining the edge curvature correction amount and the shear stress distribution, local adjustment is performed through a weighted optimization algorithm to obtain the thickness correction coefficient. For example, for the edge region with a curvature radius of 3mm, the thickness correction coefficient is usually in the range of 0.95 - 1.05; while for the region with a curvature radius less than 1mm, the correction coefficient may reach 1.1 - 1.2, indicating that a larger correction is required in the high curvature region. This step solves the accuracy problem of measuring the lens edge with complex geometric shapes by introducing a flow field analysis and stress compensation mechanism, especially improving the measurement accuracy of special optical elements such as aspherical and free-form surfaces.

[0069] When performing multi-physics field coupling calculations on the thickness correction coefficient and the edge topography vector, the finite element analysis method is used to comprehensively consider the interactions of multiple physical fields such as optics, fluid mechanics, and mechanical stress. The thickness correction coefficient is used as a weight factor and weighted combined with the edge topography vector to calculate the edge thickness distribution sequence considering the influence of multiple physical fields. This sequence reflects the thickness change of the lens edge during a complete rotation cycle. The time series averaging and circumferential statistical methods are used to eliminate the influence of random fluctuations and system noise, and the final calibration is carried out in combination with the steady-state response characteristics of the air film to obtain the edge thickness distribution. For example, by using this method to measure a molded lens with a diameter of 30 mm, the measurement accuracy of the edge thickness can reach ±0.5 μm, and the repeatability is better than ±0.3 μm, fully meeting the detection requirements of high-precision optical components. This step comprehensively considers various influencing factors through multi-physics field coupling calculations, establishes an accurate mapping relationship from the measurement signal to the actual thickness, realizes high-precision non-contact measurement of the edge thickness of the molded lens, and solves the technical problem of difficult edge detection of transparent materials in traditional methods.

[0070] The above describes the vision-based edge thickness detection method for molded lenses in the embodiments of the present invention. Next, the vision-based edge thickness detection device in the embodiments of the present invention will be described. Please refer to Figure 2 , an embodiment of the vision-based edge thickness detection device for molded lenses in the embodiments of the present invention includes: An air film control module 101, configured to introduce a visible gas into the edge region of the molded lens, control the flow rate, thickness, density, and pressure of the visible gas, form an air film disturbance layer on the circumferential surface of the edge of the molded lens, and obtain the air film thickness change amount and disturbance deformation amount in the air film disturbance layer; An optical excitation module 102, configured to select a corresponding light incident angle according to the air film thickness change amount in the air film disturbance layer, excite the disturbance deformation amount in the air film disturbance layer to generate a refractive index change, and obtain the refractive index distribution change amount and scattered light intensity change amount in the air film disturbance layer; An interference acquisition module 103, configured to configure a main light source optical path and a reference optical path based on the refractive index distribution change amount of the air film disturbance layer, collect the interference phase image corresponding to the scattered light intensity change amount, and obtain the fringe offset sequence and phase jump sequence of the interference phase image; A spectrum analysis module 104, configured to perform Fourier transform analysis according to the fringe offset sequence and phase jump sequence, extract the corresponding amplitude feature sequence and phase feature sequence, and obtain the steady-state spectrum feature through ambient temperature correction; An edge thickness calculation module 105, configured to map the amplitude feature sequence and phase feature sequence in the steady-state spectrum feature to a preset feature space, and calculate the edge thickness distribution of the molded lens.

[0071] The aboveFigure 2 From the perspective of modular functional entities, the vision-based edge thickness detection device for molded lenses in the embodiments of the present invention will be described in detail. Next, from the perspective of hardware processing, the vision-based edge thickness detection device for molded lenses in the embodiments of the present invention will be described in detail.

[0072] Figure 3 FIG. 5 is a schematic structural diagram of a vision-based edge thickness detection device for molded lenses provided by an embodiment of the present invention. The vision-based edge thickness detection device 200 may vary greatly due to different configurations or performances, and may include one or more processors 210 (for example, one or more processors) and a memory 220, and one or more storage media 230 (for example, one or more mass storage device ends) storing application programs 233 or data 232. Among them, the memory 220 and the storage media 230 may be transient storage or persistent storage. The program stored in the storage media 230 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations on the vision-based edge thickness detection device 200. Further, the processor 210 may be configured to communicate with the storage media 230 and execute a series of instruction operations in the storage media 230 on the vision-based edge thickness detection device 200 to implement the steps of the above-mentioned vision-based edge thickness detection method.

[0073] The vision-based edge thickness detection device 200 may further include one or more power supplies 240, one or more wired or wireless network interfaces 250, one or more input / output interfaces 260, and / or one or more operating systems 231, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art can understand that Figure 3 The shown structural diagram of the vision-based edge thickness detection device does not limit the vision-based edge thickness detection device provided by the present invention, and may include more or fewer components than shown in the figure, or combine some components, or have different component arrangements.

[0074] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions run on a computer, the computer is caused to execute the steps of the vision-based edge thickness detection method.

[0075] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the systems, devices, or units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0076] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0077] The foregoing are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention. Any equivalent structural transformation made by using the content of the specification and drawings of the present invention under the inventive concept of the present invention, or any direct / indirect application in other related technical fields is included in the patent protection scope of the present invention.

Claims

1. A vision-based method for detecting the edge thickness of a molded lens, characterized in that, Including: Introduce a visible gas into the edge region of the molded lens, control the flow rate, thickness, density, and pressure of the visible gas, form a gas film disturbance layer on the circumferential surface of the edge of the molded lens, and obtain the gas film thickness change amount and disturbance deformation amount in the gas film disturbance layer; According to the gas film thickness change amount in the gas film disturbance layer, select a corresponding light incident angle to excite the disturbance deformation amount in the gas film disturbance layer to generate a refractive index change, and obtain the refractive index distribution change amount and scattered light intensity change amount in the gas film disturbance layer; Based on the refractive index distribution change amount of the gas film disturbance layer, configure a main light source optical path and a reference optical path, collect the interference phase image corresponding to the scattered light intensity change amount, and obtain the fringe shift sequence and phase jump sequence of the interference phase image; Perform Fourier transform analysis according to the fringe shift sequence and phase jump sequence, extract the corresponding amplitude feature sequence and phase feature sequence, and obtain the steady-state spectrum feature through ambient temperature correction; Map the amplitude feature sequence and phase feature sequence in the steady-state spectrum feature to a preset feature space, and calculate the edge thickness distribution of the molded lens.

2. The vision-based method for detecting the edge thickness of a molded lens according to claim 1, wherein The step of introducing a visible gas into the edge region of the molded lens, controlling the flow rate, thickness, density, and pressure of the visible gas, forming a gas film disturbance layer on the circumferential surface of the edge of the molded lens, and obtaining the gas film thickness change amount and disturbance deformation amount in the gas film disturbance layer includes: Detect the linear velocity of the rotating molded lens, determine the initial flow rate of the visible gas according to the linear velocity, and calculate the velocity ratio between the flow rate of the visible gas and the linear velocity; Adjust the pressure coefficient of the visible gas according to the velocity ratio, and determine the density distribution parameter of the visible gas in combination with the ambient temperature; Calculate the gas film adhesion parameter based on the product of the density distribution parameter and the pressure coefficient, and control the introduction amount of the visible gas on the circumferential surface of the edge of the molded lens according to the gas film adhesion parameter; Perform uniformity analysis on the introduction amount of the visible gas to obtain the radial distribution coefficient and circumferential distribution coefficient in the gas film disturbance layer; Calculate the gas film thickness change amount in the gas film disturbance layer according to the radial distribution coefficient and the circumferential distribution coefficient; Perform gradient analysis on the gas film thickness change amount to obtain the disturbance deformation amount in the gas film disturbance layer.

3. The vision-based method for detecting the edge thickness of a molded lens according to claim 2, wherein The step of performing uniformity analysis on the introduction amount of the visible gas to obtain the radial distribution coefficient and circumferential distribution coefficient in the gas film disturbance layer includes: Perform a flow rate gradient analysis on the introduction amount of the visible gas along the radial direction to obtain the radial flow rate change rate, and obtain the radial gas flow modulation coefficient according to the corresponding relationship between the radial flow rate change rate and the surface curvature of the molded lens; Perform circumferential gas flow distribution analysis according to the radial gas flow modulation coefficient to obtain the circumferential flow compensation amount, and perform Coriolis force correction on the circumferential flow compensation amount in combination with the rotation speed of the molded lens to obtain the circumferential distribution coefficient; Perform gas flow stability calculation on the circumferential distribution coefficient along the edge contour of the molded lens to obtain the stability criterion value, and correct the radial gas flow modulation coefficient according to the stability criterion value to obtain the radial distribution coefficient.

4. The vision-based method for detecting the edge thickness of a molded lens according to claim 1, wherein Selecting a corresponding light incident angle according to the change amount of the air film thickness in the air film disturbance layer, exciting a refractive index change by generating a disturbance deformation amount in the air film disturbance layer, and obtaining the change amount of the refractive index distribution and the change amount of the scattered light intensity of the air film disturbance layer, including: Performing incident angle correlation analysis according to the corresponding relationship between the air film thickness change amount and the local curvature of the air film disturbance layer, obtaining an angle sensitivity coefficient, and performing local stress field correction on the angle sensitivity coefficient to obtain an angle mapping coefficient; Performing hierarchical processing on the angle mapping coefficient combined with the shear strain distribution of the air film disturbance layer, calculating the shear response factor of the light incident angle of each layer, and obtaining an optical coupling parameter according to the interaction between the shear response factor and the disturbance deformation amount; Performing light excitation on the molecular motion degrees of freedom in the air film disturbance layer according to the optical coupling parameter and the disturbance deformation amount, obtaining the molecular motion state quantity inside the air film, and combining the spatial distribution of the intermolecular force to obtain the molecular arrangement state quantity; Performing transient response analysis based on the mapping relationship between the molecular arrangement state quantity and the molecular orientation distribution, calculating the local refractive index gradient coefficient, and obtaining the refractive index distribution gradient coefficient according to the spatial distribution characteristics of the gradient coefficient; Performing zonal scanning on the air film disturbance layer according to the refractive index distribution gradient coefficient, obtaining the change amount of the regional refractive index, and performing spatial superposition processing on the change amount of the regional refractive index to obtain the change amount of the refractive index distribution; Performing light intensity modulation analysis on the change amount of the refractive index distribution combined with the spatial distribution of the light energy, obtaining the light intensity distribution coefficient, and performing energy conservation correction according to the light intensity distribution coefficient to obtain the change amount of the scattered light intensity.

5. The vision-based method for detecting the edge thickness of a molded lens according to claim 4, wherein Performing light excitation on the molecular motion degrees of freedom in the air film disturbance layer according to the optical coupling parameter and the disturbance deformation amount, obtaining the molecular motion state quantity inside the air film, and combining the spatial distribution of the intermolecular force to obtain the molecular arrangement state quantity, including: Calculating the molecular excitation energy according to the product relationship between the optical coupling parameter and the disturbance deformation amount, obtaining the molecular energy level transition coefficient, and analyzing the molecular orientation distribution of the air film based on the molecular energy level transition coefficient to obtain the molecular polarization intensity; Performing gradient analysis on the molecular polarization intensity along the air film thickness direction, obtaining the polarization field distribution quantity, and performing interaction calculation according to the polarization field distribution quantity combined with the intermolecular van der Waals force to obtain the intermolecular interaction potential distribution; Performing statistical mechanics analysis on the intermolecular interaction potential distribution, obtaining the molecular motion state quantity inside the air film, and obtaining the molecular arrangement state quantity according to the coupling relationship between the molecular motion state quantity and the local temperature field.

6. The vision-based method for detecting the edge thickness of a molded lens according to claim 1, wherein Configuring the main light source optical path and the reference optical path based on the change amount of the refractive index distribution of the air film disturbance layer, collecting the interference phase image corresponding to the change amount of the scattered light intensity, and obtaining the fringe offset sequence and the phase jump sequence of the interference phase image, including: Performing optical path difference analysis according to the spatial frequency distribution of the change amount of the refractive index distribution, obtaining the optical path spectrum coefficient, and performing phase modulation response calculation on the optical path spectrum coefficient to obtain the modulation parameter of the main light source optical path; Perform optical intensity compensation processing on the modulation parameters of the main light source optical path in combination with the change amount of the scattered light intensity, calculate the balance factor of the reference optical path, and perform coherence optimization based on the balance factor to obtain the double optical path coupling coefficient; Construct an interference field according to the double optical path coupling coefficient and the change amount of the scattered light intensity, obtain the interference fringe intensity distribution amount, and combine the spatial gradient characteristics of the intensity distribution amount to obtain the interference phase image; Extract the transient characteristics of the interference phase image, obtain the displacement amount of the fringe sequence, and obtain the fringe offset sequence according to the cumulative change law of the displacement amount of the fringe sequence; Perform phase continuity analysis according to the fringe offset sequence, obtain the coordinates of the phase mutation point, and perform periodic identification on the coordinates of the phase mutation point to obtain the phase jump period; Perform sequence reconstruction on the phase jump period in combination with the phase distribution, obtain the phase jump amplitude, and obtain the phase jump sequence according to the time series change of the phase jump amplitude.

7. The method for detecting the edge thickness of a molded lens based on vision according to claim 1, wherein, The Fourier transform analysis is performed according to the fringe offset sequence and the phase jump sequence, and the corresponding amplitude feature sequence and phase feature sequence are extracted, and the steady-state spectrum feature is obtained through environmental temperature correction, including: Perform cross-correlation analysis on the fringe offset sequence and the phase jump sequence, obtain the sequence correlation coefficient, and perform time series synchronization processing according to the correlation coefficient to obtain the synchronized sequence group; Perform short-time Fourier transform according to the synchronized sequence group, obtain the spectrum distribution matrix, and perform nonlinear decomposition on the spectrum distribution matrix to obtain the amplitude feature sequence and the phase feature sequence; Perform temperature response analysis on the amplitude feature sequence and the phase feature sequence in combination with the environmental temperature parameters, obtain the temperature correction factor, and perform spectrum compensation according to the temperature correction factor to obtain the temperature compensation sequence; Perform multi-scale filtering processing on the temperature compensation sequence, obtain the steady-state spectrum component, and obtain the steady-state spectrum feature according to the distribution characteristics of the steady-state spectrum component.

8. The vision-based method for detecting the edge thickness of a molded lens according to claim 1, wherein The amplitude feature sequence and the phase feature sequence in the steady-state spectrum feature are mapped to a preset feature space, and the edge thickness distribution of the molded lens is calculated, including: Perform feature decoupling on the amplitude feature sequence and the phase feature sequence in the steady-state spectrum feature based on the preset feature space, obtain the air film perturbation amount and the edge response amount, and obtain the physical space mapping factor according to the coupling relationship between the air film perturbation amount and the edge response amount; Perform dynamic response analysis according to the physical space mapping factor and the lens rotation angle sequence, obtain the edge rotation compensation coefficient, and perform nonlinear correction on the edge rotation compensation coefficient in combination with the air film interface stress distribution to obtain the dynamic mapping basis vector; Perform air film thickness distribution correlation analysis on the dynamic mapping basis vector, obtain the interface tension distribution parameter, and perform morphological reconstruction on the air film microstructure according to the interface tension distribution parameter to obtain the edge morphology vector; Perform air film flow field analysis according to the contour distribution of the edge morphology vector, obtain the edge curvature correction amount, and perform local optimization on the edge curvature correction amount in combination with the air film shear stress to obtain the thickness correction coefficient; Perform multi-physical field coupling calculations on the thickness correction coefficient and the edge topography vector to obtain an edge thickness distribution sequence, and based on the edge thickness distribution sequence and combined with the steady-state response characteristics of the air film, obtain the edge thickness distribution.

9. A vision-based edge thickness detection device for molded lenses, characterized in that, The vision-based edge thickness detection device for molded lenses adopts the vision-based edge thickness detection method according to any one of claims 1 to 8. The vision-based edge thickness detection device for molded lenses includes: An air film control module for introducing a visible gas into the edge region of the molded lens, controlling the flow rate, thickness, density, and pressure of the visible gas, forming an air film disturbance layer on the circumferential surface of the edge of the molded lens, and obtaining the air film thickness change amount and disturbance deformation amount in the air film disturbance layer; An optical excitation module for selecting a corresponding light incident angle based on the air film thickness change amount in the air film disturbance layer, exciting a refractive index change generated by the disturbance deformation amount in the air film disturbance layer, and obtaining the refractive index distribution change amount and scattered light intensity change amount in the air film disturbance layer; An interference acquisition module for configuring a main light source optical path and a reference optical path based on the refractive index distribution change amount of the air film disturbance layer, acquiring an interference phase image corresponding to the scattered light intensity change amount, and obtaining a fringe shift sequence and a phase jump sequence of the interference phase image; A spectrum analysis module for performing Fourier transform analysis according to the fringe shift sequence and the phase jump sequence, extracting corresponding amplitude feature sequences and phase feature sequences, and obtaining steady-state spectrum features through ambient temperature correction; An edge thickness calculation module for mapping the amplitude feature sequences and phase feature sequences in the steady-state spectrum features to a preset feature space and calculating the edge thickness distribution of the molded lens.

10. A vision-based edge thickness detection device for molded lenses, characterized in that, The vision-based edge thickness detection device for molded lenses includes: a memory and at least one processor, and instructions are stored in the memory; The at least one processor calls the instructions in the memory so that the vision-based edge thickness detection device executes the steps of the vision-based edge thickness detection method according to any one of claims 1-8.

Images