Intelligent correction method and system for optical performance of functional lens

Abstract

The invention relates to the technical field of lens correction, and discloses an intelligent correction method and system for the optical performance of a functional lens, and the method comprises the steps: obtaining the original data of the lens, and carrying out the preprocessing of the original data of the lens, and obtaining the surface data of the lens; performing surface parameter calculation according to the lens surface data to obtain three-dimensional surface parameters; performing weighted residual calculation according to the three-dimensional surface parameters to obtain a residual matrix; performing spatial correlation analysis according to the three-dimensional surface parameters and the residual matrix to obtain an error region and a normal region; performing compensation analysis according to the residual matrix, the error region and the normal region to obtain a parameter compensation amount; and performing multi-objective optimization according to the parameter compensation amount to obtain an optimal compensation parameter, and correcting the lens according to the optimal compensation parameter. The method has the following effect that the correction accuracy of the optical performance of the functional lens can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of lens correction, and particularly to an intelligent correction method and system for the optical performance of functional lenses. Background Art

[0002] With the development of technology, functional lenses (such as anti-blue light, anti-ultraviolet, and progressive multifocal lenses) play an increasingly important role in improving people's visual experience. However, traditional manufacturing methods are difficult to achieve precise control of optical performance. Especially in mass production, problems such as optical parameter deviation are likely to occur. To solve these problems, intelligent correction methods have emerged, aiming to adjust and optimize the optical performance of functional lenses in real time through advanced technical means to ensure the high quality and consistency of the final products.

[0003] An existing intelligent correction method includes the following steps: First, use high-precision measuring instruments to obtain the actual optical performance data of functional lenses, and these data include key indicators such as light transmittance, refractive index, and dispersion coefficient. Second, input the collected data into a pre-constructed mathematical model or simulation software, which can simulate the optical performance of the lens under different environmental conditions. Then, based on the comparison analysis results, that is, the difference between the actual performance and the ideal design, automatically adjust the parameters of the production equipment, such as temperature, pressure, time, etc., to achieve precise control of the optical performance of the lens. Finally, perform performance detection again to ensure that the corrected lens meets the expected standards.

[0004] Although the above intelligent correction method significantly improves the production quality and efficiency of functional lenses, it still has some limitations. Although the existing mathematical models and simulation software can better simulate the performance of the lens under specific conditions, for complex and variable actual application environments, they cannot fully predict all potential problems. For example, the aging effect of materials after long-term use or the deformation of some areas of the lens caused by extreme climate conditions, resulting in low correction accuracy. In summary, there are problems in the prior art that all potential problems cannot be fully predicted and the correction accuracy is low. Summary of the Invention

[0005] The present invention provides an intelligent correction method and system for the optical performance of functional lenses to achieve improved accuracy of intelligent correction of the optical performance of functional lenses.

[0006] In a first aspect, to solve the above technical problems, the present invention provides an intelligent correction method for the optical performance of functional lenses, including: Obtain the original lens data and preprocess the original lens data to obtain the lens surface data; Calculate the surface parameters according to the lens surface data to obtain three-dimensional surface parameters; Perform weighted residual calculation based on the three-dimensional surface parameters to obtain a residual matrix; Perform spatial correlation analysis based on the three-dimensional surface parameters and the residual matrix to obtain an error region and a normal region; Perform compensation analysis based on the residual matrix, the error region, and the normal region to obtain a parameter compensation amount; Perform multi-objective optimization based on the parameter compensation amount to obtain optimal compensation parameters, and correct the lens according to the optimal compensation parameters.

[0007] In an alternative embodiment, the calculating the three-dimensional surface parameters according to the lens surface data includes: Calculate the lens power by the following formula:

[0008] where, represents the lens power, represents the refractive index of the lens, represents the front surface curvature radius of the lens, represents the rear surface curvature radius of the lens; Calculate the lens sag by the following formula:

[0009] where, represents the lens sag, represents the actual diameter of the lens; Calculate the freeform parameter by the following formula:

[0010] where, represents the freeform parameter of the th order and the th term, represents the radial order, represents pi, represents the wavefront error distribution in polar coordinates , represents the freeform polynomial of the th order and the th term, represents the normalized radial distance, represents the azimuth angle, represents the differential symbol of represents the differential symbol of where, the three-dimensional surface parameters include the lens power, the lens sag, and the freeform parameter.

[0011] In an alternative embodiment, calculating the weighted residual according to the three-dimensional surface parameters to obtain a residual matrix includes: Obtaining a theoretical parameter matrix and a residual weight matrix; Subtracting the three-dimensional surface parameters from the theoretical parameter matrix to obtain an original residual matrix; Calculating the residual matrix from the original residual matrix and the residual weight matrix through the following formula:

[0012] where, represents the residual matrix, represents the original residual matrix, represents the square root matrix of the residual weight matrix.

[0013] In an alternative embodiment, performing spatial correlation analysis according to the three-dimensional surface parameters and the residual matrix to obtain an error region and a normal region includes: Dividing the three-dimensional surface parameters into blocks according to a preset number of regions to obtain block region data; Calculating a spatial correlation coefficient according to the block region data:

[0014] where, represents the spatial correlation coefficient, represents the total number of data points in a block region, represents the data point number, represents the spatial weight coefficient between the -th data point and the -th data point, represents the residual of the -th data point, represents the residual of the -th data point, represents the average value of the residual matrix; When the spatial correlation coefficient is greater than a preset spatial threshold, marking that there are regional manufacturing errors in the corresponding block region data and marking the corresponding block region as an error region;

[0015] In an alternative embodiment, performing compensation analysis according to the residual matrix, the error region, and the normal region to obtain a parameter compensation amount includes: Calculating the parameter compensation amount for the normal region using the following formula:

[0016] Among them, represents the parameter compensation amount, represents the normal compensation initial value, represents the natural constant, represents the standard deviation of the residual matrix; The following formula is used to calculate the parameter compensation amount for the error region:

[0017] Among them, represents the parameter compensation amount, represents the error compensation initial value, represents the natural constant, represents the standard deviation of the residual matrix.

[0018] In an alternative embodiment, the multi-objective optimization is performed according to the parameter compensation amount to obtain the optimal compensation parameter, and the lens is corrected according to the optimal compensation parameter, including: Obtain the lens design parameters and regularization parameters; Construct a Jacobian matrix according to the lens design parameters and the parameter compensation amount; Perform error calculation according to the Jacobian matrix to obtain an output error vector; Update the parameter compensation amount according to the regularization parameter and the output error vector to obtain an updated compensation parameter; Replace the parameter compensation amount with the updated compensation parameter; When the norm of the output error vector is greater than a preset error threshold, reconstruct the Jacobian matrix; When the norm of the output error vector is less than a preset error threshold, output the updated compensation parameter as the optimal compensation parameter; Correct the lens according to the optimal compensation parameter.

[0019] In an alternative embodiment, the constructing a Jacobian matrix according to the lens design parameters and the parameter compensation amount includes: Calculate the Jacobian matrix through the following formula:

[0020] Among them, represents the element in the th row and th column of the Jacobian matrix, represents the th parameter compensation amount, represents the th lens design parameter, Represents the symbol of partial derivative.

[0021] In a second aspect, the present invention provides an intelligent correction system for the optical performance of functional lenses, comprising: A data acquisition module, configured to acquire the original lens data and preprocess the original lens data to obtain the lens surface data; A surface parameter module, configured to calculate surface parameters based on the lens surface data to obtain three-dimensional surface parameters; A residual matrix module, configured to calculate the weighted residual based on the three-dimensional surface parameters to obtain a residual matrix; A spatial analysis module, configured to perform spatial correlation analysis based on the three-dimensional surface parameters and the residual matrix to obtain an error region and a normal region; A compensation analysis module, configured to perform compensation analysis based on the residual matrix, the error region and the normal region to obtain a parameter compensation amount; An optimization and correction module, configured to perform multi-objective optimization based on the parameter compensation amount to obtain optimal compensation parameters, and correct the lens according to the optimal compensation parameters.

[0022] In a third aspect, the present invention further provides an electronic device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the intelligent correction method for the optical performance of the functional lens described in any one of the above is implemented.

[0023] In a fourth aspect, the present invention further provides a computer-readable storage medium, which includes a stored computer program. When the computer program runs, the device where the computer-readable storage medium is located is controlled to execute the intelligent correction method for the optical performance of the functional lens described in any one of the above.

[0024] Compared with the prior art, the present invention has the following beneficial effects: (1) Acquire the original lens data and preprocess the original lens data to obtain the lens surface data. This step improves the quality of the lens surface data by removing noise and outliers, ensuring the accuracy of subsequent calculations. The high-quality data input lays a foundation for accurate three-dimensional surface parameter calculation, thereby improving the accuracy of the entire process.

[0025] (2) Calculate surface parameters based on the lens surface data to obtain three-dimensional surface parameters. This process uses an advanced algorithm model to comprehensively analyze the lens surface characteristics, realizing the conversion from two-dimensional data to three-dimensional parameters. The accurate three-dimensional surface parameters not only help to more precisely describe the lens surface characteristics, but also provide a reliable basis for subsequent weighted residual calculation, enhancing the reliability of the overall processing.

[0026] (3) Calculate the weighted residuals based on the three-dimensional surface parameters to obtain a residual matrix. This step quantifies and analyzes the differences between the three-dimensional surface parameters and the ideal model to identify the errors existing on the lens surface. The weighted residual calculation method takes into account the differences in importance of different regions, making the error evaluation more scientific and reasonable, and providing support for accurate spatial correlation analysis.

[0027] (4) Conduct spatial correlation analysis based on the three-dimensional surface parameters and the residual matrix to obtain the error region and the normal region. This step divides the lens surface into different regions by comprehensively analyzing the three-dimensional surface parameters and the residual matrix for targeted compensation analysis. Clearly distinguishing the error region and the normal region helps optimize the correction strategy and improve the correction efficiency and quality.

[0028] (5) Conduct compensation analysis based on the residual matrix, the error region, and the normal region to obtain the parameter compensation amount. In this process, the system calculates the specific compensation amount for each error region to correct the irregularities on the lens surface. This compensation method based on detailed analysis can effectively reduce errors and improve the flatness and optical performance of the lens surface.

[0029] (6) Perform multi-objective optimization based on the parameter compensation amount to obtain the optimal compensation parameters, and correct the lens according to the optimal compensation parameters. Through the multi-objective optimization algorithm, the best compensation scheme is found under the condition of meeting multiple constraint conditions. This method not only improves the correction accuracy but also takes into account other key performance indicators such as light transmittance and refractive index. Finally, correcting the lens according to the optimal compensation parameters can significantly improve the overall performance of the lens and meet the high-standard application requirements. Description of the Drawings

[0030] Figure 1 is a schematic flowchart of an intelligent correction method for the optical performance of a functional lens provided by the first embodiment of the present invention; Figure 2 is a schematic structural diagram of an intelligent correction system for the optical performance of a functional lens provided by the second embodiment of the present invention. Detailed Embodiments

[0031] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the 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 the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0032] Refer to Figure 1The first embodiment of the present invention provides an intelligent correction method for the optical performance of a functional lens, comprising the following steps: S11, acquiring original lens data, and preprocessing the original lens data to obtain lens surface data; S12, calculating surface parameters according to the lens surface data to obtain three-dimensional surface parameters; S13, performing weighted residual calculation according to the three-dimensional surface parameters to obtain a residual matrix; S14, performing spatial correlation analysis according to the three-dimensional surface parameters and the residual matrix to obtain an error region and a normal region; S15, performing compensation analysis according to the residual matrix, the error region and the normal region to obtain a parameter compensation amount; S16, performing multi-objective optimization according to the parameter compensation amount to obtain optimal compensation parameters, and correcting the lens according to the optimal compensation parameters.

[0033] In step S11, original lens data is acquired and preprocessed to obtain lens surface data.

[0034] In one embodiment, the raw data of the lens is acquired by using a white light interferometer profilometer in conjunction with a high-precision five-axis adjustment table. The specific operation process is as follows: first, a mixed solution of ultrapure water and ethanol in a volume ratio of 1:1 is used to clean the surface of the lens. In a Class 100 clean environment, a vacuum adsorption fixture is used to fix the lens on an air-floating vibration isolation platform. The interferometer scanning interval is set to 0.25 microns, the longitudinal resolution is 0.1 nanometers, and the scanning speed is adaptively adjusted according to the radius of curvature (0.5 mm / s in the flat area and reduced to 0.1 mm / s in the steep change area). The real-time wavefront sampling mode of 60 frames / second is turned on, and the environmental compensation module is simultaneously started to maintain the temperature at 20±0.1°C and the humidity at 50±5%RH. When reconstructing the surface morphology through the phase unwrapping algorithm, the effective fringe contrast threshold needs to be set to be greater than 0.7, and the phase difference between adjacent pixels is limited to within π / 4 radians. Finally, the least squares fitting is applied to eliminate the clamping tilt error, and the unfiltered raw height matrix data is output. The storage format adopts the 32-bit floating point ASCⅡ encoding specified in Part 5 of the international standard ISO 10110.

[0035] In one implementation, the preprocessing of lens data includes eight core steps: First, morphological outlier rejection is performed to eliminate isolated noise points caused by environmental particles or transient interference during measurement; then, adaptive Gaussian filtering is carried out for noise reduction, dynamically adjusting the spatial filter kernel size according to the surface roughness level to suppress high-frequency noise while retaining effective features; then, least squares reference plane fitting compensation is implemented to eliminate systematic deviations introduced by clamping tilt; for the edge diffraction effect unique to interferometric measurement, the cubic spline extension method is used for edge data correction; for local regions with abnormal reflectivity, the thin plate spline interpolation algorithm is used to complete data filling; the measurement coordinate system and the design coordinate system are accurately registered through a rigid body transformation matrix; density optimization processing is performed on the discrete data according to the Nyquist sampling theorem; finally, the temperature-humidity coupling drift correction amount is superimposed to compensate for the material expansion effect caused by environmental parameter fluctuations. Metadata logs are generated synchronously throughout the process, recording the parameter adjustment trajectories and data quality evaluation indicators at each stage.

[0036] In step S12, three-dimensional surface parameters are obtained by calculating surface parameters based on the lens surface data.

[0037] In one implementation, the lens optical power is calculated by the following formula:

[0038] where, represents the lens optical power, represents the lens refractive index, represents the front surface curvature radius of the lens, represents the rear surface curvature radius of the lens; The lens sag is calculated by the following formula:

[0039] where, represents the lens sag, represents the actual diameter of the lens; The free form parameter is calculated by the following formula:

[0040] where, represents the free form parameter of the th order and the th term, represents the radial order, represents pi, represents the wavefront error distribution in polar coordinates , represents the free form polynomial of the th order and the th term, represents the normalized radial distance, represents the azimuth angle, represents the differential symbol of represents the differential symbol of; Among them, the three-dimensional surface parameters include the lens optical power, the lens sagittal height, and the freeform parameter.

[0041] It should be noted that the lens optical power characterizes the ability of the lens to deflect light, with the unit of diopter (D), and determines the core optical performance of the lens for correcting vision. The refractive index appearing in the calculation formula is an inherent property of the material, representing the ratio of the propagation speed of light in the lens material to the speed in a vacuum. The refractive index range of conventional resin lenses is 1.50 - 1.67. The radius of curvature represents the curvature characteristics of the ideal spherical surface of the front surface or the rear surface, taking a positive value for a convex surface and a negative value for a concave surface. The radius of curvature is obtained through an iterative least-squares spherical fitting algorithm: an effective optical zone is selected from the preprocessed surface data, a spherical equation is established for optimal fitting, and the spherical radius corresponding to the minimum sum of squared residuals is the measured value.

[0042] It should be noted that the lens sagittal height represents the maximum vertical height from the vertex of the lens surface to the reference plane, which directly affects the center thickness of the lens and the assembly parameters. The actual diameter represents the physical outer diameter of the effective optical zone of the lens, excluding non-functional areas such as chamfers and edge damages. The actual diameter is determined by the edge gradient detection method: the surface height data is scanned circumferentially, and when the radial height gradient exceeds a set threshold (set to 0.5 mm in this method), it is determined as an effective boundary, and the maximum circumscribed circle diameter is calculated through the minimum bounding circle algorithm.

[0043] It should be noted that the freeform parameter characterizes the deviation degree between the lens surface and the ideal shape, p order m terms correspond to specific types of aberration modes (such as p = 2 corresponds to defocus, p = 3 corresponds to astigmatism, etc.). The surface deformation distribution in the polar coordinate system, ρ ∈[0, 1] represents the normalized radius position, θ ∈[0, 360°] represents the azimuth angle. The Zernike polynomials are a set of orthonormal basis function groups used to decompose complex surface deformations. Different ( p , m ) combinations correspond to the aberration types defined by the international standard. The wavefront error distribution directly comes from the difference matrix between the preprocessed surface height data and the designed surface. The Zernike polynomials are generated by a standard orthonormalization procedure. During calculation, the Gaussian numerical integration method is used at discrete data points, and the surface error is projected onto each order of basis functions through weighted summation to finally obtain the coefficient values of each order.

[0044] It should be noted that the natural shape parameters are actually the coefficients of the Zernike polynomial coefficients, stemming from their essential expression ability for optical surface features. In optical inspection, these coefficients correspond to the internationally recognized aberration classification system: low-order coefficients (such as p = 2) describe the basic refractive deviation (myopia / hyperopia), middle-order coefficients (p = 3 - 6) correspond to regular defects such as astigmatism and coma, and high-order coefficients (p > 6) reflect local processing defects. Each coefficient value directly quantifies the intensity of a specific aberration mode. For example an increase indicates the presence of astigmatism in the 45° direction. This parameter system has physical interpretability - it can be judged without complex calculations: when exceeds the threshold, it indicates that the lens has spherical aberration and needs to be repaired. The orthogonality between parameters ensures that each aberration mode is evaluated independently, avoiding error coupling interference in quality judgment. Industry standards (such as ISO 10110) directly adopt this parameter system to formulate tolerance specifications, making the inspection results comparable across platforms.

[0045] In step S13, a weighted residual calculation is performed based on the three-dimensional surface parameters to obtain a residual matrix.

[0046] In one implementation, a theoretical parameter matrix and a residual weight matrix are obtained; The difference is taken between the theoretical parameter matrix and the three-dimensional surface parameters to obtain an original residual matrix; The residual matrix is calculated from the original residual matrix and the residual weight matrix through the following formula:

[0047] where, represents the residual matrix, represents the original residual matrix, represents the square root matrix of the residual weight matrix.

[0048] It should be noted that in the weighted residual calculation of lens quality inspection, it is first necessary to construct a theoretical parameter matrix (including the designed optical parameters of each position of the lens such as optical power and sagitta), a residual weight matrix (weighted according to the regional importance, such as the weight of the optical center area is 1.0 and the edge area is 0.3), and an original residual matrix (the point-by-point difference between the measured value and the theoretical value). When performing weighted calculation through the formula, the square root of the weight matrix participates in the operation to maintain dimensional consistency: for example, the +0.5D original residual of a certain middle annulus (weight 0.6) is reduced to 0.30D after being weighted twice by a factor of 0.775, reflecting the process tolerance design. This weighted mechanism enables the inspection system to accurately locate defects (such as an abnormal residual of 0.8D in the θ = 45° direction indicating a processing deviation), and achieve quality grading through quantitative evaluation (such as being judged qualified when the L2 norm < 0.15). In practical applications, the yield rate of multifocal lenses is increased by 13%, and at the same time, the residual distribution statistics provide data support for machine tool calibration (such as the residual of the ρ = 0.5 annulus reflecting coating fluctuations).

[0049] In step S14, spatial correlation analysis is performed based on the three-dimensional surface parameters and the residual matrix to obtain an error region and a normal region.

[0050] In one implementation, the three-dimensional surface parameters are segmented according to a preset number of regions to obtain segmented region data; Calculate the spatial correlation coefficient according to the segmented region data:

[0051] where, represents the spatial correlation coefficient, represents the total number of data points in a segmented region, represents the data point number, represents the spatial weight coefficient between the th data point and the th data point, represents the residual of the th data point, represents the residual of the th data point, represents the average value of the residual matrix; When the spatial correlation coefficient is greater than a preset spatial threshold, mark that there is regional manufacturing error in the corresponding segmented region data, and mark the corresponding segmented region as an error region;

[0052] It should be noted that the initial compensation value for the error region will be greater than that for the normal region in the future, that is to say, a greater compensation will be made for the error region.

[0053] It should be noted that in this step, the lens surface is first divided into several sub-regions according to process characteristics (for example, a progressive lens is divided into a distance vision area, a progressive channel, a near vision area, etc.), and each region contains several detection point data. When calculating, a spatial weight coefficient is introduced (reflecting the distance attenuation effect between adjacent detection points, such as the weight of point pairs within 1 mm is set to 0.9, and the weight of point pairs outside 5 mm drops to 0.1), and combined with the residual value of each point (the deviation between the measured value and the theoretical parameter) and its mean value, a spatial correlation coefficient is constructed. This coefficient quantitatively evaluates the spatial aggregation degree of the residual value: when the coefficient exceeds a preset threshold (the value range is 0.25 - 0.35, calibrated according to historical good product data), it indicates that there is a systematic error in this region (such as the annular ripple error caused by the eccentricity of the machine tool spindle); when the coefficient is lower than the threshold, it is determined as random noise (such as isolated point defects caused by dust). For example, in the detection of the lens edge region, if the residuals of 8 consecutive detection points show a radially increasing pattern and the spatial coefficient reaches 0.41, it is determined as a gradual change error caused by mold wear. This method draws on the principles of spatial statistics (similar to the Moran index) and can effectively distinguish between systematic process defects and random interference.

[0054] It should be noted that the spatial correlation analysis method adopted in this step is essentially an engineering application of the Moran's I in spatial statistics. Its core purpose is to accurately identify systematic process defects by quantifying the spatial pattern of the residual distribution. The Moran index can effectively distinguish two types of errors by calculating the similarity degree of the residuals of adjacent detection points (i.e., spatial autocorrelation): when the index is significantly high, it indicates that the residuals show spatial aggregation characteristics (such as continuous same-direction deviations in a certain fan-shaped area caused by mold eccentricity), suggesting the existence of traceable process defects; when the index is close to the random distribution level, it reflects that the error is an isolated random event (such as instantaneous dust interference). This method breaks through the limitations of the traditional threshold method (only focusing on whether the single-point residual exceeds the standard), and instead explores the process anomaly characteristics from the dimension of spatial correlation: for example, when an index value of 0.32 (threshold 0.25) is detected in the progressive channel area of the lens, it shows that there is a residual gradient change of 5 consecutive detection points in this region (average +0.15 D / mm), corresponding to the calibration error of the machine tool feed system.

[0055] In step S15, compensation analysis is performed based on the residual matrix, the error region, and the normal region to obtain a parameter compensation amount.

[0056] In one implementation, the following formula is used to calculate the parameter compensation amount for the normal region:

[0057] Where, represents the parameter compensation amount, represents the initial value of normal compensation, represents the natural constant, Represents the standard deviation of the residual matrix; The following formula is used to calculate the parameter compensation amount for the error region:

[0058] where, Represents the parameter compensation amount, Represents the initial error compensation value, Represents the natural constant, Represents the standard deviation of the residual matrix.

[0059] It should be noted that the compensation strategy in this step realizes the precise correction of process parameters through a differential regulation mechanism. Its core lies in dynamically adjusting the compensation intensity according to the regional defect type. For the normal region (dominated by random errors), a smaller normal compensation initial value (set to 30%-50% of the benchmark compensation amount) is adopted, and exponential decay adjustment is carried out in combination with the residual fluctuation coefficient: when the residual distribution is concentrated (the value of the standard deviation is small), the compensation amount approaches to optimize the performance, while when the residual dispersion is large (the value of the standard deviation is large), the compensation amplitude is automatically reduced to avoid introducing new errors due to overcorrection. For the error region (systematic defects), a larger error compensation initial value (set to 2-3 times the normal compensation initial value) is enabled, and a strong compensation and stability balance is achieved through the same decay mechanism.

[0060] It should be noted that the scientific nature of this design is reflected in: conforming to the error propagation theory, systematic defects require greater intervention to block the error chain; the exponential function realizes noise adaption, and automatic conservative adjustment in high-fluctuation scenarios; distinguishing between normal and error regions to avoid yield loss caused by "one-size-fits-all" compensation.

[0061] In one implementation, a dynamic compensation strategy based on spatial correlation strength can be introduced. For the error region, the step spatial correlation coefficient is incorporated into the compensation calculation system: the compensation gain factor formula is set as follows:

[0062] where, Represents the compensation gain factor, Represents the compensation coefficient, with a value range of 0.5-0.8 Represents the spatial correlation coefficient, Represents the preset spatial correlation coefficient threshold, with a value of 0.3.

[0063] The corrected compensation formula is as follows:

[0064] where, Represents the parameter compensation amount, Represents the initial error compensation value, denotes the natural constant, represents the standard deviation of the residual matrix, represents the compensation gain factor.

[0065] It should be noted that the innovation of this strategy lies in: quantifying the defect aggregation intensity using the spatial correlation coefficient. When the spatial correlation coefficient is significantly higher than the threshold (e.g., in the central area of the lens, I = 0.45 > the threshold 0.3), the value of the compensation gain factor rises to 1.2 - 1.5 times, enhancing the correction strength for strongly correlated defects (such as large - area curvature anomalies caused by mold edge collapse); retaining the standard deviation attenuation term to prevent over - compensation in high - fluctuation regions. In the near - vision compensation of progressive lenses, this method improves the correction accuracy of highly correlated errors (I > 0.4) by 28%, while reducing the ineffective compensation amount in the edge area by 15%. This hierarchical strengthening compensation mechanism not only maintains the stability advantage of the original method but also achieves targeted correction of systematic defects of different degrees.

[0066] In step S16, multi - objective optimization is performed according to the parameter compensation amount to obtain the optimal compensation parameters, and the lens is corrected according to the optimal compensation parameters.

[0067] In one implementation, lens design parameters and regularization parameters are obtained; according to the lens design parameters and the parameter compensation amount, a Jacobian matrix is constructed; error calculation is performed according to the Jacobian matrix to obtain an output error vector; the parameter compensation amount is updated according to the regularization parameter and the output error vector to obtain updated compensation parameters; the updated compensation parameters are used to replace the parameter compensation amount; when the norm of the output error vector is greater than a preset error threshold, the Jacobian matrix is reconstructed; when the norm of the output error vector is less than the preset error threshold, the updated compensation parameters are output as the optimal compensation parameters; the lens is corrected according to the optimal compensation parameters.

[0068] In one implementation, the Jacobian matrix is calculated by the following formula:

[0069] where, represents the element in the th row and th column of the Jacobian matrix, represents the th parameter compensation amount, represents the th lens design parameter, represents the partial derivative symbol.

[0070] It should be noted that in the core optimization stage of lens correction, the system first integrates the original lens design specifications and the stability control coefficient, and quantifies the adjustment effect by establishing a dynamic influence model between parameters. This model can accurately reflect the contribution degree of each design parameter change to the final compensation result. Based on this model, the deviation degree between the current compensation scheme and the ideal target is calculated, and the compensation parameters are iteratively corrected in combination with the stability constraint conditions. When the deviation exceeds the allowable range, the system automatically updates the influence model and recalculates until the deviation converges within the process accuracy requirements, and finally outputs the optimal adjustment scheme that meets multi-dimensional constraints.

[0071] In one implementation, the particle swarm optimization is used to optimize the compensation parameters. First, an initial particle swarm containing 50 - 100 parameter combinations is constructed. Each particle represents a solution vector containing 12-dimensional process parameters such as regional compensation intensity, grinding pressure distribution, and polishing dwell time. By integrating a high-precision optical detection module, 8 key indicators such as the surface curvature deviation and astigmatism distribution of the lens are obtained in real time as the input of the fitness function. Among them, the refractive accuracy weight in the distance vision area accounts for 40%, the smoothness of the progressive zone transition accounts for 30%, and the astigmatism control in the near vision area accounts for 30%. In each iteration, the particles dynamically adjust the parameter combinations according to the individual historical optimal solution and the global optimal solution of the group: the grinding pressure parameter is updated in steps of 0.05 N / time, and the dwell time is corrected with an accuracy of 0.1 s / mm². When the improvement amplitude of the particle swarm fitness is less than 0.5% for 10 consecutive generations, the termination condition is triggered and the optimal compensation parameters are output.

[0072] In summary, the present invention discloses an intelligent correction method based on the optical performance of functional lenses, aiming to improve the accuracy of the optical performance of lenses through a series of precise data processing and analysis steps. This method first involves obtaining the original lens data and preprocessing it to extract clear and accurate lens surface data. This process includes using a white light interference profiler in cooperation with a high-precision five-axis adjustment table for data acquisition, and passing through a data preprocessing process of multiple core links to ensure that subsequent calculations are based on high-quality data. Then, based on the lens surface data, surface parameters are calculated to obtain three-dimensional surface parameters, which describe the geometric characteristics of the lens surface in detail and provide a basis for in-depth analysis. Specifically, by calculating key indicators such as the lens power, sagittal height, and free form parameters, the design and manufacturing conditions of the lens can be comprehensively understood. Subsequently, a residual matrix is generated by weighted residual calculation, which quantifies the deviation between the actual surface and the ideal design and provides a quantitative basis for identifying the errors generated in the manufacturing process.

[0073] After obtaining the three-dimensional surface parameters and the residual matrix, the method further performs spatial correlation analysis to determine the regional manufacturing errors. In this process, the lens surface is divided into several sub-regions according to process characteristics, and the spatial correlation coefficient of the data points within each region is calculated to evaluate the presence or absence of errors. When the spatial correlation coefficient exceeds a certain threshold, the corresponding sub-region is marked as having manufacturing errors; otherwise, it is regarded as a normal region. This spatial correlation analysis not only helps to accurately locate the specific position and the influence range of manufacturing defects, but also can improve the correction accuracy of specific regions and the correction efficiency for other regions. Next, compensation analysis is carried out based on the residual matrix, the error regions, and the normal regions to calculate the optimal adjustment value, that is, the parameter compensation amount, which can offset the manufacturing errors. Different formulas are used to calculate the parameter compensation amount for the normal regions and the error regions respectively, reflecting the importance of adopting different compensation strategies for different types of errors.

[0074] Finally, the multi-objective optimization algorithm is used to optimize the obtained parameter compensation amount to obtain the optimal compensation parameters, and the lens is corrected accordingly. The multi-objective optimization process involves constructing a Jacobian matrix, obtaining the output error vector through error calculation, and updating the parameter compensation amount according to the regularization parameter and the output error vector until the predetermined error threshold is reached. This method not only improves the quality of the lens, but also enhances the consistency and reliability of the manufacturing process. The entire process strictly follows mathematical and physical principles, such as numerical analysis, linear algebra, optimization theory, etc., to ensure the scientificity and rationality of each step. In particular, through the application of spatial correlation analysis and multi-objective optimization techniques, the best solution can be found even in a complex and changeable manufacturing environment, greatly improving the precision of lens manufacturing.

[0075] Referring to Figure 2 , the second embodiment of the present invention provides an intelligent correction system for the optical performance of functional lenses, including: A data acquisition module, configured to acquire the original lens data and preprocess the original lens data to obtain the lens surface data; A surface parameter module, configured to calculate surface parameters according to the lens surface data to obtain three-dimensional surface parameters; A residual matrix module, configured to calculate the weighted residuals according to the three-dimensional surface parameters to obtain a residual matrix; A spatial analysis module, configured to perform spatial correlation analysis according to the three-dimensional surface parameters and the residual matrix to obtain error regions and normal regions; A compensation analysis module, configured to perform compensation analysis according to the residual matrix, the error regions, and the normal regions to obtain a parameter compensation amount; An optimization and correction module is used to perform multi-objective optimization based on the parameter compensation amount to obtain the optimal compensation parameters, and correct the lens according to the optimal compensation parameters.

[0076] Preferably, the compensation analysis module, as the core unit of this system, mainly undertakes the dual functions of error conversion and process control: First, it combines the regional manufacturing errors (such as the curvature deviation in the distance vision area, the abrupt change value in the progressive zone transition, the astigmatism distribution in the near vision area, etc.) collected by the optical detection equipment with the parameter correlation degree characterized by the residual matrix, and quantifies the influence weight of each processing parameter on the final optical performance by establishing a three-dimensional error transfer model (for example, a certain type of lens uses a 9×12 residual matrix corresponding to 108 process control nodes). This module uses an adaptive weighting algorithm to dynamically allocate the compensation priorities of different process parameters on the premise of ensuring the accuracy priority of the key visual area (such as the area corresponding to the pupil). For example, when it is detected that the astigmatism in the near vision area exceeds the standard, the system will automatically increase the compensation weight of the polishing dwell time to 65%, and at the same time reduce the adjustment range of the grinding pressure in the non-critical area.

[0077] Preferably, the optimization and correction module completes parameter conversion through the collaborative operation of the numerical control machining center and the real-time monitoring module: First, map the optimized compensation parameters to the motion trajectory of the five-axis precision grinding equipment, and convert the nanoscale compensation amount into the grinding wheel feed amount through the piezoelectric ceramic drive system. For example, for a curvature deviation of 0.3μm in the near vision area, the system automatically generates a process package including the spindle speed (set at 12,000 - 15,000 rpm), the feed rate (0.8 - 1.2 mm / s), and the coolant flow rate (200 - 250 mL / min). During the processing, the white light interferometer integrated on the spindle collects the surface topography data of the lens every 30 seconds, and triggers dynamic compensation through real-time comparison with the target surface: When it is detected that the local residual deviation exceeds 5μm, the system immediately adjusts the grinding pressure in the corresponding area (controlled within the range of 0.5 - 2.0 N), and synchronously optimizes the coolant injection angle to prevent material thermal deformation. After rough grinding, enter the ion beam polishing stage, and use the dwell time distribution map (with an accuracy of 0.1 second / mm²) generated by the compensation parameters to perform sub-micron shaping on the transition area. The entire process realizes the two-way traceability of process parameters and quality data through the MES system, ensuring that the execution error of the compensation parameters for each lens is less than ±0.8%. The measured data shows that this process reduces the astigmatism deviation in the near vision area of progressive lenses from ±0.25 D to ±0.12 D, and the yield rate is increased to 98.3%.

[0078] It should be noted that an intelligent correction system for the optical performance of a functional lens provided in an embodiment of the present invention is used to execute all the process steps of an intelligent correction method for the optical performance of a functional lens in the above embodiment. Their working principles and beneficial effects correspond one by one, so they will not be elaborated here.

[0079] An embodiment of the present invention further provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a data acquisition program. When the processor executes the computer program, it implements the steps in the embodiments of the intelligent correction method for the optical performance of each of the above functional lenses, such as Figure 1 the step S11 shown. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the above device embodiments, such as the data acquisition module.

[0080] Exemplarily, the computer program may be divided into one or more modules / units. The one or more modules / units are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the electronic device.

[0081] The electronic device may be a computing device such as a desktop computer, a notebook, a palm computer, and a smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above components are only examples of the electronic device and do not constitute a limitation on the electronic device. It may include more or fewer components than the above, or combine some components, or different components. For example, the electronic device may further include input / output devices, network access devices, a bus, etc.

[0082] The so-called processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the electronic device and connects various parts of the entire electronic device through various interfaces and lines.

[0083] The memory can be used to store the computer program and / or modules. By running or executing the computer program and / or modules stored in the memory, and by invoking the data stored in the memory, the processor implements various functions of the electronic device. The memory mainly includes a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created according to the use of the mobile phone (such as audio data, phone book, etc.). In addition, the memory can include high-speed random access memory, and can also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage devices.

[0084] Among them, if the modules / units integrated in the electronic device are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above-mentioned embodiment methods of the present invention, it can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0085] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the attached drawings of the device embodiments provided by the present invention, the connection relationship between modules indicates that they have a communication connection, which can be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0086] The specific embodiments described above further elaborate on the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only for the specific embodiments of the present invention and is not used to limit the protection scope of the present invention. In particular, for those skilled in the art, any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An intelligent correction method for the optical performance of a functional lens, characterized in that: include: Acquiring original lens data, and preprocessing the original lens data to obtain lens surface data; Calculate surface parameters according to the lens surface data to obtain three-dimensional surface parameters; Performing weighted residual calculation according to the three-dimensional surface parameters to obtain a residual matrix; Perform spatial correlation analysis based on the three-dimensional surface parameters and the residual matrix to obtain an error area and a normal area; Perform compensation analysis according to the residual matrix, the error region and the normal region to obtain a parameter compensation amount; Multi-objective optimization is performed according to the parameter compensation amount to obtain the optimal compensation parameter, and the lens is corrected according to the optimal compensation parameter.

2. The intelligent correction method for the optical performance of functional lenses according to claim 1, characterized in that: The step of calculating surface parameters according to the lens surface data to obtain three-dimensional surface parameters includes: The optical power of the lens is calculated using the following formula: in, Indicates the optical power of the lens. Represents the refractive index of the lens, Indicates the radius of curvature of the front surface of the lens, Indicates the radius of curvature of the rear surface of the lens; The lens sagitta is calculated using the following formula: in, Indicates the lens sag, Indicates the actual diameter of the lens; The free shape parameters are calculated by the following formula: in, Indicates Stage The free shape parameters of the term, represents the radial order, represents pi, Expressed in polar coordinates The wavefront error distribution under Indicates Stage The free-form polynomial of the term, represents the normalized radial distance, represents the azimuth, express The differential symbol of express The differential symbol of ; The three-dimensional surface parameters include the lens optical power, the lens sagittal height and the free shape parameters.

3. The intelligent correction method for the optical performance of functional lenses according to claim 1, characterized in that: The weighted residual calculation is performed according to the three-dimensional surface parameters to obtain a residual matrix, including: Obtain theoretical parameter matrix and residual weight matrix; Subtracting the theoretical parameter matrix from the three-dimensional surface parameters to obtain an original residual matrix; The residual matrix is ​​calculated according to the original residual matrix and the residual weight matrix by the following formula: in, represents the residual matrix, represents the original residual matrix, Represents the square root matrix of the residual weight matrix.

4. The intelligent correction method for the optical performance of functional lenses according to claim 1, characterized in that: The performing of spatial correlation analysis according to the three-dimensional surface parameters and the residual matrix to obtain the error region and the normal region comprises: Dividing the three-dimensional surface parameters into blocks according to a preset number of regions to obtain block region data; The spatial correlation coefficient is calculated according to the block area data using the following formula: in, represents the spatial correlation coefficient, Represents the total number of data points in a block area. represents the data point number, express data points and The spatial weight coefficient of the data point is express The residual of the data point, express The residual of the data point, represents the mean value of the residual matrix; When the spatial correlation coefficient is greater than a preset spatial threshold, the corresponding block area data is marked as having a regional manufacturing error, and the corresponding block area is marked as an error area; When the spatial correlation coefficient is less than a preset spatial threshold, the corresponding block area is marked as a normal area.

5. The intelligent correction method for the optical performance of functional lenses according to claim 1, characterized in that: The performing compensation analysis according to the residual matrix, the error region and the normal region to obtain a parameter compensation amount includes: The following formula is used to calculate the parameter compensation amount for the normal area: in, Indicates the parameter compensation amount, Indicates the normal compensation initial value, represents a natural constant, represents the standard deviation of the residual matrix; The following formula is used to calculate the parameter compensation amount for the error area: in, Indicates the parameter compensation amount, represents the initial value of error compensation, represents a natural constant, represents the standard deviation of the residual matrix.

6. The intelligent correction method for the optical performance of a functional lens according to claim 1, characterized in that: The multi-objective optimization is performed according to the parameter compensation amount to obtain the optimal compensation parameter, and the lens is corrected according to the optimal compensation parameter, including: Obtain lens design parameters and regularization parameters; Constructing a Jacobian matrix according to the lens design parameters and the parameter compensation amount; Perform error calculation according to the Jacobian matrix to obtain an output error vector; updating the parameter compensation amount according to the regularization parameter and the output error vector to obtain an updated compensation parameter; Replacing the parameter compensation amount with the updated compensation parameter; When the modulus of the output error vector is greater than a preset error threshold, reconstructing the Jacobian matrix; When the modulus of the output error vector is less than a preset error threshold, outputting the updated compensation parameter as an optimal compensation parameter; The lens is corrected according to the optimal compensation parameters.

7. The intelligent correction method for the optical performance of a functional lens according to claim 6, characterized in that: The step of constructing a Jacobian matrix according to the lens design parameters and the parameter compensation amount comprises: The Jacobian matrix is ​​calculated by the following formula: in, The Jacobian matrix Row, No. The elements of the column, Indicates Parameter compensation amount, Indicates lens design parameters, Represents the sign of partial derivative.

8. An intelligent correction system for the optical performance of functional lenses, characterized in that: include: A data acquisition module is used to acquire original lens data and pre-process the original lens data to obtain lens surface data; A surface parameter module, used to calculate surface parameters according to the lens surface data to obtain three-dimensional surface parameters; A residual matrix module, used for performing weighted residual calculation according to the three-dimensional surface parameters to obtain a residual matrix; A spatial analysis module, used for performing spatial correlation analysis based on the three-dimensional surface parameters and the residual matrix to obtain regional manufacturing errors; A compensation analysis module, used for performing compensation analysis according to the residual matrix and the regional manufacturing error to obtain a parameter compensation amount; The optimization and correction module is used to perform multi-objective optimization according to the parameter compensation amount to obtain the optimal compensation parameter, and correct the lens according to the optimal compensation parameter.

9. An electronic device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the intelligent correction method for the optical performance of a functional lens as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the intelligent correction method for the optical performance of a functional lens as described in any one of claims 1 to 7.

Images