Design method for personalized customization of out-of-focus lens and lens

Abstract

The invention discloses a design method for personalized customization of an out-of-focus lens and a lens, and relates to the related technical field of lens design, and the method comprises the steps: detecting diopter distribution information and cylindricity distribution information of the retina of a user; performing mirror projection analysis according to diopter distribution information and cylindrical distribution information to obtain mirror projection data; setting boundary conditions through a user side; processing boundary conditions and the mirror projection data through a polar coordinate configuration model to generate lens polar coordinate design parameters; and sending the lens polar coordinate design parameters to a lens design end. The technical problems that in the prior art, due to the fact that standard lens design cannot meet the special diopter and cylindricity distribution requirements of individual users, the lens design precision and individuation degree are insufficient, and the myopia correction effect is poor are solved, and the technical effects of improving the lens design precision, individuation and the myopia correction effect are achieved.

Description

Technical Field

[0001] The present application relates to the technical field related to lens design, and in particular to a design method and lens for a personalized customized defocus lens. Background Art

[0002] In the process of designing customized defocus lenses, how to design according to the unique eye refraction and visual needs of each user is a key technical challenge. The current defocus lens design mainly relies on standardized refraction (i.e. the degree of myopia and hyperopia of the eye) and simple spherical lens design. Although it can meet the vision needs of most people, it cannot effectively meet the personalized vision needs of some users, especially for users with complex retinal refraction and cylindrical distribution, such as patients with high myopia, astigmatism or presbyopia. With the development of medical imaging technology and computer vision technology, it is possible to obtain detailed refractive power distribution information and cylindrical power distribution information about the user's retina through more accurate eye imaging methods, reflecting the differences in refractive characteristics of various areas of the eyeball at different positions. However, traditional design methods ignore the subtle differences in eye refraction in different areas, and it is difficult to convert these complex eye refraction data into practical and accurate lens design parameters, resulting in the visual effect of the lens may not fully meet the needs of individual users, thus affecting the accuracy and comfort of vision correction.

[0003] At the current stage, relevant technologies have the problem that standardized lens design cannot meet the special refractive power and cylindrical power distribution requirements of individual users, resulting in technical problems such as insufficient lens design accuracy and personalization, and poor myopia correction effect. Summary of the invention

[0004] The present application provides a design method and lens for personalized customized defocus lenses, thereby solving the technical problem in the prior art that the standardized lens design cannot meet the special refractive power and cylindrical power distribution requirements of individual users, resulting in insufficient lens design accuracy and personalization and poor myopia correction effect, thereby achieving the technical effect of improving lens design accuracy, personalization and myopia correction effect.

[0005] The present application provides a design method for a personalized through-focus lens, comprising: detecting the diopter distribution information and the cylindrical power distribution information of the user's retina; performing mirror projection analysis based on the diopter distribution information and the cylindrical power distribution information to obtain mirror projection data; setting boundary conditions through a user end; processing the boundary conditions and the mirror projection data through a polar coordinate configuration model to generate polar coordinate design parameters of the lens; and sending the polar coordinate design parameters of the lens to a lens design end.

[0006] In a possible implementation, the method for designing personalized customized defocus lenses also performs the following processing: constructing a plane coordinate system with the user's pupil as the origin, the horizontal direction as the first coordinate axis, and the vertical direction as the second coordinate axis; using an image collector, based on the plane coordinate system, scanning the pupil at 0°, 45°, 90° and 135°, respectively, to obtain local refractive power distribution information and local cylindrical power distribution information; inputting the local refractive power distribution information into a refractive power interpolation analysis model, and outputting the refractive power distribution information; inputting the local cylindrical power distribution information into a cylindrical power interpolation analysis model, and outputting the cylindrical power distribution information.

[0007] In a possible implementation, the design method for personalized customized defocus lenses also performs the following processing: the refractive power interpolation analysis model construction step includes: respectively collecting the local distribution data sets of refractive power records of 0°, 45°, 90° and 135° of the user's eye pupil set; traversing the local distribution data sets of refractive power records to perform interpolation calculations to obtain a global distribution data set of refractive power identification; using the global distribution data set of refractive power identification as supervision data and the local distribution data set of refractive power records as input data to train the refractive power interpolation analysis model; the cylinder interpolation analysis model construction step includes: respectively collecting the local distribution data sets of cylinder records of 0°, 45°, 90° and 135° of the user's eye pupil set; traversing the local distribution data sets of cylinder records to perform interpolation calculations to obtain a global distribution data set of cylinder identification; using the global distribution data set of cylinder identification as supervision data and the local distribution data set of cylinder records as input data to train the cylinder interpolation analysis model.

[0008] In a possible implementation, the design method for a personalized customized defocus lens further performs the following processing: obtaining a projection function:

[0009] h i = L·tan ( ang i ) ;

[0010] z i =D i ;

[0011] Among them, i Characterizes the i-th scanning angle, h i Characterizes the distance between the lens position and the eye axis at the i-th scanning angle, z i Characterizes the diopter or cylinder projected onto the mirror surface, D i Characterize the detection diopter or cylinder of the i-th scanning angle, L characterizes the distance between the center of the mirror surface and the eye axis; construct the mirror projection completion function:

[0012]

[0013] Among them, D ( x ) Characterizes equivalent diopter or equivalent cylinder, x 0 is the maximum boundary position of the instrument measurement result, x 1 is the actual maximum boundary position of the lens, D 0 is the boundary x 0 The actual measured diopter or cylinder at the position, x represents the position of the non-projection area of ​​the lens; when the lens position belongs to the projection area, the diopter distribution information and the cylinder distribution information are processed according to the projection function to obtain the first mirror projection data; when the lens position belongs to the non-projection area, the diopter distribution information and the cylinder distribution information are processed according to the mirror projection complement function to obtain the second mirror projection data; the first mirror projection data and the second mirror projection data are added to the mirror projection data.

[0014] In a possible implementation, the method for designing a personalized customized through-focus lens further performs the following processing: obtaining mirror projection recording data and boundary condition recording data; identifying lens polar coordinate identification data for the mirror projection recording data and the boundary condition recording data; and training the polar coordinate configuration model using the lens polar coordinate identification data as supervision and the mirror projection recording data and the boundary condition recording data as input.

[0015] In a possible implementation, the method for designing a personalized through-focus lens further performs the following processing: obtaining a polar coordinate identification function; and processing the mirror projection recording data and the boundary condition recording data identification according to the polar coordinate identification function to obtain the lens polar coordinate identification data.

[0016] In a possible implementation, the method for designing personalized customized through-focus lenses further performs the following processing: using the lens polar coordinate identification data as supervision and the mirror projection recording data and the boundary condition recording data as input to obtain a polar coordinate configuration model construction data set; based on the polar coordinate configuration model construction data set, training a regression decision tree to obtain a first polar coordinate configuration model, wherein the data weights of any group of the polar coordinate configuration model construction data set are the same; obtaining a first polar coordinate configuration model construction data set in which the output accuracy of the first polar coordinate configuration model is less than or equal to an output accuracy threshold; when the amount of data in the first polar coordinate configuration model construction data set is less than or equal to a data amount threshold, setting the first polar coordinate configuration model to the polar coordinate configuration model.

[0017] In a possible implementation, the method for designing personalized customized through-focus lenses further performs the following processing: when the data amount of the first polar coordinate configuration model construction data set is greater than the data amount threshold, increasing the weight of the first polar coordinate configuration model construction data set in the polar coordinate configuration model construction data set to obtain a secondary polar coordinate configuration model construction data set; training a regression decision tree based on the secondary polar coordinate configuration model construction data set to obtain a second polar coordinate configuration model and then performing a loop analysis until the data amount of the Nth polar coordinate configuration model construction data set is less than or equal to the data amount threshold, merging the first polar coordinate configuration model, the second polar coordinate configuration model, and the Nth polar coordinate configuration model to obtain the polar coordinate configuration model.

[0018] The present application also provides a lens designed by any one of the design methods for personalized customized defocus lenses.

[0019] The invention proposes a design method and lens for customized defocus lenses, detects the diopter distribution information and cylindrical distribution information of the user's retina; performs mirror projection analysis based on the diopter distribution information and the cylindrical distribution information to obtain mirror projection data; sets boundary conditions through the user end; processes the boundary conditions and the mirror projection data through the polar coordinate configuration model to generate polar coordinate design parameters of the lens; and sends the polar coordinate design parameters of the lens to the lens design end. The invention solves the technical problem that the standardized lens design in the prior art cannot meet the special diopter and cylindrical distribution requirements of individual users, resulting in insufficient lens design accuracy and personalization, and poor myopia correction effect, and achieves the technical effect of improving lens design accuracy, personalization and myopia correction effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0021] Figure 1 A schematic diagram of a flow chart of a method for designing a personalized customized defocus lens provided in an embodiment of the present application.

[0022] Figure 2 A schematic diagram of a flow chart for outputting diopter and cylinder distribution information in a design method for a personalized customized defocus lens provided in an embodiment of the present application.

[0023] Figure 3 A plane distribution diagram of grid points in a polar coordinate system in a design method for a personalized customized defocus lens provided in an embodiment of the present application. DETAILED DESCRIPTION

[0024] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0025] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0026] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of the objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, product, or server comprising a series of steps is not necessarily limited to those steps clearly listed, but may include other steps that are not clearly listed or inherent to these processes, methods, products, or devices, and unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field to which this application belongs. The terms used herein are for the purpose of describing the embodiments of this application only.

[0027] Example 1

[0028] The present application embodiment provides a design method for a personalized customized defocus lens, such as Figure 1 As shown, the method includes:

[0029] Step S100, detecting the diopter distribution information and the cylinder distribution information of the user's retina.

[0030] Preferably, a high-precision CCD (charge-coupled device) scanning device is used to non-invasively scan the user's retina to detect the refractive power distribution information and cylindrical power distribution information of the user's retina, that is, the user's eye is scanned within a range of ±25 degrees with the pupil as the center to obtain detailed refractive parameters of the inside of the user's eye, especially the retinal area. Specifically, the refractive power distribution information refers to the distribution of the refractive power (i.e., the eye's ability to focus light) of different areas on the retina. The refractive power is usually expressed in diopters (D). Different human eyes may have different refractive power in different areas (such as the central area, the peripheral area, etc.). The eye has different refractive powers in different domains, especially in some special cases, such as astigmatism or high myopia. Through high-precision CCD scanning equipment, these differences can be recorded in detail to obtain complete refractive power distribution data; cylinder refers to the astigmatism of the eye, which is used to describe the aspherical refractive problem of the eye. Different from conventional refractive power, cylinder reflects the difference in refraction in different directions of the eye. High-precision CCD scanning equipment can obtain cylinder data of different areas of the eye, especially the cylinder distribution on the retina (cylinder distribution information), which helps to design lenses to correct astigmatism more accurately.

[0031] Preferably, a CCD scanning device refers to a scanning device using charge coupled device (CCD) technology, which is widely used in image acquisition and precise measurement. In ophthalmology and eyewear design, a CCD scanning device is used to obtain high-precision eye images, especially to detect the diopter distribution and cylindrical distribution of the user's eye retina. CCD is a semiconductor technology that can convert light signals into electrical signals. It is widely used in cameras, scanners, microscopes and other imaging devices. It converts the light entering the light source (such as the light of the user's eyes) into electric charges through an array of photosensitive elements (usually photodiodes), and then transmits and converts these charges in sequence. It is a digital signal, thereby generating clear images or data. Specifically, in ophthalmic diagnosis, CCD scanning equipment can perform high-resolution scans of the eye through precise imaging technology, especially detailed imaging of the retina and other areas of the eye. By scanning the refractive power of different areas of the eyeball, the CCD scanning equipment can obtain the refractive power distribution information of the eye, which reflects how different parts of the eye refract light and helps diagnose refractive errors such as myopia and hyperopia. For patients with astigmatism (cylinder), the CCD scanning equipment can obtain the cylinder information of different areas on the retina, which helps to design glasses more accurately and compensate for the refractive errors of the retina in different directions.

[0032] Further, such as Figure 2As shown, step S100 also includes step S110, constructing a plane coordinate system with the user's eye pupil as the origin, the horizontal direction as the first coordinate axis, and the vertical direction as the second coordinate axis; step S120, using an image collector, based on the plane coordinate system, scanning the eye pupil at 0°, 45°, 90° and 135° respectively, to obtain local refractive power distribution information and local cylindrical power distribution information; step S130, inputting the local refractive power distribution information into a refractive power interpolation analysis model, and outputting the refractive power distribution information; step S140, inputting the local cylindrical power distribution information into a cylindrical power interpolation analysis model, and outputting the cylindrical power distribution information.

[0033] Preferably, the user's eye pupil is used as the coordinate origin to indicate that all measurement data are performed relative to the position of the pupil, the horizontal direction is the first coordinate axis (X axis), and the vertical direction is the second coordinate axis (Y axis), which is used to describe the distribution of the refractive power and cylindrical power of the eyeball in different directions. The image collector (such as a CCD scanner) is used to scan in the following four directions, including 0° (horizontal), scanning from the front of the eye, measuring the eyeball on the horizontal plane, 45°, scanning from the upper right at an angle of 45°, capturing the eyeball The refractive information of the upper right area of ​​the eyeball is scanned at 90° (vertical) from directly above the eye, and the eyeball is measured in the vertical direction. At 135°, the refractive information of the upper left area of ​​the eyeball is captured from a 45° angle from the upper left. The refractive power (myopia, hyperopia, etc.) and the degree and direction of astigmatism in each direction are collected as local refractive power distribution information and local cylindrical information, respectively providing the refractive state of different parts of the eyeball and the astigmatism state of the eyeball at different angles (0°, 45°, 90° and 135°). The measurement results of an example are shown in the following table:

[0034] angle -25 -24 -23 -22 -21 -20 -19 -18 -17 Diopter -5.44 -5.44 -5.47 -5.42 -5.44 -5.61 -5.53 -5.57 -5.65 Cylinder -0.78 -0.73 -0.69 -0.74 -0.71 -0.53 -0.49 -0.51 -0.45

[0035] angle -16 -15 -14 -13 -12 -11 -10 -9 -8 Diopter -5.71 -5.65 -5.71 -5.71 -5.79 -5.73 -5.74 -5.72 -5.81 Cylinder -0.42 -0.42 -0.33 -0.35 -0.24 -0.26 -0.21 -0.21 -0.18

[0036] angle -7 -6 -5 -4 -3 -2 -1 0 1 Diopter -5.89 -5.91 -5.93 -5.97 -6.06 -6.13 -6.19 -6.24 -6.05 Cylinder -0.13 -0.17 -0.19 -0.17 -0.20 -0.18 -0.16 -0.18 -0.19

[0037] angle 2 3 4 5 6 7 8 9 10 Diopter -6.08 -6.11 -6.14 -6.18 -6.08 -6.12 -6.16 -6.20 -6.09 Cylinder -0.22 -0.26 -0.31 -0.23 -0.22 -0.18 -0.28 -0.29 -0.31

[0038] angle 11 12 13 14 15 16 17 18 19 Diopter -6.08 -6.03 -5.97 -5.92 -5.87 -5.92 -5.96 -6.00 -6.09 Cylinder -0.35 -0.45 -0.56 -0.67 -0.83 -0.81 -0.80 -0.81 -0.79

[0039] angle 20 21 22 23 24 25 Diopter -6.18 -6.08 -6.02 -5.97 -5.96 -5.99 Cylinder -0.75 -0.80 -0.81 -0.81 -0.97 -0.98

[0040] Preferably, the local refractive power distribution information is input into the refractive power interpolation analysis model for processing, and finally the complete refractive power distribution information is output. Specifically, the refractive power interpolation analysis model mathematically processes these local refractive power data, fills in the gaps between different angles, and obtains the refractive power distribution of the entire eyeball surface, including calculating the refractive power of different parts of the eyeball according to the refractive power data of each scanning point (using mathematical formulas to ensure that the refractive power of each point on the eyeball surface is accurately estimated), expanding the local information to the entire area of ​​the eye, and estimating the refractive power distribution of other areas that are not directly scanned. Output complete diopter distribution information, that is, detailed changes in diopter (myopia, hyperopia) in the entire area of ​​the eyeball; similarly, input local cylinder distribution information into the cylinder interpolation analysis model for processing, and output complete cylinder distribution information. The cylinder interpolation analysis model mathematically calculates these local cylinder data to generate the cylinder distribution of the entire eyeball surface, including receiving astigmatism information obtained at different scanning angles, and using the interpolation algorithm to calculate the cylinder distribution of other areas based on known data, fill in the gaps between scanning points, and finally output the cylinder information of the entire eyeball surface in different directions. Then, according to this distribution information, personalized eyeglass lens design is carried out to accurately correct the vision defects in each area and improve visual quality and comfort.

[0041] Furthermore, the step of constructing the refractive interpolation analysis model in step S130 includes step S131, respectively collecting local distribution data sets of refractive records of 0°, 45°, 90° and 135° of the user's eye pupil set; step S132, traversing the local distribution data sets of refractive records to perform interpolation calculations to obtain a global distribution data set of refractive identification; and step S133, using the global distribution data set of refractive identification as supervision data and the local distribution data set of refractive records as input data to train the refractive interpolation analysis model.

[0042] Preferably, local distribution data sets of refractive power records at 0°, 45°, 90° and 135° of the user's eye pupil refractive power distribution information are collected respectively, including refractive information of different areas on the retina or cornea (such as myopia, hyperopia, refractive error, etc.), that is, these data describe the refractive state of different parts of the eye, and then traverse the refractive power data sets obtained from different directions (0°, 45°, 90°, 135°) and perform interpolation calculations to fill in the blank points between the local data, that is, through a mathematical interpolation algorithm (such as linear interpolation, spline interpolation, etc.), the refractive power of other unscanned areas of the eye is inferred based on the known local refractive power information, and the refractive power of the unscanned area is inferred, and finally a complete refractive power distribution data set is generated. (i.e., a global distribution dataset of refractive identification), which contains the refractive information of each position of the entire eye surface (or retina), and then uses a supervised learning method to train the refractive interpolation analysis model through input data (local distribution dataset of refractive records) and supervised data (global distribution dataset of refractive identification). Specifically, the input data and supervised data are input into the model as training data, and the model gradually learns how to infer the global refractive distribution based on the local refractive information by learning the relationship between these data. Through continuous optimization, the model can accurately infer the refractive distribution of the entire eye from the local data, and adjust the model parameters through training to improve the accuracy of the prediction, thereby obtaining a refractive interpolation analysis model.

[0043] Furthermore, step S140 is a step of constructing a cylinder interpolation analysis model, comprising step S141, respectively collecting local distribution data sets of cylinder records of 0°, 45°, 90° and 135° of the user's eye pupil set; step S142, traversing the local distribution data sets of cylinder records to perform interpolation calculations to obtain a global distribution data set of cylinder identification; step S143, using the global distribution data set of cylinder identification as supervision data and the local distribution data set of cylinder records as input data to train the cylinder interpolation analysis model.

[0044] Preferably, local distribution data sets of the cylinder records of 0°, 45°, 90° and 135° of the user's eye pupil cylinder distribution information are collected respectively to indicate the degree and direction of the eye's astigmatism at different angles, that is, these data represent the astigmatism characteristics of a certain part of the eye, and then the cylinder data sets obtained from different directions (0°, 45°, 90°, 135°) are traversed and interpolated to fill in the blank points between the local data, that is, through a mathematical interpolation algorithm (such as linear interpolation, spline interpolation, etc.), the cylinder of other unscanned areas of the eye is inferred based on the known local cylinder information, and finally a complete cylinder distribution data set (that is, a global distribution data set of cylinder identifiers) is generated. ), contains the cylinder information of each position of the entire eye surface (or retina), and then uses the supervised learning method to train the cylinder interpolation analysis model through input data (cylinder record local distribution data set) and supervision data (cylinder identifier global distribution data set). Specifically, the input data and supervision data are input into the model as training data. The model gradually learns how to infer the global cylinder distribution based on the local cylinder information by learning the relationship between these data. Through continuous optimization, the model can accurately infer the cylinder distribution of the entire eye from the local data, and adjust the model parameters through training to improve the accuracy of the prediction, and obtain the cylinder interpolation analysis model. Through the interpolation analysis model, the astigmatism and refractive power of the user's eye in different areas are comprehensively considered, and the lens is designed according to the specific situation of the user's eye, so as to provide a more comfortable and accurate vision correction effect.

[0045] Step S200: performing mirror projection analysis according to the refractive power distribution information and the cylindrical power distribution information to obtain mirror projection data.

[0046] Preferably, performing mirror projection analysis on the refractive power distribution information and the cylindrical power distribution information means projecting the refractive power and cylindrical power information of the eye onto a virtual two-dimensional plane (mirror), thereby generating projection data that can be used to design lenses, which is similar to projecting the shape of a three-dimensional object onto a plane to simplify and analyze its characteristics. Specifically, the refractive power distribution (or cylindrical power distribution) of the eye is mapped onto a plane in a certain way, and the curvature and refractive change of the eyeball will be mapped onto a mirror surface. For the refractive power distribution information, the optical properties of different areas on the surface of the eyeball can be described. For example, the center of the eyeball may have a stronger refractive power, while the peripheral area may have a weaker refractive power. For the cylindrical power distribution information, the mirror projection will reflect the direction and intensity of astigmatism. By By mapping the cylindrical power data in different directions to a plane, the astigmatism characteristics of the entire eye can be seen and the lens design can be adjusted accordingly; then coordinate conversion is performed, assuming that the eyeball is a three-dimensional object, and mirror projection converts these three-dimensional refractive power or cylindrical power data into two-dimensional data on the plane. For example, the refractive power distribution of various parts of the eyeball (which may vary along different coordinate axes) will be converted into points or curves on a two-dimensional image, and then the refractive power or cylindrical power distribution in the three-dimensional space is mapped to the plane through data mapping to form mirror projection data, which can be used to generate lens design parameters and can help lens design software accurately calculate the refractive power and astigmatism of different areas, and ultimately generate a suitable lens shape and curvature to optimize the user's visual experience.

[0047] Furthermore, step S200 further includes step S210, obtaining a projection function:

[0048] h i = L·tan(ang i );

[0049] z i =D i ;

[0050] Among them, i Characterizes the i-th scanning angle, h i Characterizes the distance between the lens position and the eye axis at the i-th scanning angle, z i Characterizes the diopter or cylinder projected onto the mirror surface, D i represents the detection diopter or cylinder of the i-th scanning angle, and L represents the distance between the center of the mirror surface and the eye axis; step S220, constructing a mirror projection completion function:

[0051]

[0052] Among them, D ( x ) Characterizes equivalent diopter or equivalent cylinder, x 0is the maximum boundary position of the instrument measurement result, x 1 is the actual maximum boundary position of the lens, D 0 is the boundary x 0 The actual measured diopter or cylinder at the position, x represents the position of the non-projection area of ​​the lens; step S230, when the lens position belongs to the projection area, the diopter distribution information and the cylinder distribution information are processed according to the projection function to obtain the first mirror projection data; step S240, when the lens position belongs to the non-projection area, the diopter distribution information and the cylinder distribution information are processed according to the mirror projection complement function to obtain the second mirror projection data; step S250, the first mirror projection data and the second mirror projection data are added to the mirror projection data.

[0053] Preferably, through the projection function: h i = L·tan(ang i ), z i =D i , project the diopter distribution information onto the actual lens for lens design, directly project the data onto the tangent plane at the vertex of the front surface of the lens, and set point (h i , z i ) is a point on the tangent plane of the lens, where ang i Characterizes the i-th scanning angle, h i Characterizes the distance between the lens position and the eye axis at the i-th scanning angle, z i Characterizes the diopter or cylinder projected onto the mirror surface, D i Characterize the detected diopter or cylinder at the i-th scanning angle, L characterizes the distance between the center of the mirror surface and the eye axis, and then the measured diopter distribution information data is projected onto the actual lens.

[0054] Preferably, only the distribution of the diopter or cylinder in a small part of the middle of the lens can be determined through the mirror projection analysis, and there is still a large area on the lens that is not defined, so the data outside the projection area is completed to complete the lens design normally. Specifically, the mirror projection completion function is constructed as follows:

[0055]

[0056] Among them, D ( x ) Characterizes equivalent diopter or equivalent cylinder, x 0 is the maximum boundary position of the instrument measurement result, x 1 is the actual maximum boundary position of the lens, D 0 is the boundary x 0The actual measured diopter or cylinder at , x represents the position of the non-projection area of ​​the lens, and the measured data is expanded to the entire lens through the mirror projection completion function.

[0057] Preferably, if the lens position belongs to the projection area (the surface part of the lens), the diopter distribution information and the cylinder distribution information are processed according to the projection function, that is, the projection function calculates the optical properties that the lens surface should have according to the lens position (that is, the coordinates of the projection area) and the diopter and cylinder data of the eye, and obtains the first mirror projection data to ensure that the optical properties of the lens (such as diopter and astigmatism compensation) match the needs of the eye; if the lens position belongs to the non-projection area (the area on the lens that is not directly in contact with the eye, or the edge area of ​​the lens, etc.), the diopter distribution information and the cylinder distribution information are processed according to the mirror projection complement function. The mirror projection data is processed by processing the distribution information and cylindrical distribution information, that is, filling the data gaps in the non-projection area to ensure that the optical properties (diopter and cylinder) of each area on the lens are processed. Even if there is no direct diopter or cylinder information in some areas, the second mirror projection data is obtained to supplement the area on the lens that is not directly involved in the diopter and cylinder data to ensure that the optical effect of the entire lens will not be inconsistent due to missing data. Finally, the first mirror projection data (data in the projection area) and the second mirror projection data (data completed in the non-projection area) are merged to form complete mirror projection data.

[0058] Step S300, setting boundary conditions through the user end.

[0059] Preferably, the user sets some specific restrictions or parameters (boundary conditions) through an interactive interface (such as an application, software or hardware device) during the lens design process. The boundary conditions may include the physical size and shape of the lens, such as the diameter, thickness, curvature, size of the frame, height of the glasses, etc.; the field of view that the lens should cover (for example, only correcting central vision or including peripheral vision); the optical design of the lens can take into account the user's comfort requirements, adjust the focal length, light transmittance, anti-blue light function, etc., and even include personalized requirements on appearance (such as lens color, reflection effect, etc.); the environmental conditions of the lens, such as whether it is used outdoors, whether anti-ultraviolet function is required, whether specific lighting or motion requirements are taken into account, etc.; the purpose of setting boundary conditions is to constrain the parameter space of the design to ensure that the generated lens design meets the specific needs of the user. By setting boundary conditions, the functionality of the lens can be ensured, and the dispersion of the lens can be reduced, and field of view distortion can be avoided, thereby optimizing the visual effect and comfort.

[0060] Step S400, processing the boundary conditions and the mirror projection data through a polar coordinate configuration model to generate polar coordinate design parameters of the lens.

[0061] Preferably, the boundary conditions and mirror projection data in the lens design are processed by the polar coordinate configuration model, that is, the refractive power and cylindrical power distribution information of the eye (i.e., mirror projection data) is combined with the design requirements (boundary conditions) set by the user, and processed in the polar coordinate system to provide accurate design parameters for the production of the lens, wherein the polar coordinate system is a two-dimensional coordinate system, and the position is defined by angle (θ) and radius (r). Different from the traditional Cartesian coordinate system (X, Y coordinates), the polar coordinate system is usually used to describe circular or symmetrical objects, especially when dealing with objects involving spherical, circular or rotational symmetry. The polar coordinate configuration model maps the design parameters of the lens into the polar coordinate system, and converts the refractive power and cylindrical power information related to the eyeball (data obtained from the mirror projection analysis) into polar coordinate design parameters that meet the lens production requirements, thereby accurately converting the surface design details of the lens (such as curvature, thickness change, refractive power change, etc.) into specific parameters suitable for production.

[0062] Preferably, by combining the boundary conditions and the mirror projection data, the polar coordinate configuration model will generate the polar coordinate design parameters of the lens. These parameters are the specific design data required for the actual lens production, and usually include the surface curvature of the lens (the curvature of each area on the lens surface affects the refraction and focusing of light. Through the polar coordinate configuration model, the optimal curvature of each area can be calculated according to the refractive power requirements of different areas), thickness variation (the thickness of the lens may vary in different areas, especially in areas requiring high refractive power adjustment. The polar coordinate model will adjust the thickness distribution of the lens according to the refractive power data), refractive power and cylindrical power distribution (the polar coordinate design parameters also include the refractive power and cylindrical power distribution at different positions on the lens to ensure that the refractive requirements of each area are met) and optical coating and functional design (according to user needs, the model also adjusts the design of the optical coating on the lens surface, such as UV protection coating, anti-reflective coating, etc.), and then combines the advantages of the polar coordinate system and processes it in the most suitable way for the lens design to ensure that each part of the lens can match the refractive power and comfort requirements of the eye, and generate the polar coordinate design parameters of the lens.

[0063] Furthermore, step S400 also includes step S410, obtaining mirror projection record data and boundary condition record data; step S420, identifying lens polar coordinate identification data for the mirror projection record data and the boundary condition record data; step S430, using the lens polar coordinate identification data as supervision and the mirror projection record data and the boundary condition record data as input to train the polar coordinate configuration model.

[0064] Preferably, the mirror projection record data describes the projection of the eye's diopter and cylinder distribution on the lens surface, usually including the diopter and cylinder information of different areas, reflecting the visual correction effect that needs to be achieved on the lens surface; the boundary condition record data is the constraint condition set for the lens design to ensure that the generated design meets the actual needs and comfort requirements; the mirror projection record data and the boundary condition record data are marked with the lens polar coordinate identification data, that is, the mirror projection record data and the boundary condition record data are mapped into the polar coordinate system to identify the optical properties and shapes required by the lens at different positions (based on the radius and angle of the polar coordinates), for example, the diopter, cylinder and other properties of the lens; The surface degree, curvature, thickness, etc., and the polar coordinate identification data of the lens are used as the target data for polar coordinate configuration model training; then the polar coordinate identification data of the lens is used as the supervision data, and the mirror projection record data and the boundary condition record data are used as the input data to supervise the training prediction model. The model learns how to derive suitable lens design parameters (such as curvature, refractive power, cylindrical power, etc.) from the input data, and the polar coordinate configuration model is obtained through multiple iterations. It can automatically calculate the suitable lens design parameters according to the mirror projection record data and boundary condition record data provided by the user, to ensure that the lens can meet the user's personalized needs to provide the best visual effect and wearing comfort.

[0065] Further, step S420 also includes step S421, obtaining a polar coordinate identification function; step S422, processing the mirror projection recording data and the boundary condition recording data identification according to the polar coordinate identification function to obtain the lens polar coordinate identification data.

[0066] Preferably, in polar coordinates, Poisson's equation becomes:

[0067]

[0068] in The entire plane is mapped to a half-band domain {0≤r<∞,0≤θ≤2π} on the r,θ plane, which is singular when r=0. To add the condition that u is bounded at r=0, in order to select the solution of interest, u can be set to satisfy: For variables r, θ take equal step lengths Δr, Δθ respectively, let: r i = ( i+0.5 ) Δri=0、1、2…,θ j =jΔθj=0, 1…j-1,

[0069]

[0070] This forms a grid point on the half-band domain ( r i ,θ j) , distributed on the r-θ plane, such as Figure 3 As shown,

[0071] For any point ( r i ,θ i) (i>0), using the central difference quotient formula:

[0072]

[0073] In summary, the difference equation can be obtained:

[0074]

[0075] Arranged:

[0076]

[0077] When i = 0, the final differential formula can be derived using the finite volume method:

[0078]

[0079] Arranged:

[0080] 2(u 1,j -u 0,j )(Δθ) 2 +4(u 0,j+1 -2u 0,j )+4u 0,j-1 =f(r 0 ,θ j )(Δr) 2 (Δθ) 2 ;

[0081] Actual calculation:

[0082] For the maximum radius r max The circular area of r i = ( i+0.5 ) Δr, i = 0, 1, 2…n, θ j =jΔθ, j=0, 1…m-1;

[0083] Divide the entire area into a (m)(n+1) matrix, where one side of the matrix is ​​known (i=n), leaving m×n unknowns (j=0, 1, 2…m-1; i=0, 1…n-1);

[0084] Let x k =u ij , where k = j*n+i, 0≤i≤n-1, 0≤j≤m-1;

[0085] Then, a linear equation system with m×n unknown quantities can be obtained by the five-point difference format.

[0086] make:

[0087] When j = 0:

[0088] When i=0, 4x (m-1)n +[-2(Δθ) 2 -8]x 0 +2(Δθ) 2 x 1 +4x n =f(r 0 ,θ 0 )(Δr) 2 (Δθ) 2 ;

[0089] When 1<=i<=n-2, Ax m-1)n+i +Bx i-1 +Cx i +Dx i+1 +Ex n+i =f(r i ,θ 0 )(Δr) 2 (Δθ) 2 ;

[0090] When i=n-1, Ax (m-1)n+i +Bx i-1 +Cx i +Ex n+i =f(r i ,θ 0 )(Δr) 2 (Δθ) 2 -Dx n0 ;

[0091] When 1<=j<=m-2:

[0092] When i=0, 4x (j-1)n +[-2(Δθ) 2 -8]x jn +2(Δθ) 2 x jn+1 +4x (j+1)n =f(r 0 ,θ j )(Δr) 2 (Δθ) 2 ;

[0093] When 1<=i<=n-2, Ax j-1)n+i +Bxjn+i-1 +Cx jn+i +Dx jn+i+1 +Ex (j+1)n+i =f(r i ,i j )(Δr) 2 (Dth) 2 ;

[0094] When i=n-1, Ax (j-1)n+i +Bx jn+i-1 +Cx jn+i +Ex (j+1)n+i =f(r i ,i j )(Δr) 2 (Dth) 2 -Dx nj ;

[0095] When j=m-1:

[0096] When i=0,4x (j-1)n +[-2(Δθ) 2 -8]x jn +2(Dth) 2 x jn+1 +4x 0 =f(r 0 ,i j )(Δr) 2 (Dth) 2 ;

[0097] When 1<=i<=n-2, Ax (j-1)n+i +Bx jn+i-1 +Cx jn+i +Dx jn+i+1 +Ex 0+i =f(r i ,i j )(Δr) 2 (Dth) 2 ;

[0098] When i=n-1, Ax (j-1)n+i +Bx jn+i-1 +Cx jn+i +Ex 0+i =f(r i ,i j )(Δr) 2 (Dth) 2 -Dx nj ;

[0099] among them, x k =u ij: is the vector height value of the i-th row and j-th column on the grid; Δr is the sampling interval in the radial direction of the polar coordinates; Δθ is the sampling interval in the angular direction of the polar coordinates; is the radius value of the i-1th node in the radial direction; is the radius value of the i-th node in the radial direction; r i = ( i+0.5 ) Δr is the radius of the i-th node, r i and are different notations with the same meaning; f(r 0 ,θ j ) is the equivalent curvature of the corresponding point on the grid, which is determined by the actual measurement results; by solving the above linear equations, the design data of the entire surface can be obtained, that is, the polar coordinate identification data of the lens can be obtained.

[0100] Furthermore, step S430 also includes step S431, using the lens polar coordinate identification data as supervision and the mirror projection record data and the boundary condition record data as input to obtain a polar coordinate configuration model construction data set; step S432, based on the polar coordinate configuration model construction data set, training a regression decision tree to obtain a first polar coordinate configuration model, wherein the data weights of any group of the polar coordinate configuration model construction data set are the same; step S433, obtaining a first polar coordinate configuration model construction data set in which the output accuracy of the first polar coordinate configuration model is less than or equal to the output accuracy threshold; step S434, when the data amount of the first polar coordinate configuration model construction data set is less than or equal to the data amount threshold, setting the first polar coordinate configuration model to the polar coordinate configuration model.

[0101] Preferably, the polar coordinate identification data of the lens is combined with the mirror projection record data and the boundary condition record data to generate a polar coordinate configuration model construction data set, which provides input for the regression decision tree model and includes the design objectives and input conditions. The regression decision tree is a supervised learning model suitable for predicting continuous values ​​(such as diopter, cylinder and other design parameters). The task of the regression decision tree is to learn how to predict the design parameters of the lens (such as curvature, thickness, etc.) from the input mirror projection data and boundary condition data based on the training data set (i.e., the polar coordinate configuration model construction data set). Specifically, based on the polar coordinate configuration model construction data set, the regression decision tree learns how to map the input data (diopter, cylinder, boundary conditions, etc.) to the correct design parameters (polar coordinate design parameters) by continuously adjusting its internal parameters, thereby obtaining a first polar coordinate configuration model, which can be used according to the user's The most accurate lens design is generated according to the needs. When the data set is constructed, all data points have the same weight, which means that each set of input data and its corresponding target output (lens design parameters) are of equal importance in the training process, and there is no particular preference for certain data points. If the output accuracy of the regression decision tree is lower than the predetermined output accuracy threshold, it means that the model performs poorly in prediction and needs further adjustment. If at a certain stage, the accuracy of the output of the regression decision tree does not meet expectations, and the size of the data set for building the first polar coordinate configuration model is less than or equal to a set data volume threshold, further training or adjustment is stopped. When the data volume reaches the threshold and the model still cannot achieve a high accuracy, the current model can be used as the final polar coordinate configuration model. Even if the prediction accuracy of the model does not reach the optimal level, it can still be used as the basic model for lens design.

[0102] Furthermore, step S434 also includes step S434a, when the data volume of the first polar coordinate configuration model construction data set is greater than the data volume threshold, increasing the weight of the first polar coordinate configuration model construction data set in the polar coordinate configuration model construction data set to obtain a secondary polar coordinate configuration model construction data set; step S434b, training a regression decision tree based on the secondary polar coordinate configuration model construction data set to obtain a second polar coordinate configuration model and then performing a loop analysis until the data volume of the Nth polar coordinate configuration model construction data set is less than or equal to the data volume threshold, merging the first polar coordinate configuration model, the second polar coordinate configuration model until the Nth polar coordinate configuration model, and obtaining the polar coordinate configuration model.

[0103] Preferably, the high-quality polar coordinate configuration model is gradually improved and finally obtained through the model optimization and iteration method in the regression decision tree training process, including optimizing the final lens design parameter model (polar coordinate configuration model) through continuous cyclic analysis, adjusting the weight of the data set, and merging models of multiple stages. When the data volume of the first polar coordinate configuration model to construct the data set is greater than the set data volume threshold, it means that the data volume is large enough and the model can extract more complex patterns and relationships from it. The weight of the first polar coordinate configuration model to construct the data set is increased, and the knowledge from the data set is concentratedly learned, thereby improving the model's fitting effect on these specific data, and then obtaining the secondary polar coordinate configuration model to construct the data set; then the weight is increased. The secondary polar coordinate configuration model is used to construct a data set, and the regression decision tree is continued to be trained to obtain the second polar coordinate configuration model, which is more accurate than the model in the first stage; iterative analysis is performed, that is, after multiple rounds of training and adjustment, multiple polar coordinate configuration models are finally obtained, until the Nth round, the size of the constructed data set meets a certain standard (that is, the amount of data is less than or equal to the data amount threshold), and then the trained first polar coordinate configuration model, the second polar coordinate configuration model, and the Nth polar coordinate configuration model are merged into a polar coordinate configuration model, which can combine the advantages of each round of models, optimize the design parameters of the lens to the greatest extent, and can most accurately reflect the refractive power, cylindrical power distribution and boundary condition requirements of the eye, thereby providing accurate parameters for lens design.

[0104] Step S500, sending the lens polar coordinate design parameters to the lens design end.

[0105] Preferably, the optimized and trained lens design data (specifically, the polar coordinate design parameters of the lens) are transmitted to the actual lens design end for further processing and production. Specifically, the lens design end usually refers to a platform for lens design and production, including a user interface for customized glasses, for inputting vision data, selecting frame styles, etc., and equipment for manufacturing lenses, such as optical grinders, laser engraving machines, etc., which can accurately process the lenses according to the design parameters of the lenses, ensure that the design of the lenses accurately meets the visual needs of the users, and can be smoothly converted into production data, and finally produce personalized customized lenses.

[0106] Example 2

[0107] The present application also provides a lens designed by any one of the design methods for personalized customized defocus lenses.

[0108] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A design method for a personalized defocus lens, characterized in that: The method comprises: Detecting the diopter distribution information and cylindrical distribution information of the user's retina; Performing mirror projection analysis according to the refractive power distribution information and the cylindrical power distribution information to obtain mirror projection data; Through the user end, boundary conditions are set; Processing the boundary conditions and the mirror projection data through a polar coordinate configuration model to generate polar coordinate design parameters of the lens; The lens polar coordinate design parameters are sent to a lens design terminal.

2. A method for designing a personalized defocus lens as claimed in claim 1, characterized in that: Detect the user's retinal diopter distribution information and cylindrical distribution information, including: A plane coordinate system is constructed with the user's eye pupil as the origin, the horizontal direction as the first coordinate axis, and the vertical direction as the second coordinate axis; By means of an image collector, based on the plane coordinate system, the pupil of the eye is scanned at 0°, 45°, 90° and 135° respectively to obtain local diopter distribution information and local cylinder distribution information; Inputting the local refractive power distribution information into a refractive power interpolation analysis model, and outputting the refractive power distribution information; The local cylinder distribution information is input into a cylinder interpolation analysis model, and the cylinder distribution information is output.

3. A method for designing a personalized defocus lens as claimed in claim 2, characterized in that: include: The diopter interpolation analysis model construction step comprises: Collect local distribution data sets of diopter records of 0°, 45°, 90° and 135° of the user's eye pupil set respectively; Traversing the local distribution data set of diopter records to perform interpolation calculation to obtain a global distribution data set of diopter identifiers; Using the global distribution data set of diopter identification as supervision data and the local distribution data set of diopter records as input data, training the diopter interpolation analysis model; The cylinder interpolation analysis model construction step comprises: Collect local distribution data sets of cylinder records of 0°, 45°, 90° and 135° of the user's eye pupil set respectively; Traversing the local distribution data set of cylinder records to perform interpolation calculation to obtain a global distribution data set of cylinder identifiers; The cylinder identification global distribution data set is used as supervision data, and the cylinder record local distribution data set is used as input data to train the cylinder interpolation analysis model.

4. A method for designing a personalized defocus lens as claimed in claim 1, characterized in that: Performing mirror projection analysis according to the refractive power distribution information and the cylindrical power distribution information to obtain mirror projection data includes: Get the projection function: h i =L·tan ( the i ) ; z i =D i ; Among them, i Characterizes the i-th scanning angle, h i Characterizes the distance between the lens position and the eye axis at the i-th scanning angle, z i Characterizes the diopter or cylinder projected onto the mirror surface, D i Characterizes the detection diopter or cylinder of the i-th scanning angle, and L characterizes the distance between the center of the mirror surface and the eye axis; Construct the mirror projection completion function: Among them, D ( x ) Represents the equivalent diopter or equivalent cylinder, x0 is the maximum boundary position of the instrument measurement result, x1 is the actual maximum boundary position of the lens, D0 is the actual measured diopter or cylinder at the boundary x0, and x represents the position of the non-projection area of ​​the lens; When the lens position belongs to the projection area, the diopter distribution information and the cylindrical distribution information are processed according to the projection function to obtain first mirror projection data; When the lens position belongs to the non-projection area, the refractive power distribution information and the cylindrical power distribution information are processed according to the mirror projection completion function to obtain second mirror projection data; The first mirror projection data and the second mirror projection data are added into the mirror projection data.

5. The method for designing a personalized defocus lens according to claim 1, characterized in that: The polar coordinate configuration model building step comprises: Obtaining mirror projection recording data and boundary condition recording data; Marking lens polar coordinate identification data for the mirror projection recording data and the boundary condition recording data; The polar coordinate configuration model is trained using the lens polar coordinate identification data as supervision and the mirror projection record data and the boundary condition record data as input.

6. A method for designing a personalized defocus lens as claimed in claim 5, characterized in that: Marking the lens polar coordinate identification data for the mirror projection recording data and the boundary condition recording data includes: Get polar coordinate identity function; According to the polar coordinate identification function, the mirror projection record data and the boundary condition record data identification are processed to obtain the lens polar coordinate identification data.

7. A method for designing a personalized defocus lens as claimed in claim 5, characterized in that: Taking the lens polar coordinate identification data as supervision and the mirror projection record data and the boundary condition record data as input, training the polar coordinate configuration model includes: Using the lens polar coordinate identification data as supervision, and using the mirror projection record data and the boundary condition record data as input, to obtain a polar coordinate configuration model construction data set; Constructing a data set based on the polar coordinate configuration model, training a regression decision tree, and obtaining a first polar coordinate configuration model, wherein the data weights of any group of data sets constructed by the polar coordinate configuration model are the same; Obtaining a first polar coordinate configuration model construction data set whose output accuracy in the first polar coordinate configuration model is less than or equal to an output accuracy threshold; When the data amount of the first polar coordinate configuration model constructed data set is less than or equal to a data amount threshold, the first polar coordinate configuration model is set as the polar coordinate configuration model.

8. A method for designing a personalized defocus lens as claimed in claim 7, characterized in that: Also includes: When the data volume of the first polar coordinate configuration model construction data set is greater than the data volume threshold, increasing the weight of the first polar coordinate configuration model construction data set in the polar coordinate configuration model construction data set to obtain a secondary polar coordinate configuration model construction data set; A data set is constructed according to the secondary polar coordinate configuration model, and a regression decision tree is trained. After obtaining the second polar coordinate configuration model, the analysis is repeated until the amount of data in the data set constructed by the N-th polar coordinate configuration model is less than or equal to the data amount threshold, and the first polar coordinate configuration model, the second polar coordinate configuration model, and the N-th polar coordinate configuration model are merged to obtain the polar coordinate configuration model.

9. A lens, characterized in that: The design is performed by a design method for a personalized customized defocus lens as described in any one of claims 1 to 8.

Images