阅读说明：本技术 一种分步测量光学镜片曲率半径和检测镜片缺陷的方法 (Method for measuring curvature radius of optical lens and detecting lens defect step by step ) 是由 窦健泰 龚渭 路森 邓晓龙 冉翔羽 厉淑贞 胡友友 于 2021-08-12 设计创作，主要内容包括：本发明提出了一种分步测量光学镜片曲率半径和检测镜片缺陷的方法,包括(1)初始曲率半径求解：根据采集的干涉图T(x,y)的暗环直径,求得初始曲率半径R-(0)；(2)曲线拟合：以R-(0)仿真牛顿环T-(0)(x,y),将T(x,y)和T-(0)(x,y)二阶极坐标变换后消除常数项,再叠加形成莫尔条纹,对莫尔条纹进行曲线拟合,得到待测镜片新的曲率半径R-(1)；(3)移相求波前：对R-(1)仿真的干涉图进行多步移相,得到多幅干涉图,分别与T(x,y)叠加,由多幅莫尔条纹得到波前差数据,进一步根据波前差分法求得待测镜片的高精确曲率半径R-(2)；(4)缺陷提取：利用R-(2)去掉干涉图T(x,y)中由曲率半径解得的波前数据,即可得到待测镜片的缺陷数据。(The invention provides a method for measuring the curvature radius of an optical lens and detecting the defects of the lens step by step, which comprises the following steps of (1) solving the initial curvature radius: according to the diameter of the dark ring of the acquired interference pattern T (x, y), the initial curvature radius R is obtained 0 (ii) a (2) And (3) curve fitting: with R 0 Simulated Newton's ring T 0 (x, y) A, B, and T (x, y) and T 0 Eliminating constant terms after (x, y) second-order polar coordinate transformation, superposing to form moire fringes, and performing curve fitting on the moire fringes to obtain a new curvature radius R of the lens to be measured 1 (ii) a (3) Phase-shifting wavefront determination: to R 1 The simulated interferogram is subjected to multi-step phase shifting to obtain a plurality of interferograms, the interferograms are respectively superposed with T (x, y), and waves are obtained from a plurality of Moire fringesFront difference data, further obtaining the high-precision curvature radius R of the lens to be measured according to the wave front difference method 2 (ii) a (4) Defect extraction: by means of R 2 And removing the wavefront data obtained by solving the curvature radius in the interference pattern T (x, y) to obtain the defect data of the lens to be detected.)

1. A method for step-by-step measurement of radius of curvature of an optical lens and detection of lens defects, comprising the steps of:

(1) solving the initial curvature radius: according to the diameter of the dark ring of the acquired interference pattern T (x, y), the initial curvature radius R is obtained₀；

(2) And (3) curve fitting: with R₀Simulated Newton's ring T₀(x, y) A, B, and T (x, y) and T₀(x, y) eliminating constant terms after polar coordinate transformation, superposing to form moire fringes, and performing curve fitting on the moire fringes to obtain a new curvature radius R of the lens to be measured₁；

(3) Phase-shifting wavefront determination: to R₁The simulated interferogram is subjected to multi-step phase shifting to obtain a plurality of interferograms which are respectively superposed with T (x, y), a plurality of moire fringes are used for obtaining wave front difference data, and further the high-accuracy curvature radius R of the lens to be measured is obtained according to a wave front difference method₂；

(4) Defect extraction: by means of R₂And removing the wavefront data obtained by solving the curvature radius in the interference pattern T (x, y) to obtain the defect data of the lens to be detected.

2. The method for step-by-step measurement of the curvature radius of an optical lens and detection of lens defects as claimed in claim 1, wherein the specific solution of the initial curvature radius in step (1) is: importing the collected Newton's ring interference pattern T (x, y) of the optical lens to be measured into a computer, extracting 5-10 groups of dark ring data from the interference pattern, calibrating the pixels, and calculating to obtain the actual diameter D of the m-level dark ring_mAnd n-stage dark ring actual diameter D_nAnd deducing an initial curvature radius solving expression of the optical lens to be measured according to the known laser wavelength lambda:

taking the average value as the initial curvature radius R of the lens to be measured₀。

3. The method for step-by-step measurement of the radius of curvature of an optical lens and detection of lens defects according to claim 1, wherein the curve fitting in step (2) specifically comprises:

step 1, according to the initial curvature radius R₀Simulated virtual Newton's ring T₀(x, y) transforming T by second order polar coordinate transformation₀(x, y) and T (x, y) become the line carrier frequency T₀(ρ, θ) and T (ρ, θ), from which the constant term of the interferogram is removed, to obtain T'₀(ρ, θ) and T '(ρ, θ), and mixing T'₀(ρ, θ) and T' (ρ, θ) are superimposed to form moire fringes S (ρ, θ), and the correlation expression is as follows:

expression of collected newton's ring interferograms: t (ρ, θ) is a (ρ, θ) + b (ρ, θ) cos (2 π f ρ), where a (ρ, θ) and b (ρ, θ) are the background intensity and fringe contrast, respectively, of T (x, y) in second-order polar coordinates, f is 1/(R λ), ρ is x²+y²In the formula, f is a variable to be solved, rho is the distance between the coordinate of any point of the Newton ring under the rectangular coordinate (x, y) and the center (0,0) of the Newton ring;

interferogram expression after removal of constant term:wherein T is_minAnd T_maxThe minimum and maximum values of T (ρ, θ), respectively;

simulated newton's ring interferogram expression: t is₀(ρ,θ)＝a₀(ρ,θ)+b₀(ρ,θ)·cos(2πf₀ρ) where a₀(p, theta) and b₀(p, theta) are each T₀(x, y) background intensity and fringe contrast in second order polar coordinates, f₀＝1/(R₀·λ)，ρ＝x²+y²；

Interferogram expression after removal of constant term:wherein T is_0(min)And T_0(max)Are respectively T₀A minimum value and a maximum value of (ρ, θ);

moire fringe expression superimposed after removing constant term: s (ρ, θ) ═ T '(ρ, θ) · T'₀(ρ,θ)＝cos(2πfρ)·cos(2πf₀ρ)；

Step 2, according to S (rho, theta), taking the corresponding gray value data of S (rho, theta) at a plurality of theta angles and using the functionTo fit S (ρ, θ) for each θ, where f₀For a known line carrier frequency coefficient, f is a variable to be solved, a is an unknown constant, and a range of f is determined with an error of +/-5%, namely 0.95f₀≤f≤1.05f₀Obtaining R values under a plurality of theta values by f 1/(R & lambda), and averaging to obtain a new curvature radius R of the lens to be measured₁。

4. The method for step-by-step measuring the radius of curvature of an optical lens and detecting lens defects according to claim 1, wherein the phase shifting wavefront determination in step (3) specifically comprises:

step 1: with R₁Simulating a plurality of Newton's ring interferograms with four-step phase shift, wherein the expressions are respectively as follows:

wherein: a is₁(x, y) is backScene intensity, b₁(x, y) is the fringe contrast,

step 2: mixing the above T₁(x,y)，T₂(x,y)，T₃(x,y)，T₄(x, y) are respectively superposed with T (x, y) to obtain corresponding moire fringes, and the moire fringes are subjected to low-pass filtering to obtain an expression containing wavefront difference information as follows:

the wavefront difference data is obtained by a four-step phase shifting method:

wherein s is₁(x,y)，s₂(x,y)，s₃(x,y)，s₄(x, y) is T (x, y) and T, respectively₁(x, y), T (x, y) and T₂(x, y), T (x, y) and T₃(x, y), T (x, y) and T₄(x, y) the moire fringes formed are subjected to low-pass filtering to obtain an expression containing wavefront difference information;

and step 3: from wave front difference w-w₁Relation to radius of curvature R:

the approximate derivation yields:

the final derived radius of curvature is:

and 4, step 4: from radius of curvature R to w-w₁Performing plane fitting on the data of R by using a least square method to obtain the high-precision radius of curvature R of the lens to be measured₂。

5. The method for step-by-step measurement of the radius of curvature of an optical lens and detection of lens defects according to claim 1, wherein the defect extraction in step (4) specifically comprises:

step 1: calibrating pixels, wherein P is D/L, D is the actual diameter of the lens to be measured, unit mm, L is the pixel diameter for collecting an interference pattern, and P is the actual size corresponding to the size of each unit pixel;

step 2: dividing the defect detection into a plurality of regions, and utilizing the R obtained in the step (3)₂Removing the radius of curvature R contained in T (x, y)₂Obtaining the data about the defects in the lens to be detected by the spherical information, observing the image after binarization processing, if the image is not defective, the image is a circle with uniform brightness, if the image is defective, the image has dark spots, and further observing the size of the dark spots in each area to obtain the pixel number s of the dark spots (defects);

and step 3: calculating the actual size S of each dark spot as Ps, wherein P is the actual size corresponding to each unit pixel size, and S is the number of pixels of the dark spot;

and 4, step 4: and comparing the lens defect standard according to the number of the dark spots of each area and the S value of each dark spot, and evaluating the condition of the lens defect.

6. The method for step-by-step measurement of the radius of curvature of the optical lens and the detection of the lens defect according to claim 1, wherein the interference pattern is collected by a detection device of the optical lens to be detected, the collimated light beam transmitted by the spectroscope (4) in the detection device is partially reflected by the rear surface of the spherical mirror (5) to be detected, the transmitted light is reflected by the surface of the first plane reflector (6), and the test light is formed by the spherical mirror (5) to be detected, so that the collected interference pattern contains the information of the surface and the internal defect of the spherical mirror (5) to be detected.

Technical Field

The invention relates to the field of optical detection, in particular to a method for measuring the curvature radius of an optical lens step by step and detecting the defects of the lens.

Background

Spherical optical lenses are the most common optical elements in optical systems, and play an important role in various optical systems. The radius of curvature is an important parameter characterizing a spherical optical lens, which has an important influence on the overall imaging quality of the optical system. In addition to the radius of curvature, surface defects and internal defects of spherical optical lenses also seriously affect the imaging quality.

Chinese patent publication No. CN110455221A discloses an optical path structure and equipment for rapidly measuring curvature radius of an optical lens, which comprises a light-emitting LED, a switching type lens unit, a prism, a semi-reflecting and semi-transparent lens, protective glass, a diaphragm and a CCD camera. The curvature radius of the measured sample is obtained by combining CCD image acquisition and a sub-pixel fitting circle algorithm and combining an industrial personal computer. Firstly, multiple reflections exist in the structure to change the optical track, so that the application range is widened, but transmission loss exists, and the imaging quality is influenced; secondly, the method for measuring the curvature radius in the patent is that the curvature radius obtained by performing sub-pixel fitting on a circle after further processing an image through an industrial personal computer has high precision, the precision is greatly influenced by the image quality, and the stability is reduced.

Chinese patent publication No. CN110793467A discloses an optical lens curvature radius precision detection device, which comprises a bottom plate, a fixing plate, an LED lamp, a grating sheet, a slide way, a slide block, and the like. Moire fringes are formed by adjusting the position of the grating sheet near the spherical center of the optical lens, and the curvature radius is calculated. The method has multiple debugging steps and is relatively complicated, a ronchi fringe method is adopted for measuring the curvature radius, and the method is limited by accurate positioning of the position of the center of the sphere in the aspect of measuring accuracy.

Chinese patent publication No. CN107860776A discloses a lens defect detection device and method, including: light source, reference arm, sample arm, controller and fiber coupler. And after interfering the sample signal light returned by the original path with the reference signal light returned by the original path of the reference arm to form an interference light signal, comparing the interference light signal with a standard mirror signal curve after processing, and evaluating the defect condition. The method realizes the detection of all parts of the whole optical lens, has comprehensive detection of defects, but needs to move the sample lens for many times in the detection process, and has long time consumption and low detection quality.

Chinese patent publication No. CN211905142U discloses a full-automatic eyeglass lens detection apparatus, comprising: the device comprises an upper frame, a lower frame, a fan filter unit, a three-color lamp, a display, a feeding conveyor belt, a discharging conveyor belt, a projection imaging image acquisition system and a refraction imaging image acquisition system. The defect condition of the optical lens can be well reflected by the shadow imaging system and the refraction imaging system. The method improves the precision and accuracy of image detection, but the experimental light source needs to be customized to obtain a high-quality measurement result, and the overall cost is high.

Disclosure of Invention

The purpose of the invention is as follows: the invention aims to provide a method for measuring the curvature radius of an optical lens step by step and detecting the defects of the lens, which has the advantages of high measurement precision, wide applicability, high efficiency, good stability, low cost and the like.

The technical scheme is as follows: the invention comprises the following steps:

(1) solving the initial curvature radius: according to the diameter of the dark ring of the acquired interference pattern T (x, y), the initial curvature radius R is obtained₀；

(2) And (3) curve fitting: with R₀Simulated Newton's ring T₀(x, y) A, B, and T (x, y) and T₀(x, y) eliminating constant terms after polar coordinate transformation, superposing to form moire fringes, and performing curve fitting on the moire fringes to obtain a new curvature radius R of the lens to be measured₁；

(3) Phase-shifting wavefront determination: to R₁The simulated interferogram is subjected to multi-step phase shifting to obtain a plurality of interferograms which are respectively superposed with T (x, y), a plurality of moire fringes are used for obtaining wave front difference data, and further the high-accuracy curvature radius R of the lens to be measured is obtained according to a wave front difference method₂；

(4) Defect extraction: by means of R₂And removing the wavefront data obtained by solving the curvature radius in the interference pattern T (x, y) to obtain the defect data of the lens to be detected.

The specific solving method of the initial curvature radius in the step (1) comprises the following steps: importing the collected Newton's ring interference pattern T (x, y) of the optical lens to be measured into a computer, extracting 5-10 groups of dark ring data from the interference pattern, calibrating the pixels, and calculating to obtain the actual diameter D of the m-level dark ring_mAnd n-stage dark ring actual diameter D_nAnd deducing an initial curvature radius solving expression of the optical lens to be measured according to the known laser wavelength lambda:

taking the average value as the initial curvature radius R of the lens to be measured₀。

The curve fitting in the step (2) specifically includes:

step 1, according to the initial curvature radius R₀Simulated virtual Newton's ring T₀(x, y) transforming T by second order polar coordinate transformation₀(x, y) and T (x, y) become the line carrier frequency T₀(ρ, θ) and T (ρ, θ), from which the constant term of the interferogram is removed, to obtain T'₀(ρ, θ) and T '(ρ, θ), and mixing T'₀(ρ, θ) and T' (ρ, θ) are superimposed to form moire fringes S (ρ, θ), and the correlation expression is as follows:

expression of collected newton's ring interferograms: t (ρ, θ) is a (ρ, θ) + b (ρ, θ) cos (2 π f ρ), where a (ρ, θ) and b (ρ, θ) are the background intensity and fringe contrast, respectively, of T (x, y) in second-order polar coordinates, f is 1/(R λ), ρ is x²+y²In the formula, f is a variable to be solved, rho is the distance between the coordinate of any point of the Newton ring under the rectangular coordinate (x, y) and the center (0,0) of the Newton ring;

interferogram expression after removal of constant term:wherein T is_minAnd T_maxThe minimum and maximum values of T (ρ, θ), respectively;

simulated newton's ring interferogram expression: t is₀(ρ,θ)＝a₀(ρ,θ)+b₀(ρ,θ)·cos(2πf₀ρ) where a₀(p, theta) and b₀(p, theta) are each T₀(x, y) background intensity and fringe contrast in second order polar coordinates, f₀＝1/(R₀·λ)，ρ＝x²+y²；

Interferogram expression after removal of constant term:wherein T is_0(min)And T_0(max)Are respectively T₀A minimum value and a maximum value of (ρ, θ);

moire fringe expression superimposed after removing constant term: s (ρ, θ) ═ T '(ρ, θ) · T'₀(ρ,θ)＝cos(2πfρ)·cos(2πf₀ρ)；

Step 2, according to S (rho, theta), taking the corresponding gray value data of S (rho, theta) at a plurality of theta angles and using the functionTo fit S (ρ, θ) for each θ, where f₀For a known line carrier frequency coefficient, f is a variable to be solved, a is an unknown constant, and a range of f is determined with an error of +/-5%, namely 0.95f₀≤f≤1.05f₀Obtaining R values under a plurality of theta values by f 1/(R & lambda), and averaging to obtain a new curvature radius R of the lens to be measured₁。

The phase-shifting wavefront calculation in the step (3) specifically includes:

step 1: with R₁Simulating a plurality of Newton's ring interferograms with four-step phase shift, wherein the expressions are respectively as follows:

wherein: a is₁(x, y) is background light intensity, b₁(x, y) is the fringe contrast,

step 2: mixing the above T₁(x,y)，T₂(x,y)，T₃(x,y)，T₄(x, y) are respectively superposed with T (x, y) to obtain corresponding moire fringes, and the moire fringes are subjected to low-pass filtering to obtain an expression containing wavefront difference information as follows:

is obtained by a four-step phase-shifting methodWavefront difference data:

wherein s is₁(x,y)，s₂(x,y)，s₃(x,y)，s₄(x, y) is T (x, y) and T, respectively₁(x, y), T (x, y) and T₂(x, y), T (x, y) and T₃(x, y), T (x, y) and T₄(x, y) the moire fringes formed are subjected to low-pass filtering to obtain an expression containing wavefront difference information;

and step 3: from wave front difference w-w₁Relation to radius of curvature R:

the approximate derivation yields:

the final derived radius of curvature is:

and 4, step 4: from radius of curvature R to w-w₁Performing plane fitting on the data of R by using a least square method to obtain the high-precision radius of curvature R of the lens to be measured₂。

The defect extraction in the step (4) specifically comprises the following steps:

step 1: calibrating pixels, wherein P is D/L, D is the actual diameter of the lens to be measured, unit mm, L is the pixel diameter for collecting an interference pattern, and P is the actual size corresponding to the size of each unit pixel;

step 2: dividing the defect detection into a plurality of regions, and utilizing the R obtained in the step (3)₂Removing the radius of curvature R contained in T (x, y)₂Obtaining the data about the defects in the lens to be detected, observing the image after binarization processing, if the image is not defective, the image is a circle with uniform brightness, if the image is defective, the image has dark spots, and further observing each areaThe dark spot size of the domain, obtaining the pixel number s of the dark spots (defects);

and step 3: calculating the actual size S of each dark spot as Ps, wherein P is the actual size corresponding to each unit pixel size, and S is the number of pixels of the dark spot;

and 4, step 4: and comparing the lens defect standard according to the number of the dark spots of each area and the S value of each dark spot, and evaluating the condition of the lens defect.

The interference pattern is collected by a detection device of the optical lens to be detected, collimated light beams transmitted by a spectroscope in the detection device are partially reflected by the rear surface of the spherical mirror to be detected, transmitted light is reflected by the surface of the first plane reflector, and test light is formed by the spherical mirror to be detected, so that the collected interference pattern contains surface and internal defect information of the spherical mirror to be detected.

Has the advantages that: the invention avoids the multiple acquisition and complex debugging process of the interferogram, and a single interferogram contains the wavefront information and the defect information of the curvature radius of the lens to be detected, improves the measurement precision of the curvature radius by solving step by step, and simultaneously can obtain the defect information of the lens to be detected, thereby realizing the detection of the lens to be detected with high precision, high efficiency, high stability and low cost.

Drawings

FIG. 1 is a flow chart of the present invention;

FIG. 2 is a diagram illustrating an apparatus for inspecting an optical lens according to the present invention;

FIG. 3 is a defect detection partition diagram.

Detailed Description

The invention will be further explained with reference to the drawings.

As shown in fig. 1, the process of the present invention includes: (1) solving the initial curvature radius: according to the diameter of the dark ring of the acquired interference pattern T (x, y), the initial curvature radius R is obtained₀(ii) a (2) And (3) curve fitting: with R₀Simulated Newton's ring T₀(x, y) A, B, and T (x, y) and T₀Eliminating constant terms after (x, y) second-order polar coordinate transformation, superposing to form moire fringes, and performing curve fitting on the moire fringes to obtain a new curvature radius R of the lens to be measured₁(ii) a (3) Phase-shifting wavefront determination: to R₁4-step phase shifting is carried out on the simulated interferogram to obtain 4 interferograms, the 4 interferograms are respectively superposed with T (x, y), wave front difference data are obtained through 4 Moire fringes, and the high-accuracy curvature radius R of the lens to be measured is further obtained according to a wave front difference method₂(ii) a (4) Defect extraction: by means of R₂And removing the wavefront data obtained by solving the curvature radius in the acquired interference pattern T (x, y), so as to obtain the defect data of the lens to be detected, and evaluating the defect condition of the lens to be detected through binarization processing.

The specific process is as follows:

(1) initial radius of curvature solution

As shown in fig. 2, the detection apparatus for an optical lens to be detected of the present invention includes a he-ne laser 1, a spatial filter 2, a collimating system 3, a beam splitter 4, a spherical mirror 5 to be detected, a first plane mirror 6, a second plane mirror 7, an imaging system 8, and a CCD detector 9, wherein the co-optical axes of the he-ne laser 1, the spatial filter 2, the collimating system 3, the beam splitter 4, the spherical mirror 5 to be detected, and the first plane mirror 6 are sequentially arranged, the co-optical axes of the second plane mirror 7, the imaging system 8, and the CCD detector 9 are arranged on the reflection light path of the beam splitter 4, and all optical elements are coaxial and equal in height relative to the substrate. A beam of coherent light is emitted by a helium-neon laser 1, is filtered by a spatial filter 2, is incident to a collimation system 3 to form a collimated light beam, is divided into transmitted light and reflected light by a spectroscope 4, the transmitted collimated light beam is partially reflected by the rear surface of a spherical mirror 5 to be tested, and the rest of the transmitted light is reflected by the surface of a first plane reflector 6 and forms test light through the spherical mirror 5 to be tested; the reflected light of the beam splitter 4 forms standard light after being reflected by the surface of the second plane reflector 7; the standard light and the test light enter the imaging system 8 through the beam splitter 4 and are imaged on the target surface of the CCD detector 9, and the image gray scale information recorded on the target surface of the CCD detector 9 is a Newton's ring interferogram T (x, y) containing the wave front information and the defect information of the curvature radius of the spherical mirror 5 to be tested.

Importing the collected Newton's ring interference pattern T (x, y) of the optical lens to be measured into a computer, extracting 5-10 groups of dark ring data from the interference pattern, calibrating the pixels, and calculating to obtain the actual diameter D of the m-level dark ring_mAnd n-stage dark ring actual diameter D_nThe wavelength of the laser, λ, is known,an initial curvature radius solving expression of the optical lens to be measured can be deduced:

obtaining 5-10 Rs according to 5-10 groups of data, and taking the average value as the initial curvature radius R of the lens to be measured₀。

(2) Fitting of curves

Step 1: obtaining the initial curvature radius R according to the step (1)₀Simulating virtual Newton's ring T₀(x, y) transforming T by second order polar coordinate transformation₀(x, y) and T (x, y) become the line carrier frequency T₀(ρ, θ) and T (ρ, θ), from which the constant term of the interferogram is removed, to obtain T'₀(ρ, θ) and T '(ρ, θ), and mixing T'₀(ρ, θ) and T' (ρ, θ) are superimposed to form moire fringes S (ρ, θ), and the correlation expression is as follows:

expression of collected newton's ring interferograms: t (ρ, θ) is a (ρ, θ) + b (ρ, θ) cos (2 π f ρ), where a (ρ, θ) and b (ρ, θ) are the background intensity and fringe contrast, respectively, of T (x, y) in second-order polar coordinates, f is 1/(R λ), ρ is x²+y²In the formula, f is a variable to be solved, rho is the distance between the coordinate of any point of the Newton ring under the rectangular coordinate (x, y) and the center (0,0) of the Newton ring;

interferogram expression after removal of constant term:wherein, T_minAnd T_maxThe minimum and maximum values of T (ρ, θ), respectively;

simulated newton's ring interferogram expression: t is₀(ρ,θ)＝a₀(ρ,θ)+b₀(ρ,θ)·cos(2πf₀ρ) where a₀(p, theta) and b₀(p, theta) are each T₀(x, y) background intensity and fringe contrast in second order polar coordinates, f₀＝1/(R₀·λ)，ρ＝x²+y²；

Interferogram expression after removal of constant term:wherein, T_0(min)And T_0(max)Are respectively T₀A minimum value and a maximum value of (ρ, θ);

moire fringe expression superimposed after removing constant term: s (ρ, θ) ═ T '(ρ, θ) · T'₀(ρ,θ)＝cos(2πfρ)·cos(2πf₀ρ)。

Step 2: based on S (rho, theta), the corresponding S (rho, theta) gray value data is respectively taken when theta is pi/3, 2 pi/3, pi, 4 pi/3 and 5 pi/3, and the data is processed by a functionTo fit S (ρ, θ) for each θ, where f₀The linear carrier frequency coefficient is known, f is a variable to be solved, and a is an unknown constant; for an accurate and fast fit, a range of f is determined with an error of + -5%, i.e. 0.95f₀≤f≤1.05f₀(ii) a And each theta corresponds to the data of S (rho, theta) to make f_θFitting a group of f and a, obtaining R values under 5 theta values by f being 1/(R & lambda), and obtaining a new curvature radius R of the lens to be measured by averaging₁。

(3) Phase-shift wavefront determination

Step 1: with R₁Simulating 4 Newton's ring interferograms with four-step phase shift, wherein the expressions are respectively as follows:

wherein: a is₁(x, y) is background light intensity, b₁(x, y) is the fringe contrast,

step 2: mixing the above T₁(x,y)，T₂(x,y)，T₃(x,y)，T₄(x, y) are respectively superposed with T (x, y) to obtain 4 moire fringes, and the 4 moire fringes are subjected to low-pass filtering to obtain an expression containing wavefront difference information, wherein the expression comprises the following steps:

the wavefront difference data is obtained by a four-step phase shifting method:

wherein s is₁(x,y)，s₂(x,y)，s₃(x,y)，s₄(x, y) is T (x, y) and T, respectively₁(x, y), T (x, y) and T₂(x, y), T (x, y) and T₃(x, y), T (x, y) and T₄(x, y) the moire fringes formed are subjected to low-pass filtering to obtain an expression containing wavefront difference information;

and step 3: from wave front difference w-w₁Relation to radius of curvature R:

the approximate derivation yields:

the final derived radius of curvature is:

and 4, step 4: from radius of curvature R to w-w₁Performing plane fitting on the data of R by using a least square method to obtain the high-precision radius of curvature R of the lens to be measured₂。

(4) Defect extraction

As shown in fig. 3, the defect detection partition of the present invention includes: edge 5 mm's three district is entered to the edge, and edge 5mm is entered to edge 15 mm's second district to and a central part's district, concrete defect detects and is:

step 1: performing pixel calibration, wherein P is D/L, wherein D is the actual diameter of the lens to be measured and is unit mm, L is the pixel diameter for collecting an interference pattern, and P is the actual size corresponding to the size of each unit pixel;

step 2: r obtained from wavefront by third step phase shift₂Removing the radius of curvature R contained in T (x, y)₂Obtaining the data about the defects in the lens to be detected by the spherical information, observing the image after binarization processing, if the image is not defective, the image is a circle with uniform brightness, if the image is defective, dark spots with different shapes exist, and further observing the sizes of the dark spots of the first area, the second area and the third area to obtain the pixel number s of the dark spots (defects);

and step 3: calculating the actual size S of each dark spot as Ps, wherein P is the actual size corresponding to each unit pixel size, and S is the number of pixels of the dark spot;

and 4, step 4: and comparing the lens defect standard according to the number of the dark spots of each area of the first area, the second area and the third area and the S value of each dark spot, and evaluating the lens defect condition.