Lens surface type reconstruction method and device

Abstract

The invention discloses a lens surface type reconstruction method and device, and relates to the technical field of optical measurement. The objective of the invention is to solve the problems of poor measurement effect and strict requirements on an operation environment when an existing measurement method is used for a lens with a complex shape and a large inclination angle. Comprising a binocular system, a to-be-measured lens, an optical vibration isolation platform, a beam shaping system, a laser and an electric displacement table. The to-be-measured lens is positioned above the optical vibration isolation platform and is used for generating Tyndall scattering light beams under the irradiation of the laser; the two pairs of binocular systems are symmetrically arranged inside and outside the optical vibration isolation platform, the distances between the binocular systems and the to-be-measured lens are equal, and the binocular systems are used for shooting scattering images from the incident plane direction or the emergent plane direction of the to-be-measured lens; the electric displacement platform is arranged on the lower side of the optical vibration isolation platform through a track; the laser device is arranged on the electric displacement table, moves along a preset route under the action of the electric displacement table and realizes multi-angle and multi-position laser irradiation on the lens to be measured; and the light beam shaping system is arranged on the laser device and is used for shaping the laser from the laser device.

Description

Technical Field

[0001] The present invention relates to the field of optical measurement technology, and more specifically to a lens surface reconstruction method and device. Background Art

[0002] The measurement technology of optical lens surface topography has been widely used in many fields, including optical design, manufacturing, quality control and scientific research. Traditional optical lens surface topography measurement methods mainly rely on optical interference and mechanical triggering methods. There are many defects in the existing optical lens surface topography measurement methods: limited applicability, limited measurement angle and range, etc. Among them, the optical interference measurement method requires high-precision equipment and cumbersome processing, and the mechanical trigger measurement method is easy to damage the lens and has a slow scanning speed.

[0003] In summary, the existing measurement methods have the problems of poor measurement effect and strict requirements on the operating environment when used for lenses with complex shapes and large tilt angles. Summary of the invention

[0004] The embodiment of the present invention provides a lens surface reconstruction method and device, which are used to solve the problems that the existing measurement methods have poor measurement effect and strict requirements on the operating environment when measuring lenses with complex shapes and large inclination angles.

[0005] The embodiment of the present invention provides a lens surface reconstruction device, including a binocular system, a lens to be measured, an optical vibration isolation platform, a beam shaping system, a laser and an electric displacement stage;

[0006] A lens to be tested, which is located above the optical vibration isolation platform and is used to generate a Tyndall scattered light beam under the irradiation of laser;

[0007] A binocular system, wherein two pairs of binocular systems are symmetrically arranged inside and outside the optical vibration isolation platform, and are equidistant from the lens to be tested, and are used to capture a scattered image from the incident surface direction or the exit surface direction of the lens to be tested;

[0008] An electric displacement stage, which is arranged on the lower side of the optical vibration isolation platform through a track;

[0009] A laser, which is arranged on the electric displacement stage, moves along a preset route under the action of the electric displacement stage and realizes multi-angle and multi-position laser irradiation of the lens to be measured;

[0010] The beam shaping system is arranged on the laser and is used for shaping the laser light from the laser.

[0011] Preferably, the binocular system comprises a first binocular system and a second binocular system;

[0012] The first binocular system is arranged on one side above the optical vibration isolation platform, and is used to capture the Tyndall scattering image from the exit surface of the lens to be tested;

[0013] The second binocular system is arranged inside the optical vibration isolation platform, and the second binocular system is symmetrical to the first binocular system, and is used to capture the Tyndall scattering image from the incident surface of the lens to be tested.

[0014] Preferably, the laser is a short-wave laser with a wavelength of 450 nm.

[0015] Preferably, the track consists of two orthogonal tracks; or

[0016] The track consists of two orthogonal tracks and a rotation axis; or

[0017] The track consists of two orthogonal tracks and a pan / tilt platform.

[0018] The embodiment of the present invention further provides a lens surface reconstruction method, comprising:

[0019] Calibrate the relative position of the binocular system, adjust the laser emission angle and beam shaping system so that the laser beam waist position coincides with the center position of the lens to be tested;

[0020] Preprocessing the exit surface scattering image and the incident surface scattering image of the lens to be tested acquired by the binocular system, extracting and matching the exit endpoint of the preprocessed exit surface scattering image and the incident endpoint of the incident surface scattering image according to a feature point extraction and matching algorithm to obtain the exit endpoint coordinates and the incident endpoint coordinates, and unifying the exit endpoint coordinates and the incident endpoint coordinates into a world coordinate system according to the relative position of the binocular system;

[0021] Controlling the electric translation stage to drive the laser to move on a preset track, so that the laser beam scans the lens of the binocular system in a set manner, and the binocular system obtains scanning scattering images of different positions of the lens to be tested;

[0022] Point cloud data is obtained according to the characteristic points of the scanned scattering image, and the point cloud data is processed, feature extracted, fitted and reconstructed to obtain a three-dimensional surface model of the lens to be tested.

[0023] Preferably, the set scanning mode includes point scanning, line scanning, and surface scanning;

[0024] The method of scanning the laser beam on the lens of the binocular system in a set manner specifically includes:

[0025] Selecting a reference point on the lens of the binocular system, and controlling the movement of the electric translation stage so that the laser beam scans a certain range around the reference point, so that the binocular system acquires a scanning scattering image;

[0026] After completing the reference point scanning, the electric translation stage is controlled to move along a certain axis direction so that the laser beam scans along a certain axis direction, and the binocular system obtains a scanning scattered image on a line, gradually completing the scanning of the entire lens.

[0027] Preferably, the preset tracks include: dual orthogonal tracks, dual orthogonal tracks and rotation axes, dual orthogonal tracks and a pan / tilt platform;

[0028] The controlling of the electric translation stage to drive the laser to move on a preset track specifically includes:

[0029] When the electric translation stage drives the laser to move on the dual orthogonal tracks, the scanning sequence of the laser beam moves point by point along the X or Y axis direction to complete the scanning of a line, and the binocular system obtains the scanning scattering image on a line, gradually completing the scanning of the entire lens;

[0030] When the electric translation stage drives the laser to move on the dual orthogonal tracks and the rotation axis, the incident angle of the laser is adjusted through the rotation axis, so that the binocular system can obtain scanning scattering images at different angles;

[0031] When the electric translation stage drives the laser to move on the dual orthogonal tracks and the pan-tilt platform, the incident angle of the laser is adjusted through the pan-tilt platform so that the laser beam can cover every part of the lens surface.

[0032] An embodiment of the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes any one of the lens surface reconstruction methods described above.

[0033] An embodiment of the present invention provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes any one of the lens surface reconstruction methods described above.

[0034] The lens surface reconstruction device and method provided by the embodiment of the present invention include a binocular system, a lens to be measured, an optical vibration isolation platform, a beam shaping system, a laser and an electric displacement stage; the lens to be measured is located above the optical vibration isolation platform and is used to generate a Tyndall scattered beam under the irradiation of a laser; the binocular system, two pairs of binocular systems are symmetrically arranged inside and outside the optical vibration isolation platform, and are equidistant from the lens to be measured, and are used to capture a scattered image from the incident surface or the exit surface of the lens to be measured; the electric displacement stage is arranged on the lower side of the optical vibration isolation platform through a track; the laser is arranged on the electric displacement stage, moves along a preset route under the action of the electric displacement stage and realizes multi-angle and multi-position laser irradiation of the lens to be measured; the beam shaping system is arranged on the laser and is used to shape the laser from the laser. A lens surface reconstruction device is built through preset structures and positions. A short-wave laser beam is used to irradiate the lens to be measured so that the lens to be measured generates Tyndall effect scattered light, and the scattered light field distribution images in the incident surface direction and the exit surface direction are captured by a binocular system. Combined with feature point extraction and matching algorithms, high-precision reconstruction of the lens surface is achieved. It overcomes the problem of limited measurement angles in traditional measuring devices and can measure free-form surface lenses with large inclination angles at different angles. By adjusting the electric displacement stage of the laser, multi-angle and multi-position laser irradiation of the laser beam can be achieved. At the same time, by adjusting the binocular system inside and outside the optical vibration isolation platform, image capture of the lens to be measured can be achieved, and the upper and lower surfaces of the lens to be measured can be measured simultaneously, which can be suitable for the measurement of free-form surface lenses with complex structures. The device uses the scattered light characteristics of the Tyndall effect to simplify the surface measurement process of the lens to be measured, and also improves the surface detection efficiency of the lens to be measured. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0036] Figure 1 It is a schematic structural diagram of a lens surface reconstruction device based on the Tyndall effect of the present invention;

[0037] Figure 2 This is a flow chart of a lens surface reconstruction method based on the Tyndall effect of the present invention;

[0038] Figure 3 The Tyndall effect scattered light beam images of five samples captured by the camera of the embodiment of the present invention;

[0039] Figure 4 This is an error diagram of the flat glass thickness measurement experiment results according to an embodiment of the present invention;

[0040] Among them, there are a first binocular system ~1, a second binocular system ~2, a lens to be measured ~3, an optical vibration isolation platform ~4, a beam shaping system ~5, a laser ~6, and an electric displacement stage ~7. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] Figure 1 The schematic diagram of the structure of the lens surface reconstruction device based on the Tyndall effect of the present invention is shown below. Figure 1 The lens surface reconstruction device provided by the embodiment of the present invention is described in detail with reference to the example of FIG. Figure 1 As shown, the device mainly includes: a binocular system, a lens to be tested 3, an optical vibration isolation platform 4, a beam shaping system 5, a laser 6 and an electric displacement stage 7.

[0043] Specifically, the lens 3 to be tested is located above the optical vibration isolation platform 4, and is used to generate a Tyndall scattered light beam under the irradiation of the laser; two pairs of binocular systems are symmetrically arranged inside and outside the optical vibration isolation platform 4, and the distances between the two pairs of binocular systems and the lens 3 to be tested are equal; a pair of double-sided systems located on the upper surface of the optical vibration isolation platform 4 are used to capture the exit surface scattering image from the exit surface direction of the lens 3 to be tested, and a pair of double-sided systems located inside the optical vibration isolation platform 4 are used to capture the incident surface scattering image from the incident surface direction of the lens 3 to be tested; the electric displacement stage 7 is arranged on the lower side of the optical vibration isolation platform 4 through a track and is located below the laser to be tested. The electric displacement stage 7 is used to set the laser 6. Under the action of the electric displacement stage 7, the laser 6 can move along the preset route of the electric displacement stage 7, and realize the multi-angle and multi-position irradiation of the lens 3 to be tested by the laser 6; the beam shaping system 5 is arranged on the laser 6, and the beam shaping system 5 is used to shape the laser from the laser 6.

[0044] It should be noted that the binocular system provided in the embodiment of the present invention may include a pair of binocular systems, which are symmetrically arranged on the upper part and inside the optical vibration isolation platform 4; it may also include two pairs of binocular systems, one pair of binocular systems is arranged on the upper part of the optical vibration isolation platform 4, and one pair of binocular systems is arranged inside the optical vibration isolation platform 4; in the embodiment of the present invention, the specific number of binocular systems is not limited.

[0045] When the binocular system includes a pair of binocular systems, the pair of binocular systems may be a first binocular system 1 and a second binocular system 2. Specifically, the first binocular system 1 is arranged on one side above the optical vibration isolation platform 4, and is used to capture the exit surface scattering image from the exit surface direction of the lens 3 to be tested; the second binocular system 2 is arranged inside the optical vibration isolation platform 4, and is used to capture the incident surface scattering image from the incident surface direction of the lens 3 to be tested, and the second binocular system 2 is symmetrical to the first binocular system 1, that is, the distances between the first binocular system 1 and the lens 3 to be tested and the second binocular system 2 and the lens 3 to be tested are equal.

[0046] In the embodiment of the present invention, the laser 6 is a short-wave laser 6, whose wavelength may be 450 nm. A beam shaping system 5 is arranged in front of the laser 6, and the beam waist position of the laser beam may be changed by the beam shaping system 5 so that the beam waist position of the laser beam coincides with the center position of the lens 3 to be measured.

[0047] The lens surface reconstruction device provided by the embodiment of the present invention is constructed by presetting the structure and position. In the device, a short-wave laser beam is used to irradiate the lens to be measured so that the lens to be measured generates Tyndall effect scattered light, and the scattered light field distribution images in the incident surface direction and the exit surface direction are captured by a binocular system. In combination with feature point extraction and matching algorithms, high-precision reconstruction of the lens surface is achieved. The device overcomes the problem of limited measurement angle in traditional measuring devices and can measure free-form surface lenses with large inclination angles at different angles. By adjusting the electric displacement stage of the laser, laser irradiation of the laser beam at multiple angles and multiple positions can be achieved. At the same time, by adjusting the binocular system inside and outside the optical vibration isolation platform, image capture of the lens to be measured can be achieved, and the upper surface and the lower surface of the lens to be measured can be measured simultaneously, which can be suitable for the measurement of free-form surface lenses with complex structures. The device uses the scattered light characteristics of the Tyndall effect to simplify the surface shape measurement process of the lens to be measured, and also improves the surface shape detection efficiency of the lens to be measured.

[0048] In order to more clearly introduce the lens surface reconstruction device provided by the embodiment of the present invention, the following is combined with Figure 2 Introduce lens surface reconstruction methods, such as Figure 2 As shown, the method comprises the following steps:

[0049] Step 101, calibrating the relative position of the binocular system, adjusting the laser emission angle and the beam shaping system so that the laser beam waist position coincides with the center position of the lens to be measured;

[0050] Step 102, preprocessing the exit surface scattering image and the incident surface scattering image of the lens to be tested acquired by the binocular system, extracting and matching the exit endpoint of the preprocessed exit surface scattering image and the incident endpoint of the incident surface scattering image according to a feature point extraction and matching algorithm, obtaining the exit endpoint coordinates and the incident endpoint coordinates, and unifying the exit endpoint coordinates and the incident endpoint coordinates into a world coordinate system according to the relative position of the binocular system;

[0051] Step 103, controlling the electric translation stage to drive the laser to move on a preset track, so that the laser beam scans the lens of the binocular system according to a set method, and the binocular system obtains scanning scattering images of different positions of the lens to be tested;

[0052] Step 104 , obtaining point cloud data according to the feature points of the scanned scattering image, processing, feature extraction, fitting and reconstruction of the point cloud data, and obtaining a three-dimensional surface model of the lens to be tested.

[0053] Before step 101, the binocular system is calibrated. In the embodiment of the present invention, the binocular system is calibrated by the following method:

[0054] Step 101-1, using a checkerboard calibration plate as a calibration tool. In actual applications, the checkerboard calibration plate is composed of regular black and white squares. It is used for shooting when calibrating the binocular system. It is necessary to ensure that the size of the checkerboard calibration plate is appropriate, and the size of the squares included in the checkerboard must be able to be clearly captured by the two binocular systems.

[0055] Step 101-2, place the checkerboard calibration plate in the field of view that can be seen by both left cameras, and take multiple sets of calibration plate image pairs at different angles and distances. Each set of image pairs contains images taken by the left camera of the first binocular system and the second binocular system. Make sure that the images cover different field of view areas. It is recommended to collect at least 20 sets of image pairs to ensure the accuracy of calibration.

[0056] Step 101-3, perform internal calibration on the left camera of the first binocular system and the left camera of the second binocular system respectively. The internal calibration includes determining the focal length, optical center, distortion coefficient, etc. of the binocular system. Use a calibration algorithm, such as the "calibrateCamera" function of OpenCV, to perform internal calibration on the binocular systems respectively, and obtain the internal parameter matrix and distortion parameters.

[0057] Step 101-4, after completing the intrinsic calibration of the binocular system, perform stereo calibration (extrinsic calibration) between the two left cameras. This step determines the relative position and posture relationship of the two left cameras by calculating the rotation matrix (R) and translation vector (T) of the two left cameras. Use the "stereoCalibrate" function of OpenCV to optimize the calculation of the extrinsic parameters between the binocular system by inputting the previously collected image pairs.

[0058] Step 101-5, according to the calculated intrinsic and extrinsic parameters, the images of the two left cameras are geometrically rectified and stereo rectified, and the images are adjusted to a reference plane so that the same physical point is located in the same row in the two images. The "stereoRectify" and "initUndistortRectifyMap" functions of OpenCV are used to generate the rectified mapping table and perform the rectified processing on the images.

[0059] Step 101-6, after correction, take a new image pair, and verify the matching effect between the two left cameras through a feature matching algorithm (such as SIFT (Scale-Invariant Feature Transform), ORB (Oriented FAST and RotatedBRIEF, feature point detection and description algorithm), etc.) to ensure the accuracy of stereo calibration.

[0060] Step 101-7, after stereo calibration and correction, the left cameras of the first binocular system and the second binocular system are fixed on an optical vibration isolation platform to ensure that the relative positions of the cameras remain unchanged during the experiment.

[0061] Step 101-8, by repeating the experiment multiple times, verify whether the calibration effect between the left cameras is stable, and ensure that the calibration result under the experimental conditions has high accuracy and repeatability.

[0062] In step 101, after the relative position calibration of the binocular system is completed according to the above method, the internal and external parameters and relative position relationship of the binocular system can be obtained. The relative position relationship here is (R 3 |T 3), where R represents the rotation matrix, which describes the rotation relationship between the binocular systems. In stereo calibration, the R matrix is ​​the key data, which defines the rotation transformation from the perspective of one binocular system to the perspective of another binocular system; T represents the translation vector, which describes the translation between the two binocular systems, that is, the distance between them. The length of the T vector (i.e., the Euclidean distance) represents the baseline length of the two binocular systems. The calibration data (such as the binocular system internal parameter matrix K, distortion coefficient, rotation matrix, translation vector, stereo-corrected mapping matrix, and corrected projection matrix) are recorded for subsequent data processing to ensure the accuracy and consistency of the binocular system.

[0063] Furthermore, the emission angle of the laser is adjusted and the laser beam is shaped by the beam shaping system so that the waist position of the laser beam coincides with the center position of the lens to be measured. It should be noted that the wavelength of the laser here is 450nm. After the laser beam is adjusted by the beam shaping system, the beam diameter is 0.1mm.

[0064] In step 102, the binocular system located above the optical vibration isolation platform obtains the exit surface scattering image of the lens to be tested, and the binocular system located inside the optical vibration isolation platform obtains the incident surface scattering image of the lens to be tested. In the embodiment of the present invention, the captured exit surface scattering image and incident surface scattering image are first converted into grayscale images, and then Gaussian filtering is performed to remove noise to obtain a pre-processed exit surface scattering image and a pre-processed incident surface scattering image; further, the SIFT algorithm is used to extract feature points of the pre-processed exit surface scattering image and the pre-processed incident surface scattering image, and the incident endpoint coordinates X extracted from the incident surface scattering image are obtained. in , through the epipolar theorem to eliminate the interference of stray light spots, determine the only matching point X′ in and its spatial coordinates The coordinates of the exit endpoint corresponding to the exit surface scattering image are determined by the same method. out With the same name point X′ out get In this embodiment, X in is the coordinate of the incident endpoint, and its spatial coordinate vector in the camera coordinate system is in the form of X out is the coordinate of the exit endpoint, and its spatial coordinate vector in the camera coordinate system is in the form of

[0065] Furthermore, according to the relative position of the binocular system (R 3 |T 3 ), unify the coordinates of the exit endpoint and the incident endpoint into the world coordinate system, and its world coordinates can be expressed as

[0066] In step 103, the electric translation stage is controlled to drive the laser to move on a preset track. In the embodiment of the present invention, the track mainly includes three types: composed of two orthogonal tracks, composed of two orthogonal tracks and a rotation axis, and composed of two orthogonal tracks and a pan-tilt platform.

[0067] In practical applications, since a laser needs to be set on the electric translation stage, and the laser needs to be driven by the electric translation stage to scan the lens included in the binocular system, the track needs to be introduced in combination with the scanning method. Prior to this, the steps of scanning the lens with a laser beam are introduced first:

[0068] Point scanning: First, a reference point is selected on the lens surface of the binocular system, and the laser beam is scanned in a small range around the reference point by controlling the slight movement of the electric translation stage. Then the binocular system obtains the scanned scattered image and obtains point cloud data based on the feature points of the scanned scattered image. It should be noted that point scanning is mainly used to calibrate or verify scanning accuracy.

[0069] Line scanning: After completing the point scanning, the electric translation stage can control the laser to move along a certain axis (such as the X axis), and then collect data point by point, so as to form a scanning scattering image on a line and obtain point cloud data based on the characteristic points of the scanning scattering image. This linear scanning can be performed alternately in the X-axis and Y-axis directions, gradually covering the entire lens surface of the binocular system.

[0070] Surface scanning: After completing a line scan in the X-axis direction, move to the next Y-axis position and continue linear scanning along the X-axis until the entire lens surface is covered. This grid-based scanning method ensures that scattered images and feature point data can be collected at each position.

[0071] Furthermore, after the laser scanning method and the electric translation stage structure, the influence of various tracks on the laser scanning method includes the following types:

[0072] Tracks consisting of two dual orthogonal tracks: When the motorized translation stage consists of two orthogonal linear tracks, the scanning sequence can move point by point along the X-axis direction to complete the scanning of a line, then move to the next position in the Y-axis direction, and continue scanning in the X-axis direction, eventually covering the entire lens surface. This scanning method is suitable for flat lenses or surfaces of regular shapes. First, complete the scanning of a line, and then perform surface scanning line by line.

[0073] It consists of two orthogonal tracks and a rotation axis: If a rotation axis is added to two orthogonal linear tracks, the motorized stage can adjust the angle of the laser through the rotation axis while scanning the linear tracks. This design is suitable for scanning irregular lens surfaces, especially when it is necessary to obtain scattered images at different angles. The scanning order can be flexibly determined according to the shape of the scanned object by adjusting the rotation axis.

[0074] It consists of two orthogonal tracks and a pan-tilt stage. If the motorized translation stage design includes a pan-tilt stage, the incident angle of the laser can be dynamically adjusted during the scanning process. This design is suitable for scanning complex lens surfaces, such as spherical or multi-curved lenses. The scanning sequence is more flexible under this design, and the incident angle can be adjusted through the pan-tilt stage so that the laser beam can cover every part of the lens surface.

[0075] In step 104, the electric translation stage is controlled to move the laser to scan point by point along the surface of the lens. The above scanning method is repeated after each movement to obtain scanning scattering images at different positions. Point cloud data can be obtained based on the feature points of the scanning scattering images. The point cloud data is processed, feature extracted, fitted and reconstructed to obtain a three-dimensional surface model of the lens to be tested.

[0076] Specifically, point cloud data processing, fitting and lens 3D reconstruction include the following methods:

[0077] Point cloud data is a set of three-dimensional coordinate points obtained by scanning the lens surface. When processing point cloud data, the following steps are mainly involved:

[0078] Data preprocessing: Preprocessing is a key step to ensure the quality of point cloud data. It mainly includes: denoising, downsampling, point cloud alignment and feature extraction. Among them, denoising uses filters to remove noise in cloud data; for over-dense point cloud data, the voxel grid method can be used for downsampling to reduce the amount of data and retain the main feature points; for point cloud data obtained by multiple scans, the iterative nearest point algorithm is used for alignment to ensure that all point cloud data are in a unified coordinate system; representative feature points are extracted from the processed point cloud data for subsequent surface reconstruction. In this step, the principal component analysis method can be used to extract the main direction of the point cloud, or the SIFT algorithm can be used to extract key points.

[0079] Point cloud data fitting is to approximate the real shape of the lens surface through mathematical models. Common methods include: surface fitting, B-spline surface fitting and least squares fitting. Among them, surface fitting uses mathematical tools such as polynomial fitting and spline interpolation to fit the point cloud data. In particular, for regular lens surfaces, quadratic surfaces or high-order polynomials can be used for fitting to obtain the smooth surface of the lens; B-spline surface fitting is suitable for lens surfaces with complex shapes. B-spline surfaces construct a smooth surface through control points, and the surface shape can be flexibly adjusted to adapt to complex point cloud data; in the surface fitting process, the least squares method is often used to solve the fitting equation to minimize the error between the fitting surface and the point cloud data, thereby obtaining the optimal surface shape.

[0080] After completing the fitting of the point cloud data, the reconstruction process of the lens surface includes the following steps: triangulation reconstruction, smoothing and accuracy verification. Specifically, the fitted point cloud data is converted into a mesh model. The commonly used method is Delaunay triangulation, which forms a three-dimensional model of the lens surface by connecting the point cloud into a triangular mesh; the preliminary reconstructed lens surface is smoothed to eliminate sharp corners or irregular areas on the surface. The Laplace smoothing algorithm or bilateral filtering method can be used to smooth the surface to retain important geometric details; the accuracy verification verifies the accuracy of the reconstructed surface morphology by comparing it with the original point cloud data. The average distance and maximum deviation between the reconstructed surface and the original point cloud can be calculated to ensure the accuracy of the reconstruction.

[0081] In order to verify the feasibility and measurement accuracy of the lens surface reconstruction device and method provided by the embodiment of the present invention, the present invention implements a transparent flat material thickness measurement experiment. The experimental device can be referred to Figure 1 The device parameters included in this experiment are as follows:

[0082] The laser uses Herschel-Ray LDD-450 semiconductor laser with a wavelength of 450nm. After being shaped by the beam shaping system, the parallel beam diameter of the laser is 0.1mm.

[0083] The binocular system includes two CCD cameras, which are DMK42AUC03 of IMAGINGSOURCE, with a pixel size of 3.75 μm×3.75 μm and a resolution of 1280 pixel×960 pixel; the CCD camera includes a fixed-focus lens of WTL280-03X10;

[0084] The calibration board is a high-precision ceramic chessboard QDF12×12-21×21-0.5×0.5-bk, with a size of 10mm×10mm, a cell size of 0.5mm×0.5mm, and an accuracy of 1μm;

[0085] The experiment was conducted in a dust-free, closed, dark environment, and ensured that the smooth surface of the sample to be tested was free of dust adsorbed in the air. The samples to be tested in the experiment were five K19 flat glasses of different thicknesses but the same material, and the refractive index of the sample to be tested was 1.5256. Each sample to be tested was measured 9 times.

[0086] The experimental steps to verify the thickness measurement of transparent flat plate materials based on the Tyndall effect are as follows:

[0087] Step 201, projecting a short-wave laser beam onto the silicon dioxide substrate lens of the sample to be tested, to excite the sample to be tested to emit Tyndall effect scattered light;

[0088] Step 202, the binocular camera system captures a large number of exit surface scattering images and incident surface scattering images at different positions during the movement of the laser;

[0089] Step 203, converting the coordinates of the exit endpoint in the exit surface scattering image and the coordinates of the incident endpoint in the incident surface scattering image from the spatial coordinate system to the world coordinate system; the binocular camera system captures a large number of scanning scattering images at different positions during the movement of the laser;

[0090] Step 204, the computing unit preprocesses the acquired scanning scattering image to extract visual features of the scanning scattering image;

[0091] Step 205 : Based on the extracted visual features, the computing unit reconstructs the lens surface using a reconstruction model and an iterative optimization algorithm to generate a three-dimensional surface model of the sample to be tested.

[0092] Figure 4 The images of the scattered beams of the five samples captured by the camera provided in the embodiment of the present invention are shown in Table 1. The five samples were measured multiple times using the Fogale Nanotech low-coherence interferometer, and the average value of the multiple measurement results was used as the nominal value of the sample thickness, which was 1.980mm, 3.864mm, 5.887mm, 7.903mm, and 9.839mm, respectively. The comparison between the measurement method based on the Tyndall effect and the low-coherence interferometer results is shown in Tables 2 and 3. Figure 4 shown.

[0093] Table 1 Flat glass thickness measurement experimental results

[0094]

[0095]

[0096] Table 2 Flat glass thickness measurement experimental results error analysis measurement

[0097] Sample 1 Sample 2 Sample 3 Sample 4 Sample 5 Nominal thickness 1.98 3.864 5.887 7.903 9.839 Measured thickness 1.983 3.887 5.884 7.891 9.857 Absolute error 0.003 0.023 -0.003 -0.012 0.018 Relative error 0.16％ 0.61％ -0.05％ -0.15％ 0.18％

[0098] It can be seen from Table 2 that the maximum deviation of the measurement results of the transparent flat plate measurement method based on the Tyndall effect is 23μm compared with the low coherence interferometer, and the relative error is within 1%. It can be seen that the measurement method based on the visual characteristics of the Tyndall effect can be applied to the thickness measurement of transparent flat plate materials.

[0099] Based on the above experimental results, the feasibility and measurement accuracy of the surface measurement method based on the visual feature of the Tyndall effect are verified. For free-form surfaces, the technical route of surface reconstruction based on the lens Tyndall effect makes full use of the information of the scanning scattering image in the incident direction and the scanning scattering image in the outgoing direction, which can not only fuse the surface information of the two surfaces, but also use the ray tracing model combined with the Fresnel law for overall optimization.

[0100] Through the above experimental verification, it is proved that the lens surface reconstruction method based on the Tyndall effect provided by the embodiment of the present invention has a promising application in measuring the thickness of transparent materials and the surface shape of free-form surface lenses, providing a solid foundation for the practical application of this technology. Through the above implementation methods, the visual feature matching and spatial reconstruction technology based on the Tyndall effect can realize the accurate measurement and spatial reconstruction of the lens surface shape, and provide reliable technical support for the design, manufacture and detection of free-form surface optical lenses.

[0101] Another embodiment of the present invention further provides a computer device, comprising: a processor and a memory; the memory is used to store computer program code, and the computer program code includes computer instructions; when the processor executes the computer instructions, the electronic device executes each step of the lens surface reconstruction method shown in the above method embodiment.

[0102] Another embodiment of the present invention further provides a computer-readable storage medium, which stores computer instructions. When the computer instructions are executed on a computer device, the computer device executes each step of the lens surface reconstruction method shown in the above method embodiment.

[0103] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0104] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A lens surface reconstruction device, characterized in that: It includes binocular system, lens to be tested, optical vibration isolation platform, beam shaping system, laser and electric displacement stage; A lens to be tested, which is located above the optical vibration isolation platform and is used to generate a Tyndall scattered light beam under the irradiation of laser; A binocular system, wherein two pairs of binocular systems are symmetrically arranged inside and outside the optical vibration isolation platform, and are equidistant from the lens to be tested, and are used to capture a scattered image from the incident surface direction or the exit surface direction of the lens to be tested; An electric displacement stage, which is arranged on the lower side of the optical vibration isolation platform through a track; A laser, which is arranged on the electric displacement stage, moves along a preset route under the action of the electric displacement stage and realizes multi-angle and multi-position laser irradiation of the lens to be measured; The beam shaping system is arranged on the laser and is used for shaping the laser light from the laser.

2. The device according to claim 1, characterized in that The binocular system includes a first binocular system and a second binocular system; The first binocular system is arranged on one side above the optical vibration isolation platform, and is used to capture the Tyndall scattering image from the exit surface of the lens to be tested; The second binocular system is arranged inside the optical vibration isolation platform, and the second binocular system is symmetrical to the first binocular system, and is used to capture the Tyndall scattering image from the incident surface of the lens to be tested.

3. The device according to claim 1, characterized in that The laser is a short-wave laser with a wavelength of 450nm.

4. The device according to claim 1, characterized in that The track consists of two orthogonal tracks; or The track consists of two orthogonal tracks and a rotation axis; or The track consists of two orthogonal tracks and a pan / tilt platform.

5. A lens surface reconstruction method, characterized in that: include: Calibrate the relative position of the binocular system, adjust the laser emission angle and beam shaping system so that the laser beam waist position coincides with the center position of the lens to be tested; Preprocessing the exit surface scattering image and the incident surface scattering image of the lens to be tested acquired by the binocular system, extracting and matching the exit endpoint of the preprocessed exit surface scattering image and the incident endpoint of the incident surface scattering image according to a feature point extraction and matching algorithm to obtain the exit endpoint coordinates and the incident endpoint coordinates, and unifying the exit endpoint coordinates and the incident endpoint coordinates into a world coordinate system according to the relative position of the binocular system; Controlling the electric translation stage to drive the laser to move on a preset track, so that the laser beam scans the lens of the binocular system in a set manner, and the binocular system obtains scanning scattering images of different positions of the lens to be tested; Point cloud data is obtained according to the characteristic points of the scanned scattering image, and the point cloud data is processed, feature extracted, fitted and reconstructed to obtain a three-dimensional surface model of the lens to be tested.

6. The method according to claim 5, characterized in that The setting scanning mode includes point scanning, line scanning and surface scanning; The method of scanning the laser beam on the lens of the binocular system in a set manner specifically includes: Selecting a reference point on the lens of the binocular system, and controlling the movement of the electric translation stage so that the laser beam scans a certain range around the reference point, so that the binocular system acquires a scanning scattering image; After completing the reference point scanning, the electric translation stage is controlled to move along a certain axis direction so that the laser beam scans along a certain axis direction, and the binocular system obtains a scanning scattered image on a line, gradually completing the scanning of the entire lens.

7. The method according to claim 5, characterized in that The preset tracks include: dual orthogonal tracks, dual orthogonal tracks and rotation axes, dual orthogonal tracks and a pan / tilt platform; The controlling of the electric translation stage to drive the laser to move on a preset track specifically includes: When the electric translation stage drives the laser to move on the dual orthogonal tracks, the scanning sequence of the laser beam moves point by point along the X or Y axis direction to complete the scanning of a line, and the binocular system obtains the scanning scattering image on a line, gradually completing the scanning of the entire lens; When the electric translation stage drives the laser to move on the dual orthogonal tracks and the rotation axis, the incident angle of the laser is adjusted through the rotation axis, so that the binocular system can obtain scanning scattering images at different angles; When the electric translation stage drives the laser to move on the dual orthogonal tracks and the pan-tilt platform, the incident angle of the laser is adjusted through the pan-tilt platform so that the laser beam can cover every part of the lens surface.

8. A computer device, characterized in that: The computer device includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the lens surface reconstruction method as described in any one of claims 5-7.

9. A computer-readable storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a processor, the processor executes the lens surface reconstruction method as described in any one of claims 1 to 7.

Images