A method for measuring large-size reflective surfaces

By combining a controllable light source array with a camera array, and utilizing the law of reflection and iterative optimization, the problems of low efficiency and insufficient accuracy in the measurement of large-size reflective surfaces were solved, achieving fast and accurate measurement results.

CN115218813BActive Publication Date: 2024-11-08TIANJIN UNIV
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
CN202110424676.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-04-20
Publication Date
2024-11-08
Estimated Expiration
2041-04-20

AI Technical Summary

Technical Problem

Existing technologies struggle to efficiently and accurately measure the shape of large reflective surfaces, especially in military, aerospace, and industrial production. Traditional equipment is costly and inefficient, while optical measurement methods have limited accuracy in large fields of view.

Method used

By combining a controllable light source array with a camera array, the surface shape of a large reflective surface is reconstructed through calculation of the law of reflection and iterative optimization. The centroid coordinates of the light source are calculated using images captured by the camera array, and multiple iterative calculations are performed in combination with the reflected light rays and normal directions to reduce errors.

Benefits of technology

It enables rapid and accurate measurement of reflective surfaces, with measurement speeds down to the second level and accuracy down to the micrometer level. It is suitable for large-size reflective surfaces and simplifies the operation process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115218813B_ABST
    Figure CN115218813B_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of reflective surface measurement, and discloses a large-size reflective surface measurement method, comprising the following steps: S1: obtaining reflective surface images at multiple points and processing to obtain the centroid coordinates of the luminous points; S2: calculating the intersection of an outgoing light ray and a large-size reflective surface according to the centroid coordinates; S3: obtaining the direction of the incident light ray, and calculating the normal vector of the reflective surface to be measured; S4: reconstructing the surface shape of the reflective surface to be measured; the method can be used for measuring a large-size surface to be measured, has simple operation, and has a fast detection speed; and the method is based on the law of reflection for calculation, has high sensitivity to the surface shape of the large-size reflective surface, has a small error after verification, and can achieve micron-level measurement accuracy.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of reflective surface measurement, and in particular to a large-size reflective surface measurement method. Background Art

[0002] Large-size reflective surfaces are widely used in military, aerospace, and industrial production and life. For example, reflective telescopes, household flat mirrors, large-size tiles for construction, etc. are all reflective surfaces. Generally speaking, these devices need to ensure a certain surface accuracy to avoid quality problems during use. For example, if the surface quality of a reflective astronomical telescope is poor, its optical imaging performance will be seriously affected. Household flat mirrors need to meet flatness requirements to avoid imaging distortion. Tiles need to be evaluated for flatness before being put on the market to avoid problems such as breakage during use.

[0003] Manufacturing accuracy alone cannot meet the accuracy requirements of large-scale reflective surfaces, so before such structures are put into use, measurements are required to prevent defective products from being put into use. However, due to the large size of the reflective surface, large-scale reflective surfaces require a high measurement range. At present, the commonly used equipment for large-scale measurement is mostly large-scale three-coordinate measuring instruments, which are costly and have low measurement efficiency and cannot meet the measurement speed requirements. In some optical application fields, the measurement accuracy needs to reach the micron level. Traditional optical measurement methods, such as line structured light, grating structured light and other measurement methods based on optical triangulation principles, are limited in measurement accuracy by the camera array field of view and camera array resolution. Under a large measurement field of view, the accuracy is often only at the level of hundreds of microns. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention aims to provide a method for measuring a large-size reflective surface.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] A method for measuring a large-size reflective surface comprises the following steps:

[0007] S1: Obtain images of the reflective surface at multiple points and process them to obtain the centroid coordinates of the luminous points;

[0008] S2: Calculate the intersection of the outgoing light and the large-size reflective surface based on the barycentric coordinates;

[0009] S3: Get the direction of the incident light and calculate the normal vector of the reflective surface to be measured;

[0010] S4: Reconstruct the shape of the reflective surface to be measured.

[0011] In the present invention, preferably, after step S4, in order to reduce the error, step S5 is performed: the reconstructed reflective surface model data is brought into step S3 to re-perform several iterative calculations to obtain the final test result.

[0012] In the present invention, preferably, before step S1, a measurement system needs to be built, and the controllable luminous point light source and the camera array are installed on the same side of the large-size reflective surface to be measured, and the placement position of the camera array and the distribution of the controllable luminous point light source array are determined by calculation.

[0013] In the present invention, preferably, building a measurement system specifically includes the following steps:

[0014] S11: installing the controllable luminous point light source and the camera array on the same side of the large-size reflective surface to be measured, so that the camera array can capture the controllable luminous point light source through the large-size reflective surface;

[0015] S12: construct a simulated optical path, control the camera array field of view according to the camera array detection accuracy requirements, and calculate the camera array placement position based on the camera array imaging matrix;

[0016] S13: Calculate the camera array light direction corresponding to the sampling point of the large-size reflective surface through the camera array position, and determine the incident light direction based on the reflection law, that is, controllable light emission direction;

[0017] S14: Calculate the intersection of the incident light and the approximate plane where the controllable luminous point light source array is located, so as to determine the reference distribution position of the controllable luminous point light source array, and arrange the controllable luminous point light source array.

[0018] In the present invention, preferably, in step S1, the controllable luminous point light source array is controlled to emit light, and the camera array is used to shoot to obtain an image. The position of the controllable luminous point light source array is calculated by a binarization algorithm and a connected domain algorithm, and based on the position area, the centroid coordinates of the luminous point corresponding to the image are calculated.

[0019] In the present invention, preferably, in step S2, the barycentric coordinates are mapped based on the camera array imaging matrix to obtain a spatial straight line equation, that is, an outgoing light ray.

[0020] In the present invention, preferably, the intersection of the spatial straight line equation and the large-scale reflective surface shape equation is calculated.

[0021] In the present invention, preferably, in step S3, the intersection point and the light-emitting point are connected to obtain the direction of the incident light, and the normal direction vector is obtained according to the law of reflection.

[0022] In the present invention, preferably, in step S4, a numerical integration operation is performed on the normal direction vector to reconstruct the large-scale reflective surface shape.

[0023] In the present invention, preferably, in S5, the surface shape calculated in step S4 replaces the theoretical model of the large-size reflective surface in step S2, and then steps S2-S5 are repeated. The error is reduced through multiple iterations, so that the final result gradually converges to obtain the final measurement effect.

[0024] Compared with the prior art, the present invention has the following beneficial effects:

[0025] First, this method is based on a controllable light-emitting point light source array as a feature point. The measurement range can be expanded with the distribution range of the light-emitting points, and can be used to measure large-sized surfaces to be measured. Secondly, this method constructs a camera array and a controllable light-emitting point light source array, and the camera can complete the calculation by taking a small number of images. Compared with the existing technology, the operation is simple and the detection speed is fast. At the same time, this method is based on the specular reflection characteristics of the reflective surface for measurement, and can achieve second-level measurement. Moreover, this method is based on the law of reflection for calculation, and has high sensitivity to the surface shape of large-sized reflective surfaces. It has been verified that the error is small and can achieve micron-level measurement accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 The present invention is a flow chart of a large-size reflective surface measurement method.

[0027] Figure 2 A schematic diagram of the structure of a measurement system for a large-size reflective surface measurement method according to the present invention.

[0028] Figure 3 Schematic diagram of the principle of LED imaging in the camera.

[0029] Figure 4 Schematic diagram of the surface normal vector calculation based on ray tracing.

[0030] Figure 5 This is a schematic diagram of the surface shape restoration principle of a large-size reflective surface measurement method described in the present invention.

[0031] Figure 6 This is the iterative convergence process of the measurement error of the large-size reflective surface measurement method described in the present invention. DETAILED DESCRIPTION

[0032] 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.

[0033] It should be noted that when a component is referred to as being "fixed to" another component, it may be directly on the other component or there may also be a component centered. When a component is considered to be "connected to" another component, it may be directly connected to the other component or there may also be a component centered. When a component is considered to be "set on" another component, it may be directly set on the other component or there may also be a component centered. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only.

[0034] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0035] See also Figure 1 A preferred embodiment of the present invention provides a large-size reflective surface measurement method, which mainly constructs a large-size reflective surface testing system, and simulates and calculates the surface shape of the reflective surface through the cooperation of a light-emitting point light source and a camera array, thereby completing the measurement of the large-size reflective surface; the measurement method includes the following steps, S1: obtaining a reflective surface image at multiple points and processing to obtain the centroid coordinates of the light-emitting points; S2: calculating the intersection of the outgoing light and the large-size reflective surface according to the centroid coordinates; S3: obtaining the direction of the incident light and calculating the normal vector of the reflective surface to be measured; S4: integrating the normal vector to reconstruct the surface shape of the reflective surface to be measured.

[0036] In this embodiment, before performing the test step, a measurement system needs to be built, and the controllable luminous point light source and the camera array need to be installed on the same side of the large-size reflective surface to be measured, and the placement of the camera array and the distribution of the controllable luminous point light source array need to be determined by calculation.

[0037] Specifically, Figure 2 As shown, the measurement system includes a controllable light point light source array 2, a large-size reflective surface 1 and a camera array 3, wherein the controllable light point light source array 2 is composed of a large number of LED light sources, which is used to provide feature points to assist in subsequent surface shape calculation. The camera array 3 is used to collect the feature points provided by the controllable light point light source array 2 and calculate the specific coordinates of different controllable light point light sources. The shooting principle of the system is as follows Figure 3 As shown, the reflective surface can be partially divided into small plane mirrors, and the mirror image 4 of the controllable luminous point light source array 2 in the reflective surface equivalent to the plane mirror, the camera array 3 can capture the mirror image 4 of the luminous point light source array 2, and the surface shape data of the large-size reflective surface 1 can be obtained by further calculation through the mirror image 4.

[0038] Furthermore, building a measurement system specifically includes the following steps:

[0039] S11: installing the controllable luminous point light source and the camera array 3 on the same side of the large-size reflective surface 1 to be measured, so that the camera array 3 can capture the controllable luminous point light source through the large-size reflective surface 1;

[0040] S12: Construct a simulation optical path, control the field of view of the camera array 3 according to the detection accuracy requirement of the camera array 3, and calculate the placement position of the camera array 3 based on the imaging matrix of the camera array 3. The expression of the imaging matrix of the camera array 3 is:

[0041]

[0042] where x c ,yc,z c is the object point coordinate in the camera array 3 coordinate system, x u ,y u is the coordinate of the image point corresponding to the object point in the image plane coordinate system of camera array 3, and f is the distance between the optical center of camera array 3 and CCD (charge coupled device);

[0043] S13: Calculate the direction of the reflected light that can be captured by the camera array 3 corresponding to the sampling point (x0, y0, z0) on the large-size reflective surface 1 through the position of the camera array 3, and determine the direction of the incident light based on the law of reflection, that is, the controllable light emission direction, and the calculation method is:

[0044] R=I-2(N·I)N,

[0045] Wherein R is the unit vector of incident light, I is the unit vector of reflected light, and N is the normal vector of the corresponding reflection point on the large-size reflective surface 1.

[0046] S14: Using a plane to replace the curved surface where the controllable light emitting point light source array 2 is located to facilitate subsequent calculations, calculate the intersection of the incident light and the approximate plane where the controllable light emitting point light source array 2 is located, and the calculation formula is:

[0047]

[0048] Among them, a, b, c, d are the coefficients of the approximate plane equation where the controllable luminous point light source array 2 is located, x0, y0, z0 are the coordinates of the sampling points of the large-size reflective surface 1, (m, n, p) is the spatial vector of the incident light corresponding to each sampling point, x, y, z are the calculated intersection points, and these intersection points are the theoretical distribution positions of the controllable luminous point array, thereby determining the reference distribution position of the controllable luminous point light source array 2, and arranging the controllable luminous point light source array 2 according to the distribution.

[0049] Furthermore, during the actual installation process, a controllable luminous point light source array 2 arranged in a checkerboard shape can be used. In this case, the theoretical distribution position of the controllable luminous point light source array 2 calculated in step S14 is used as a reference value for the distribution density of the controllable luminous point light source array 2.

[0050] Furthermore, when setting up the measurement system, it is required that the sampling point density of the surface to be measured is greater than 1 / mm, the camera field of view is less than 200mm, in order to ensure the measurement accuracy, and the camera is tilted at 45° for shooting.

[0051] In this embodiment, after the measurement system is built, the measurement of the reflective surface to be measured is started. In step S1, the controllable light-emitting point light source array 2 is first controlled to light up in sequence, and the camera captures the corresponding image. The position of the controllable light-emitting point light source array 2 is calculated by the binarization algorithm and the connected domain algorithm. Based on the position area, the centroid coordinates of each light-emitting point are calculated, thereby establishing a corresponding relationship between the camera and the controllable light-emitting point light source. The calculation formula is:

[0052]

[0053]

[0054] Among them C x and C y is the calculated center of gravity coordinate, D ix and D iy is the pixel coordinate, V i is the gray value of the corresponding pixel.

[0055] Furthermore, in the constructed measurement system, after the controllable luminous point light source array 2 is turned on, the camera takes pictures. In each image taken by the camera, in addition to the main luminous point corresponding to the camera, other interfering luminous points in the surrounding area are also captured. First, the image is processed using a binary algorithm to make the luminous points on the reflective surface stand out, and then the connected domain algorithm is used to locate the luminous points in the image. Then, all the located luminous points are calculated to determine the coordinates of the luminous points corresponding to the camera, that is, the center of gravity coordinates.

[0056] Step S2, then solve based on the theoretical model of the large-size reflective surface 1, for each light-emitting point photographed by the camera array 3, the equation of a straight line in space can be mapped based on the imaging matrix of the camera array 3, and the expression is:

[0057]

[0058] Where r x 、r y 、r z, x0, y0, z0 are the constraint coefficients of the straight line, t is the intermediate variable of the parametric equation, that is, the distance between the straight line and the reference point, and the intersection point (x, y, z) of the two is calculated based on the straight line equation and the surface shape equation of the large-size reflective surface 1.

[0059] Furthermore, the centroid coordinates are substituted into the imaging matrix of the camera array 3 in step S12 to obtain the straight line equation, which is the expression of the outgoing light of the large-size reflective surface, and the intersection of the outgoing light and the theoretical surface shape equation of the large-size reflective surface is calculated, that is, the reflective point on the large-size reflective surface.

[0060] Step S3, connect the intersection point with the actual position of the light emitting point to obtain the direction of the incident light, and obtain the normal direction vector according to the reflection law, which is recorded as (P x ,P y ,P z ).

[0061] Step S4, after obtaining the surface normal vector to be determined, numerical integration is performed:

[0062]

[0063]

[0064] Reconstruct large-scale reflective surface shape.

[0065] In step S5, the theoretical model of the large-size reflective surface 1 in step S2 is replaced by z calculated in step S4, and steps S2-S5 are repeated. The error is reduced through multiple iterations, so that the final result gradually converges to obtain the final measurement effect.

[0066] Working principle:

[0067] First, build a measurement system, and design the layout of light sources and cameras according to the surface characteristics and accuracy requirements of the surface to be measured. Generally, the sampling point density of the surface to be measured is greater than 1 / mm, and the camera field of view is less than 200mm to ensure the measurement accuracy. Then control the camera to tilt 45° for shooting. Based on the distribution of sampling points under the ideal model of the surface to be measured, combined with the camera imaging formula, calculate the direction of the reflected light corresponding to the sampling point that can be captured by the camera. Based on the direction of the reflected light and the ideal surface equation of the large-size reflective surface 1, and the law of reflection, calculate the theoretical direction of the incident light. Arrange the controllable luminous point light source array 2 at a position approximately equal to the camera, and use an ideal plane as an approximation of the surface of the controllable luminous point light source array 2. Then calculate the intersection of the incident light obtained in the previous step and the ideal plane, which is the theoretical distribution position of the controllable luminous point light source. Then calculate the position and density of the controllable luminous point light source, and arrange the controllable luminous point light source array 2 based on this density.

[0068] After the measurement system is built, the measurement process is as follows:

[0069] First, the controllable luminous point light source array 2 is controlled to light up in sequence, and the camera array 3 captures the image. The position of the controllable luminous point light source array 2 is extracted from the captured image through the binarization algorithm and the connected domain algorithm, and the centroid coordinates of each luminous point are calculated in the position area, thereby establishing a corresponding relationship between the camera and the controllable luminous point light source.

[0070] After completing the mapping point calculation of the controllable luminous point light source on the camera, such as Figure 4 As shown, first, the direction and position of the outgoing light ray 9 are determined according to the mapping point 5 of the controllable luminous point light source on the camera CCD plane and the equivalent optical center 6 of the camera lens, and then based on the estimated surface shape 7 of the large-size reflective surface 1, the intersection point of the outgoing light ray 9 and the large-size reflective surface 1 is calculated, and the intersection point is the reflection point 8. Finally, the reflection point 8 and the position of the actual controllable luminous point light source are connected to obtain the incident light ray 10, and the angular bisector of the incident light ray and the reflected light ray is the surface normal vector 11 to be determined.

[0071] like Figure 5 As shown, after obtaining the surface normal vector to be determined, the surface shape 12 is obtained through integral operation. The partial derivatives of the corresponding point shape f in the x and y directions are calculated based on the surface normal vector. The shape of the reflective surface to be measured is restored by numerical integral operation. Since the shape of the large-scale reflective surface is in an ideal state in the initial restoration process, in order to calculate the normal vector, there is a certain deviation in the initial calculation result. The operation is repeated many times through iterative optimization method until the final result is stable.

[0072] A 500*500 LED array was built, and a 5×5 camera array was built at the same time. The large-size optical surface was photographed at an angle of 45°. After using this method, the error variation during multiple iterative optimization processes is as follows: Figure 6 As shown, it is proved that the measurement accuracy can reach micron level after multiple iterative optimizations.

[0073] The above description is a detailed description of the preferred feasible embodiments of the present invention, but the embodiments are not intended to limit the scope of the patent application of the present invention. All equivalent changes or modified changes completed under the technical spirit suggested by the present invention should fall within the patent scope covered by the present invention.

Claims

1. A method for measuring large-size reflective surfaces. It is characterized in that The steps include: S0: Build the measurement system: S11: installing the controllable luminous point light source and the camera array on the same side of the large-size reflective surface to be measured, so that the camera array can capture the controllable luminous point light source through the large-size reflective surface; S12: construct a simulated optical path, control the camera array field of view according to the camera array detection accuracy requirements, and calculate the camera array placement position based on the camera array imaging matrix; S13: Calculate the camera array light direction corresponding to the sampling point of the large-size reflective surface through the camera array position, and determine the incident light direction based on the reflection law, that is, controllable light emission direction; S14: calculating the intersection of the incident light and the approximate plane where the controllable light-emitting point light source array is located, thereby determining the reference distribution position of the controllable light-emitting point light source array, and arranging the controllable light-emitting point light source array; S1: Control the controllable light-emitting point light source array to emit light, and the camera array takes pictures to obtain the reflective surface image. The position area of ​​the controllable light-emitting point light source array is calculated by the binarization algorithm and the connected domain algorithm. Based on the position area, the centroid coordinates of the light-emitting point corresponding to the image are calculated; S2: Calculate the intersection of the outgoing light and the large-size reflective surface based on the barycentric coordinates; S3: Get the direction of the incident light and calculate the normal vector of the reflective surface to be measured; S4: Reconstruct the shape of the reflective surface to be measured.

2. A method for measuring a large-size reflective surface according to claim 1, It is characterized in that After step S4, in order to reduce the error, step S5 is performed: the reconstructed reflective surface model data is brought into step S3 to re-perform several iteration calculations to obtain the final test result.

3. A method for measuring a large-size reflective surface according to claim 2, It is characterized in that In step S2, the barycentric coordinates are mapped based on the camera array imaging matrix to obtain a spatial straight line equation, that is, an outgoing light ray.

4. A method for measuring a large-size reflective surface according to claim 3, It is characterized in that Calculate the intersection of the equation of a line in space and the equation of the shape of a large reflective surface.

5. A method for measuring a large-size reflective surface according to claim 1, It is characterized in that In step S3, the intersection point and the light-emitting point are connected to obtain the direction of the incident light, and the normal direction vector is obtained according to the law of reflection.

6. A method for measuring a large-size reflective surface according to claim 4, It is characterized in that In step S4, a numerical integration operation is performed on the normal direction vector to reconstruct the large-scale reflective surface shape.

7. A method for measuring a large-size reflective surface according to claim 2, It is characterized in that In S5, the theoretical model of the large-size reflective surface in step S2 is replaced by the surface shape calculated in step S4, and steps S2-S5 are repeated. The error is reduced through multiple iterations, so that the final result gradually converges to obtain the final measurement effect.

Citation Information

Patent Citations

  • Stereo vision measuring apparatus based on binocular omnidirectional visual sense sensor

    CN101308018A

  • Geometrical dimensional measurement apparatus and method for large-sized tunnel tunnel segment component

    CN101387494A

  • Ultra-large workpiece circular section shape and azimuthal measurement apparatus and method

    CN101387501A

  • High reflective object surface light field deflection technique measuring system and method

    CN105806257A

  • Heliostat shape rapid detection system and method thereof

    CN106197312A