An adaptive sub-aperture stitching wavefront detection method based on slope variance

Abstract

An adaptive sub-aperture stitching wavefront detection method based on slope variance is disclosed, belonging to the technical field of wavefront detection. To address the difficulty in balancing stitching accuracy and efficiency in existing methods, this method includes: global sparse coarse scanning: determining the number and position of sub-apertures based on low overlap, dividing the wavefront of a large-aperture optical system or component into several sub-apertures, and controlling a displacement stage to sequentially acquire data; local gradient calculation and variance evaluation: performing preliminary stitching on the data obtained from the sparse coarse scanning, resolving the global slope field, and calculating the local slope variance; high-risk area identification and target densification; adaptive targeted fine scanning data acquisition: controlling the displacement stage to acquire corresponding targeted fine scanning sub-aperture data at the generated targeted fine scanning coordinate points; fusing the sparse coarse scanning data and targeted fine scanning data, and performing final stitching using least squares average error and the Zernike mode method to obtain the original aperture wavefront data. This invention ensures both stitching accuracy and efficiency.

Description

Technical Field

[0001] This invention belongs to the technical field of wavefront detection, and relates to phase detection, and particularly to an adaptive sub-aperture splicing wavefront detection method based on slope variance, which is applied to the detection of optical components and systems. Background Technology

[0002] Large-aperture optical systems are playing an increasingly important role in the field of optical engineering. Their superior light-gathering performance and long focal lengths enable them to play a crucial role in deep space exploration, military security, and other fields. They are also widely used in laser fusion, remote sensing, and space optics. The effectiveness of large-aperture optical systems in these fields depends on the precision of their internal optical components, and the fabrication and testing of these components remains a significant challenge in the industry.

[0003] Direct detection using a wavefront detector is a common method in wavefront detection, but it requires the aperture of the wavefront detector to be the same as or slightly larger than the aperture of the system or component under test. This requirement significantly increases the cost and difficulty in measuring large-aperture optical systems and components.

[0004] Using the sub-aperture stitching method, the wavefront of a large-aperture optical system or component can be detected by a small-aperture wavefront detector. By stitching together the wavefront information of individual small-aperture sub-apertures, the complete original large-aperture wavefront information can be restored, thus avoiding the shortcomings of direct detection methods.

[0005] In the traditional sub-aperture stitching method, the overlap rate of the sub-apertures is fixed, and the range of the overlapping area is also fixed. This raises the question of how to choose the overlap rate. If the overlap rate is too low, Zernike fitting divergence may occur in the high gradient area of ​​the surface, resulting in reduced stitching accuracy. If the overlap rate is too high, although distortion is suppressed by the "large number averaging effect", the operation is cumbersome, time-consuming, and more susceptible to the influence of environmental temperature drift.

[0006] The choice of overlap rate affects both accuracy and efficiency. Determining the overlap rate is a crucial step when splicing sub-apertures.

[0007] Chinese patent publication number "CN117288116A" is titled "A Sub-Aperture Stitching Method for Optical Elements Based on Self-Adjusting Secondary Particle Swarm Optimization." This invention uses the measured sub-aperture error range as a basis, iteratively calculates compensation for the sub-aperture error through a particle swarm optimization algorithm, adjusts the sub-aperture error coefficient, and then re-stitches the sub-apertures accordingly. The particle swarm optimization algorithm used in this invention is an iterative search algorithm, which is time-consuming under multivariable and high-resolution aberration conditions and may encounter local optima, limiting detection efficiency. Reducing the number of iterations to decrease time may fail to find the optimal solution, thus affecting accuracy. Therefore, this invention suffers from a difficulty in balancing stitching accuracy and efficiency. Summary of the Invention

[0008] To address the existing problem of balancing stitching accuracy and efficiency, this invention proposes an adaptive sub-aperture stitching wavefront detection method based on slope variance. This method no longer uses a fixed sampling overlap rate, but instead improves efficiency while maintaining accuracy through the combination of different overlap rates.

[0009] This invention provides the following technical solutions:

[0010] An adaptive sub-aperture stitching wavefront detection method based on slope variance, comprising the following steps:

[0011] Step S1: Global sparse coarse scan;

[0012] The number and position of sub-apertures are determined based on the low overlap rate. The wavefront of a large-aperture optical system or component is divided into several sub-apertures, and the displacement stage is controlled to collect data sequentially.

[0013] Step S2: Local gradient calculation and variance evaluation;

[0014] The data obtained from the sparse coarse scan are initially stitched together to extract the global slope field and calculate the local slope variance.

[0015] Step S3: High-risk area identification and targeted encryption;

[0016] Regions with extremely high spatial frequency and drastic changes in local slope gradient are marked as high-risk regions. Based on these regions, fine-scan coordinates are generated to increase the local overlap rate.

[0017] Step S4: Adaptive targeted fine scanning data acquisition;

[0018] The control stage collects the corresponding target fine scan sub-aperture data at the generated target fine scan coordinate points;

[0019] Step S5: Fuse the sparse coarse scan data with the targeted fine scan data, and finally stitch them together using the least squares average error and Zernike mode method to obtain the original aperture wavefront data.

[0020] The specific steps of step S2 are as follows:

[0021] S2-1. Calculate the local slope magnitude matrix based on the global slope field obtained after preliminary stitching:

[0022] ;

[0023] In the formula This is a local slope magnitude matrix. , For preliminary splicing or local areas , Slope data of the direction;

[0024] S2-2. Based on this amplitude matrix, the slope variance in each local region can be calculated. As an assessment factor for high-risk areas:

[0025] ;

[0026] In the formula For the slope variance, To evaluate the number of valid pixels within a local evaluation region, For the traversal index of the pixel (from 1 to ... ), To assess the first in the region The slope magnitude of each pixel To assess all within the region The mathematical average of the slope magnitudes of each pixel.

[0027] The specific steps of step S3 are as follows:

[0028] S3-1, Selecting the first global slope variance data One quantile, used as a dynamic threshold High-risk areas are local slope variances. Spatial region;

[0029] S3-2. Determine the high-risk baseline coordinates and neighborhood set. Assume that the retrieved high-risk baseline coordinates are... The region where the coarse scan sub-aperture is located conforms to Then, the center coordinates of the coarse scan sub-aperture are extracted as the high-risk reference coordinates, denoted as... Then, in the coarse scan sub-aperture, the high-risk reference coordinates are retrieved. For all adjacent sub-apertures, mark their center coordinates as follows: The set of all neighboring sub-apertures that meet the conditions is called the neighborhood set, denoted as . ;

[0030] If a sub-aperture meets the following formula, then the sub-aperture is considered a neighboring sub-aperture:

[0031] ;

[0032] In the formula, This is the step distance for the coarse scan. For mechanical tolerance;

[0033] S3-3. Determine the new target fine scan coordinates according to the following formula. ;

[0034] ;

[0035] ;

[0036] The above formula means that the high-risk reference coordinates Its neighborhood set Coordinates of the center of all neighboring sub-apertures Calculate the midpoint of the vector;

[0037] S3-4. Perform pupil mask boundary constraints to prevent some coordinates of the system-generated target fine scanning coordinates from exceeding the effective physical aperture range of the device under test:

[0038] ;

[0039] In the formula The diameter of the component under test is denoted as . To provide a preset safety margin; the target precision scanning coordinates satisfy this formula. Perform a subsequent scan to remove coordinates that are not satisfied.

[0040] Step S5 specifically involves: after acquiring data again in step S4, combining it with the initial data acquired in S1, the original wavefront can be reconstructed using newly added targeted fine scan data and the original coarse scan data. The stitched wavefront is then obtained using the least squares averaging error and Zernike mode method. A wavefront is defined as a surface composed of all points in a light wave with the same phase; that is, a wavefront is essentially an isophase surface. Wavefront phase parameters can be obtained from the stitched wavefront.

[0041] ;

[0042] The formula is the key formula for least squares equalization error, where Let be the sub-aperture pose compensation vector to be solved. Sub-aperture overlap topology connectivity matrix The generalized inverse matrix, This is the relative slope error vector of neighboring sub-apertures within the overlapping region.

[0043] ;

[0044] The formula is the key formula of the Zernike pattern method, where Let be the Zernike mode coefficient vector to be solved. This is the partial derivative matrix of Zernike polynomial theory. This is the global slope column vector.

[0045] The beneficial effects of this invention are:

[0046] Based on the coarse-fine cascaded targeted resampling scheme, this invention achieves detection accuracy close to that of a global high overlap fine scan with a data acquisition time close to that of a coarse scan, while ensuring stitching accuracy and efficiency, and improving the timeliness and resistance to environmental temperature drift of wavefront detection.

[0047] In regions with smooth slopes, a low overlap rate is employed to improve efficiency by reducing data redundancy and shortening the time required for data acquisition. This also reduces the accumulation of measurement noise and prevents overfitting due to local noise. In high-risk regions with steep slopes, the overlap rate is increased by adding targeted densification sub-apertures, thereby enlarging the overlapping area and improving measurement accuracy in these high-risk regions. This, in turn, enhances overall measurement accuracy, ultimately achieving both high efficiency and high stitching accuracy.

[0048] By using physical local adaptive resampling, the requirement for ultra-high precision displacement stages in sub-aperture stitching technology is reduced. Lower precision and lower cost displacement stages can be used, effectively reducing detection costs while ensuring accuracy, thus achieving the effect of cost reduction and efficiency improvement. Attached Figure Description

[0049] Figure 1 This is a flowchart of an adaptive sub-aperture splicing wavefront detection method based on slope variance according to the present invention;

[0050] Figure 2 This is a flowchart illustrating the coarse-fine cascaded targeted resampling principle of an adaptive sub-aperture splicing wavefront detection method based on slope variance according to the present invention.

[0051] Figure 3 The diagram shows a comparison of the sub-aperture arrangement of the adaptive sub-aperture splicing wavefront detection method based on slope variance according to the present invention. In the diagram, A is the global fine scan sub-aperture arrangement, B is the sparse coarse scan sub-aperture arrangement, and C is the coarse and fine fusion sub-aperture arrangement.

[0052] Figure 4 This is a local slope gradient map of an adaptive sub-aperture splicing wavefront detection method based on slope variance according to the present invention.

[0053] Figure 5The images show the original wavefront truth map and wavefront phase comparison map of the adaptive sub-aperture splicing wavefront detection method based on slope variance according to the present invention. In the image A, the global fine scan wavefront phase map is shown, the image B, the sparse coarse scan wavefront phase map is shown, and the image C, the coarse and fine fused wavefront phase map is shown.

[0054] Figure 6 The images show a comparison of the reconstructed wavefront residuals of the adaptive sub-aperture splicing wavefront detection method based on slope variance according to the present invention. Image A is the global fine scan residual image, image B is the sparse coarse scan residual image, and image C is the coarse-fine fusion residual image. Detailed Implementation

[0055] The present invention will be further described in detail below with reference to the accompanying drawings.

[0056] like Figure 1 , Figure 2 As shown, an adaptive sub-aperture stitching wavefront detection method based on slope variance is proposed, which includes:

[0057] Step S1: Global Sparse Coarse Scan. Based on the aperture and sub-aperture diameter of the optical system or component under test, the algorithm automatically calculates the number and position coordinates of sub-apertures required for a low-overlap sparse coarse scan, and acquires the corresponding wavefront data using a wavefront detector.

[0058] Step S2: Local gradient calculation and variance evaluation. The collected sparse coarse scan data is initially stitched together to obtain the global slope field and calculate the local slope variance. Specific steps include:

[0059] S2-1. Calculate the local slope magnitude matrix based on the global slope field obtained after preliminary stitching:

[0060] ;

[0061] In the formula This is a local slope magnitude matrix. , For preliminary splicing or local areas , Slope data for the direction.

[0062] S2-2. Based on the local slope magnitude matrix, calculate the slope variance of each local region. As an assessment factor for high-risk areas:

[0063] ;

[0064] In the formula For the slope variance, To evaluate the number of valid pixels within a local evaluation region, For the traversal index of the pixel (from 1 to ... ), To assess the first in the region The slope magnitude of each pixel To assess all within the region The mathematical average of the slope magnitudes of each pixel.

[0065] Step S3: High-risk area identification and targeted encryption. Regions with extremely high spatial frequency and drastic changes in local slope gradient are marked as high-risk areas. Based on these regions, fine-scan coordinates are generated to increase the local overlap rate. Specific steps include:

[0066] S3-1, Selecting the first global slope variance data One quantile, used as a dynamic threshold The high-risk area is the local slope variance. The spatial region.

[0067] S3-2. Determine the high-risk baseline coordinates and neighborhood set. Assume that the retrieved high-risk baseline coordinates are... The region where the coarse scan sub-aperture is located conforms to Then, the center coordinates of the coarse scan sub-aperture are extracted as the high-risk reference coordinates, denoted as... Then, in the coarse scan sub-aperture, the high-risk reference coordinates are retrieved. For all adjacent sub-apertures, mark their center coordinates as follows: The set of all neighboring sub-apertures that meet the conditions is called the neighborhood set, denoted as . .

[0068] If a sub-aperture meets the following formula, then the sub-aperture is considered a neighboring sub-aperture:

[0069] ;

[0070] In the formula, This is the step distance for the coarse scan. For mechanical tolerance.

[0071] S3-3. Determine the new target fine scan coordinates according to the following formula. .

[0072] ;

[0073] ;

[0074] The above formula means that the high-risk reference coordinates Its neighborhood set Coordinates of the center of all neighboring sub-apertures Perform midpoint calculation on the vector.

[0075] In S3-3, the geometric midpoint is used to calculate the targeted fine scan coordinates, thereby improving the local overlap rate. For even higher overlap requirements, the number of fine scan sub-apertures can be dynamically increased through multiple recursive interpolations. If the sparse coarse scan is arranged in an equilateral triangular mesh, the coordinates of three adjacent high-risk coarse scan sub-apertures can be used to calculate the corresponding targeted fine scan coordinates using the geometric centroid formula. Based on different initial mesh arrangements, the corresponding targeted fine scan coordinates can be generated using methods such as geometric center and higher-order interpolation.

[0076] S3-4. Perform pupil mask boundary constraints to prevent some coordinates of the system-generated target fine scanning coordinates from exceeding the effective physical aperture range of the device under test:

[0077] ;

[0078] In the formula The diameter of the component under test is denoted as . To allow for a preset safety margin, the target precision scan coordinates satisfy this formula. Perform a subsequent scan to remove coordinates that are not satisfied.

[0079] Step S4: Adaptive Targeted Fine Scanning Data Acquisition. The displacement stage is controlled to acquire the corresponding targeted fine scanning sub-aperture data at the generated targeted fine scanning coordinate points.

[0080] Step S5: The targeted fine scanning sub-aperture data and sparse coarse scanning sub-aperture data collected in steps S1 and S4 are fused together, and the final stitching is performed using the least squares average error and Zernike mode method to obtain the original aperture wavefront data.

[0081] ;

[0082] The formula is the key formula for least squares equalization error, where Let be the sub-aperture pose compensation vector to be solved. Sub-aperture overlap topology connectivity matrix The generalized inverse matrix, This is the relative slope error vector of neighboring sub-apertures within the overlapping region.

[0083] ;

[0084] The formula is the key formula of the Zernike pattern method, where Let be the Zernike mode coefficient vector to be solved. This is the partial derivative matrix of Zernike polynomial theory. This is the global slope column vector.

[0085] Example:

[0086] An adaptive sub-aperture stitching wavefront detection method based on slope variance, comprising the following steps:

[0087] In this embodiment, the wavefront parameters were measured using a wavefront detector of the corresponding aperture, and the RMS value was 1.1447. The PV value is 6.3529. We treat it as the truth value for easy comparison.

[0088] Step S1: Global sparse coarse scan. Based on the aperture of the device under test (90mm) and the sub-aperture aperture (50mm), the algorithm automatically calculates the 7 sub-apertures required for a sparse coarse scan with a 20% overlap rate. The position coordinates are shown in Table 1. Then, the corresponding wavefront data are acquired by a wavefront detector.

[0089] Table 1. Coordinates of sparse coarse scan sub-aperture positions

[0090]

[0091] Step S2: Local gradient calculation and variance evaluation. The collected sparse coarse scan data is initially stitched together to obtain the global slope field and calculate the local slope variance. Specific steps include:

[0092] S2-1. Calculate the local slope magnitude matrix based on the global slope field obtained after preliminary stitching:

[0093] ;

[0094] In the formula This is a local slope magnitude matrix. , For preliminary splicing or local areas , Slope data for the direction.

[0095] S2-2. Based on the local slope magnitude matrix, calculate the slope variance of each local region. As an assessment factor for high-risk areas:

[0096] ;

[0097] In the formula For the slope variance, To evaluate the number of valid pixels within a local evaluation region, For the traversal index of the pixel (from 1 to ... ), To assess the first in the region The slope magnitude of each pixel To assess all within the region The mathematical average of the slope magnitudes of each pixel.

[0098] The local slope variances corresponding to each sub-aperture are shown in Table 2:

[0099] Table 2. Local slope variance table for each sub-aperture

[0100]

[0101] Step S3: High-risk area identification and targeted encryption. Regions with extremely high spatial frequency and drastic changes in local slope gradient are marked as high-risk areas. Based on these regions, fine-scan coordinates are generated to increase the local overlap rate. Specific steps include:

[0102] S3-1. Select the 75th percentile of the global slope variance data, i.e., 2.36e-14, as the dynamic threshold. The high-risk area is the local slope variance. The spatial region.

[0103] S3-2. Determine the high-risk baseline coordinates and neighborhood set. Assume that the retrieved high-risk baseline coordinates are... The region where the coarse scan sub-aperture is located conforms to Then, the center coordinates of the coarse scan sub-aperture are extracted as the high-risk reference coordinates, denoted as... Then, in the coarse scan sub-aperture, the high-risk reference coordinates are retrieved. For all adjacent sub-apertures, mark their center coordinates as follows: The set of all neighboring sub-apertures that meet the conditions is called the neighborhood set, denoted as . .

[0104] If a sub-aperture meets the following formula, then the sub-aperture is considered a neighboring sub-aperture:

[0105] ;

[0106] In the formula, This is the step distance for the coarse scan. For mechanical tolerance.

[0107] The high-risk reference coordinates were ultimately determined to be the center coordinates of sub-apertures 3 and 6, which are (-20, 34.64) and (20, 34.64). The neighborhood set was determined to include sub-apertures 2, 3, 5, 6, and 7.

[0108] S3-3. Determine the new target fine scan coordinates according to the following formula. .

[0109] ;

[0110] ;

[0111] The above formula means that the high-risk reference coordinates Its neighborhood set Coordinates of the center of all neighboring sub-apertures Perform midpoint calculation on the vector.

[0112] The coordinates of the five newly added sub-apertures are shown in Table 3.

[0113] Table 3. Newly Added Targeted Fine Scanning Sub-Aperture Position Coordinates

[0114]

[0115] S3-4. Perform pupil mask boundary constraints to prevent some coordinates of the system-generated target fine scanning coordinates from exceeding the effective physical aperture range of the device under test:

[0116] ;

[0117] In the formula The diameter of the component under test is denoted as . To allow for a preset safety margin, the target precision scan coordinates satisfy this formula. Perform a subsequent scan to remove coordinates that are not satisfied.

[0118] Step S4: Adaptive Targeted Fine Scanning Data Acquisition. The displacement stage is controlled to acquire the corresponding targeted fine scanning sub-aperture data at the generated targeted fine scanning coordinate points.

[0119] Step S5: The targeted fine-scan sub-aperture data and sparse coarse-scan sub-aperture data collected in steps S1 and S4 are fused together. The final stitching is performed using least-squares average error reduction and the Zernike mode method to obtain the original aperture wavefront data. The final data comparison is shown below.

[0120] like Figure 3 As shown in the figure, the sub-aperture arrangement is divided into three groups, including:

[0121] Group A: Global fine scan;

[0122] Group B: Sparse coarse scan;

[0123] Group C: Coarse-fine fusion, which means adding a new targeted fine-scanning sub-aperture on the basis of sparse coarse scanning sub-apertures.

[0124] Figure 4 The figure shows a local slope gradient. As can be seen from the figure, the slope gradient at the top is significantly greater than that at other locations, which is the basis for selecting high-risk sub-apertures.

[0125] Figure 5 The following is a comparison of the original wavefront true value and the wavefront phase under the three sampling strategies:

[0126] The RMS for Group A is 1.1445. The PV is 6.3768. .

[0127] The RMS for Group B is 1.1402. The PV is 6.2912 .

[0128] The RMS for Group C is 1.1438. The PV is 6.3904 .

[0129] Figure 5 The diagram contains wavefront phase maps under three sampling strategies. It represents the wavefront phase characteristics reconstructed by stitching together the data, with colors indicating phase variations. The given RMS values ​​represent the average statistical deviation of all points on the entire wavefront from the ideal state, while the PV values ​​represent the absolute value of the maximum phase difference, reflecting the extreme maximum deviation of the wavefront.

[0130] Figure 6 The following is a comparison of the reconstructed wavefront residuals under three sampling strategies:

[0131] The RMS of the residuals for group A is 0.00966. The residual PV is 0.0958. .

[0132] The RMS of the residuals for group B is 0.02417. The residual PV is 0.1474. .

[0133] The RMS of the residuals for group C is 0.01145. The residual PV is 0.0986. .

[0134] The data above shows that Group A, with its global fine scanning, achieved the highest spatial sampling rate of 0.00966. Meanwhile, the residual RMS of group C, which used a coarse-fine cascade targeted resampling scheme, reached 0.01145 despite a 37% reduction in the sampling sub-aperture size. This significantly outperforms sparse coarse scanning and is comparable to global fine scanning. This verifies the invention's ability to balance stitching accuracy and efficiency, improving efficiency while maintaining a certain level of stitching accuracy, reducing equipment requirements, and effectively decreasing testing costs.

Claims

1. An adaptive sub-aperture splicing wavefront detection method based on slope variance, characterized in that, The method includes the following steps: Step S1: Global sparse coarse scan; The number and position of sub-apertures are determined based on the low overlap rate. The wavefront of a large-aperture optical system or component is divided into several sub-apertures, and the displacement stage is controlled to collect data sequentially. Step S2: Local gradient calculation and variance evaluation; The data obtained from the sparse coarse scan are initially stitched together to extract the global slope field and calculate the local slope variance. Step S3: High-risk area identification and targeted encryption; Regions with extremely high spatial frequency and drastic changes in local slope gradient are marked as high-risk regions. Based on these regions, fine-scan coordinates are generated to increase the local overlap rate. Step S4: Adaptive targeted fine scanning data acquisition; The control stage collects the corresponding target fine scan sub-aperture data at the generated target fine scan coordinate points; Step S5: Fuse the sparse coarse scan data with the targeted fine scan data, and finally stitch them together using the least squares average error and Zernike mode method to obtain the original aperture wavefront data.

2. The adaptive sub-aperture stitching wavefront detection method based on slope variance according to claim 1, characterized in that, The specific steps of step S2 are as follows: S2-1. Calculate the local slope magnitude matrix based on the global slope field obtained after preliminary stitching: ； In the formula This is a local slope magnitude matrix. , For preliminary splicing or local areas , Slope data of the direction; S2-2. Based on this amplitude matrix, the slope variance in each local region can be calculated. As an assessment factor for high-risk areas: ； In the formula For the slope variance, To evaluate the number of valid pixels within a local evaluation region, For the traversal index of the pixel (from 1 to ... ), To assess the first in the region The slope magnitude of each pixel To assess all within the region The mathematical average of the slope magnitudes of each pixel.

3. The adaptive sub-aperture splicing wavefront detection method based on slope variance according to claim 1, characterized in that, The specific steps of step S3 are as follows: S3-1, Selecting the first global slope variance data quantiles, used as dynamic thresholds High-risk areas are local slope variances. Spatial region; S3-2. Determine the high-risk baseline coordinates and neighborhood set. Assume that the retrieved high-risk baseline coordinates are... The region where the coarse scan sub-aperture is located conforms to Then, the center coordinates of the coarse scan sub-aperture are extracted as the high-risk reference coordinates, denoted as... Then, in the coarse scan sub-aperture, the high-risk reference coordinates are retrieved. For all adjacent sub-apertures that are adjacent to each other, mark their center coordinates as follows: The set of all neighboring sub-apertures that meet the conditions is called the neighborhood set, denoted as . ; If a sub-aperture meets the following formula, then the sub-aperture is considered a neighboring sub-aperture: ； In the formula, This is the step distance for the coarse scan. For mechanical tolerance; S3-3. Determine the new target fine scan coordinates according to the following formula. ; ； ； The above formula means that the high-risk reference coordinates Its neighborhood set Coordinates of the center of all neighboring sub-apertures Calculate the midpoint of the vector; S3-4. Perform pupil mask boundary constraints to prevent some coordinates of the system-generated target fine scanning coordinates from exceeding the effective physical aperture range of the device under test: ； In the formula The diameter of the component under test is denoted as . To provide a preset safety margin; the target precision scanning coordinates satisfy this formula. Perform a subsequent scan to remove coordinates that are not satisfied.

4. The adaptive sub-aperture stitching wavefront detection method based on slope variance according to claim 1, characterized in that, The specific steps of step S5 are as follows: After collecting data again in step S4, the data is combined with the initial data collected in S1. The original wavefront can be restored by adding new targeted fine scan data and original coarse scan data. The spliced ​​wavefront is obtained by using the least squares averaging error and Zernike mode method. ； The formula is the key formula for least squares equalization error, where Let be the sub-aperture pose compensation vector to be solved. Sub-aperture overlap topology connection matrix The generalized inverse matrix, This is the relative slope error vector of neighboring sub-apertures within the overlapping region; ； The formula is the key formula of the Zernike pattern method, where Let be the Zernike mode coefficient vector to be solved. This is the partial derivative matrix of Zernike polynomial theory. This is the global slope column vector.

Images