Computed hologram efficient compression coding method, verification method and file generation method

Abstract

The invention discloses a computer-generated hologram efficient compression coding method, a computer-generated hologram verification method and a computer-generated hologram file generation method, belongs to the field of optical detection, and solves the technical problems of low calculation speed, low controllability of coding precision and coding file size and low compression efficiency caused by unreasonable error estimation before coding in the prior art. Obtaining a stripe line distribution schematic diagram based on a CGH phase distribution function; aiming at stripe lines in different forms in the stripe line distribution schematic diagram, respectively determining sampling search modes corresponding to the stripe lines, carrying out wavefront error pre-evaluation analysis before coding, and determining sampling parameters; sampling the stripe line by using a sampling search mode and a sampling parameter corresponding to the stripe line, and obtaining discrete sampling points through Newton iteration calculation; and distance-limiting thinning processing is carried out on the discrete sampling points by using a DP algorithm to obtain a stripe line sparse point set, and coding compression is completed. The method is used for realizing the controllability and balance of the coding precision and the size of the coding file.

Description

Technical Field

[0001] This invention relates to the field of optical inspection, specifically to a method for efficient compression encoding, verification, and file generation of computational holograms. Background Technology

[0002] Computational holograms (CGHs), as core compensation devices for zero-position compensation detection of freeform surfaces, have seen rapid development in related technologies in recent years. Encoding is a crucial step in transitioning them from optical design to actual fabrication. The main idea of ​​encoding is to use a pre-designed phase function to solve for all phase contour lines with a phase interval of π, thereby constructing a stripe pattern and ultimately converting it into a layout file that can be input into a photolithography machine. However, since the processing files that machines can recognize can only consist of discrete data points, not continuous curve functions, the most critical step in encoding is to obtain a suitable set of discrete sampling points that can describe the stripe lines. But the stripe feature size is much smaller than the CGH aperture, leading to problems such as excessively large encoding data volume and low encoding efficiency. Therefore, a suitable encoding method that can achieve efficient compression is needed.

[0003] Currently, mainstream compression coding methods suffer from low compression efficiency and slow computation speed. For example, in the prior art, Chinese patent document CN107705338A discloses a "high-precision detection computational hologram coding method based on non-maximum suppression," which uses the Teh-Chin non-maximum suppression algorithm to achieve compression coding. This algorithm is characterized by identifying and retaining local curvature maxima of the digital curve, deleting non-maximum points, achieving compression thinning, and increasing data compression efficiency through multiple iterations. However, this technical solution is complex in principle and only uses the maximum coding wavefront error under extreme conditions for global estimation. The error estimation method is not reasonable enough, resulting in inappropriate selection of sampling step size, which greatly increases coding complexity, significantly increases computational resource consumption, and reduces the controllability of coding accuracy and encoded file size.

[0004] Qiu Hongwei et al. proposed a "variable step size search computational hologram encoding method", which sets an error tolerance based on the correspondence between encoding error, phase gradient, and sampling step size. Although the sampling step size is dynamically changed according to the phase gradient, the compression efficiency of this technical solution is low, resulting in an excessive amount of encoded data, which makes data transmission and reading inconvenient.

[0005] In summary, existing technologies suffer from slow computation speed, low controllability of encoding accuracy and file size, and low compression efficiency due to unreasonable error estimation before encoding. Therefore, it is essential to propose an encoding method that achieves high-efficiency compression, improves encoding speed by combining it with encoding error estimation, and achieves a controllable balance between encoding accuracy and file size. Summary of the Invention

[0006] This invention solves the technical problems of existing technologies, such as slow calculation speed, low controllability of encoding accuracy and encoded file size, and low compression efficiency due to unreasonable error estimation before encoding.

[0007] The present invention provides an efficient compression coding method for computational holograms, comprising the following steps:

[0008] Step 1: Obtain the phase distribution function of CGH;

[0009] Step 2: Based on the phase distribution function of CGH, obtain a schematic diagram of the stripe distribution;

[0010] Step 3: For stripes of different shapes in the stripe distribution diagram, determine the corresponding sampling search method, perform wavefront error pre-evaluation analysis before encoding, and determine the sampling parameters.

[0011] Step 4: Sample the stripe lines using the sampling search method and sampling parameters described in Step 3, and obtain discrete sampling points through Newton's iteration calculation.

[0012] Step 5: Use the DP algorithm to perform distance-limited thinning processing on the discrete sampling points to obtain a sparse point set of stripe lines, thus completing the encoding compression.

[0013] Furthermore, in this embodiment of the invention, the CGH-based phase distribution function in step 2, which yields a schematic diagram of the stripe distribution, is specifically as follows:

[0014] By setting a low-density display factor to compress the phase distribution function of the CGH, a schematic diagram of the stripe distribution is obtained.

[0015] Furthermore, in this embodiment of the invention, step 3, which involves determining the corresponding sampling search method for different morphologies of stripes in the stripe distribution diagram, specifically involves:

[0016] The different types of stripes include stripes with no rotation in the x-direction, no rotation in the y-direction, and stripes with closed rotation.

[0017] For stripes without rotation in the x-direction, sampling is performed along the x-direction; for stripes without rotation in the y-direction, sampling is performed along the y-direction; and for stripes with rotational closure, sampling is performed using polar coordinates.

[0018] Furthermore, in this embodiment of the invention, step 3, which involves performing a pre-evaluation analysis of the wavefront error before encoding to determine the sampling parameters, specifically includes:

[0019] Based on the phase distribution function of CGH, the fringe curvature radius distribution and fringe slope distribution are obtained. The sampling step size is then set to obtain the sampling position difference δ. s (x,y), obtain the distance difference δ between the encoded curve and the actual curve caused by the DP algorithm. DP The difference between the encoded curve and the actual curve caused by the DP algorithm should satisfy the following:

[0020] δ DP ≤t;

[0021] Where t is the threshold of the DP algorithm;

[0022] The conditions that the coding wavefront error should meet are:

[0023]

[0024] Where m is the set diffraction order, λ is the light wavelength, and D(x,y) is the fringe density distribution;

[0025] Based on the wavefront error pre-evaluation analysis before encoding described above, sampling parameters can be determined according to the encoding accuracy requirements. These sampling parameters include the sampling step size L. s And the threshold t of the DP algorithm.

[0026] Furthermore, in this embodiment of the invention, the sampling along the x-direction specifically refers to:

[0027] Based on the phase distribution function of CGH, the highest value L of the fringe line order is obtained. max The lowest value L of the fringe order min Set the sampling step size for the interval [L] min ,L max For each order of the stripe line in the image, iterate through the horizontal coordinate x and solve for the corresponding vertical coordinate y to obtain the discrete sampling points sampled along the x-direction;

[0028] The sampling along the y-direction specifically refers to:

[0029] Using the same method as sampling along the x-direction, traverse the y-coordinates and solve for the corresponding x-coordinates to obtain the discrete sampling points along the y-direction.

[0030] Furthermore, in this embodiment of the invention, the sampling using polar coordinates specifically refers to:

[0031] Determine the poles in polar coordinates, set the sampling step size, and for the interval [L]... min ,L max For each order of the stripe line in the equation, the polar angle is traversed in the interval [0, 2π], and the corresponding polar radius is solved to obtain the discrete sampling points in polar coordinates.

[0032] Furthermore, in this embodiment of the invention, the distance-limited thinning process in step 5 specifically includes:

[0033] Set a thinning threshold, connect the first and last sampling points of each stripe line to obtain the corresponding connecting lines, calculate the maximum distance from the sampling point on each stripe line to the corresponding connecting line, and determine whether to delete the sampling point based on the maximum distance. The sampling points that are not deleted are taken as the sparse point set of the stripe line.

[0034] Furthermore, in this embodiment of the invention, the method of determining whether to delete a sampling point based on the maximum distance specifically refers to:

[0035] Determine whether the maximum distance from the sampling point on each stripe to the corresponding connecting line is less than the thinning threshold. If so, delete all sampling points between the two connecting sampling points on the corresponding stripe. If not, divide the corresponding stripe into two segments at the sampling point corresponding to the maximum distance and re-determine until the maximum distance from the sampling point on each stripe to the corresponding connecting line is less than the thinning threshold, and then end the determination.

[0036] The coded wavefront error verification method of this invention is implemented based on any of the methods described above, specifically as follows:

[0037] Based on the phase distribution function of CGH, it is possible to construct the encoded wavefront error function at any point on each sampling edge (the line segment interval formed by each adjacent sampling point):

[0038]

[0039] in, Let be the phase distribution function of CGH, (x,y) be a point on the sampling edge, and L be the fringe order;

[0040] The maximum value of the encoded wavefront error function on each sampling edge is taken as the encoded wavefront error of the corresponding sampling edge. By traversing all sampling edges, the encoded wavefront error distribution of the entire holographic region is obtained.

[0041] The method for generating encoded processing files according to the present invention is implemented based on any of the methods described above, specifically as follows:

[0042] After encoding and compression, the sampled points from the sparse point set of the stripe lines are used to construct stripe polygons and written into the GDSII file to obtain the encoded processing file.

[0043] This invention solves the technical problems of existing technologies, such as slow calculation speed, low controllability of encoding accuracy and encoded file size, and low compression efficiency due to unreasonable error estimation before encoding. Specific beneficial effects include:

[0044] This invention proposes an efficient compression coding method for computational holograms. Based on the relationship between the coding wavefront error and the morphological characteristics, fringe density, and sampling step size, an error pre-evaluation method is given before coding. By reasonably setting the sampling step size, discrete sampling calculation of the CGH coding curve is performed, thereby reducing sampling requirements, avoiding waste of computational resources, and achieving fast coding. Subsequently, taking advantage of the characteristics of the Douglas-Puk algorithm (DP) to achieve limited-distance thinning of the curve and the proportional relationship between coding error and curve position error, the DP algorithm is applied to CGH coding to perform limited-distance thinning of the high-density curve after pre-sampling. This can achieve compression processing under a set error threshold. This method not only has high compression efficiency, but also achieves a balance and controllability between the size of the encoded file and the coding accuracy. Attached Figure Description

[0045] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0046] Figure 1 This is a flowchart of the efficient compression coding method for computational holograms as described in Implementation Method 1;

[0047] Figure 2 This is a schematic diagram of the encoding position error calculation under rectangular coordinate sampling (taking sampling along x as an example) as described in Implementation Method 3;

[0048] Figure 3 This is a schematic diagram of the calculation of encoding position error under polar coordinate sampling as described in Implementation Method 3;

[0049] Figure 4 This is a schematic diagram of the DP algorithm calculation process described in Implementation Method 4;

[0050] Figure 5 This is a schematic diagram of the main holographic stripe distribution with a display density of 1 / 300 as described in Embodiment 7;

[0051] Figure 6 This is a schematic diagram of the aligned holographic stripe distribution with a display density of 1 / 1000 as described in Embodiment 7;

[0052] Figure 7 This is a schematic diagram of the holographic stripe distribution with a display density of 1 / 20 as described in Embodiment 7;

[0053] Figure 8 The file size of the main holographic compressed file and the corresponding coding wavefront error under different DP thresholds as described in Implementation Method 7;

[0054] Figure 9 It is the master holographic encoded wavefront error distribution for the final evaluation and verification described in Implementation Method Seven;

[0055] Figure 10 The surface shape detection result of the CGH generated by the encoding described in Implementation Method 7 is the surface shape detection result of the aspherical surface to be inspected. Detailed Implementation

[0056] Various embodiments of the present invention will now be clearly and completely described with reference to the accompanying drawings. The embodiments described with reference to the drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0057] Implementation Method 1. A method for efficient compression coding of computational holograms, comprising the following steps:

[0058] Step 1: Obtain the phase distribution function of CGH;

[0059] Step 2: Based on the phase distribution function of CGH, obtain a schematic diagram of the stripe distribution;

[0060] Step 3: For stripes of different shapes in the stripe distribution diagram, determine the corresponding sampling search method, perform wavefront error pre-evaluation analysis before encoding, and determine the sampling parameters.

[0061] Step 4: Sample the stripe lines using the sampling search method and sampling parameters described in Step 3, and obtain discrete sampling points through Newton's iteration calculation.

[0062] Step 5: Use the DP algorithm to perform distance-limited thinning processing on the discrete sampling points to obtain a sparse point set of stripe lines, thus completing the encoding compression.

[0063] Existing technologies suffer from technical problems such as slow calculation speed, low controllability of encoding accuracy and encoded file size, and low compression efficiency due to unreasonable error estimation before encoding.

[0064] To address the aforementioned technical problems, this embodiment provides an efficient compression coding method for computational holograms, such as... Figure 1 As shown, step 1 obtains the continuous optical field phase distribution function of the CGH through optical design. To determine the basic characteristics of the CGH stripe lines, step 2 involves observing the actual morphology of the stripe lines, setting a low-density display factor α (α<1), and adjusting the phase distribution function. To compress the stripes, you can use the following formula to calculate and view the stripe pattern:

[0065]

[0066] Therefore, this implementation utilizes the relationship between sampling parameters, fringe curve morphology, and sampling position error to pre-evaluate the coding wavefront error, thereby setting appropriate sampling parameters. While ensuring the accuracy of discrete sampling points, it accelerates the coding speed. Furthermore, based on the characteristic that the CGH coding wavefront error is proportional to the curve position error, it performs distance-limited thinning processing on the discrete sampling points after CGH coding pre-sampling. This compression processing under a set error threshold can achieve a balance and controllability between the coding file size and coding accuracy. It solves the technical problems of existing technologies, such as slow calculation speed, low controllability of coding accuracy and coding file size, and low compression efficiency due to unreasonable error estimation before coding.

[0067] Implementation Method 2. This implementation method further defines the efficient compression coding method for computational holograms described in Implementation Method 1. In step 3, for different morphologies of stripes in the stripe distribution diagram, the corresponding sampling search method is determined respectively, specifically as follows:

[0068] The different types of stripes include stripes with no rotation in the x-direction, no rotation in the y-direction, and stripes with closed rotation.

[0069] For stripes without rotation in the x-direction, sampling is performed along the x-direction; for stripes without rotation in the y-direction, sampling is performed along the y-direction; and for stripes with rotational closure, sampling is performed using polar coordinates.

[0070] By solving the sampling equation using Newton's iterative method, we obtain each discrete sampling data point of the fringe curve under set sampling parameters. The fringe lines in CGH are essentially the phase contour lines of the phase function of its optical field, meaning that the points (x, y) on the fringe lines should satisfy the nonlinear equation. Where L = 0, ±1, ±2, ..., L represents the fringe order. Different sampling search methods are set according to different fringe curve shapes.

[0071] The sampling along the x-direction, based on the phase distribution function of CGH, yields the highest value L of the fringe line order. max The lowest value L of the fringe order min Set the sampling step size for the interval [L] min ,L max For each order of the stripes in the diagram, iterate through the x-coordinates and solve for the corresponding y-coordinates to obtain the discrete sampling points along the x-direction, specifically:

[0072] Based on the phase distribution function of CGH, the highest value L of the fringe line order is obtained. max The lowest value L of the fringe order min In this context, the stripe lines refer to the two side lines of each stripe. Two adjacent stripe lines form a first-order stripe, as shown in the following formula:

[0073]

[0074] in, This represents the maximum value of the phase distribution. This represents the minimum value of the phase distribution. The sampling step size L is set. s For the interval [L] min ,L max For each stripe order in the image, the horizontal coordinate x is traversed with a set sampling step size, and then the corresponding vertical coordinate y is solved through Newton's iteration to obtain the discrete sampling points in the x-direction, as shown in the following formula:

[0075]

[0076] in, y0 is the sign of the partial derivative and y0 is the initial value of the iteration.

[0077] The sampling along the y-direction uses the same method as the sampling along the x-direction: traversing the ordinate y, solving for the corresponding abscissa x, to obtain the discrete sampling points along the y-direction, specifically:

[0078] Using the same sampling method along the x-direction, the sampling step size L is set. s For the interval [L] min ,L max For each stripe order in the equation, the vertical coordinate y is traversed with a set sampling step size, and then the corresponding horizontal coordinate x is solved through Newton's iteration to obtain the discrete sampling points in the y direction, as shown in the following equation:

[0079]

[0080] The method described above uses polar coordinates for sampling, determines the polar poles, and sets the sampling step size for the interval [L]. min ,L max For each order of the fringe line in the diagram, the polar angle is traversed over the interval [0, 2π], and the corresponding polar radius is calculated to obtain the discrete sampling points in polar coordinates. Specifically:

[0081] Determine the phase extremum point (x) c ,y c As the pole of the polar coordinate system, the sampling step size L is set. s For the interval [L] min ,L max For each order of the fringe line in the image, the polar angle θ is traversed over the interval [0, 2π]. The equation is solved using Newton's iteration to obtain the polar radius r, and the polar angle θ = θ + L for the next sampling is updated simultaneously. s / r, and so on, to perform the calculation until all samples are collected, as shown in the following formula:

[0082]

[0083] By iteratively calculating the above formula, high-precision discrete sampling points can be obtained.

[0084] Implementation Method 3. This implementation method further defines the efficient compression coding method for computational holograms described in Implementation Method 1. Specifically, step 3, which involves pre-evaluating and analyzing the wavefront error before coding to determine the sampling parameters, includes:

[0085] Based on the phase distribution function of CGH, the fringe curvature radius distribution and fringe slope distribution are obtained. The sampling step size is then set to obtain the sampling position difference δ. s (x,y), obtain the distance difference δ between the encoded curve and the actual curve caused by the DP algorithm. DP The difference between the encoded curve and the actual curve caused by the DP algorithm should satisfy the following:

[0086] δ DP ≤t;

[0087] Where t is the threshold of the DP algorithm.

[0088] The conditions that the coding wavefront error should meet are:

[0089]

[0090] Where m is the set diffraction order, λ is the wavelength of light, and D(x,y) is the fringe density distribution.

[0091] Based on the wavefront error pre-evaluation analysis before encoding described above, sampling parameters can be determined according to the encoding accuracy requirements. These sampling parameters include the sampling step size L. s And the threshold t of the DP algorithm.

[0092] After determining the sampling search method, the encoded wavefront error can be estimated based on the specific fringe shape, fringe density, and set sampling parameters, as follows:

[0093] First, based on the phase distribution function Taking the partial derivatives with respect to x and y respectively, we can obtain the fringe curvature radius distribution R(x,y) and the fringe slope distribution k(x,y):

[0094]

[0095] Taking sampling along the x-direction as an example, the difference in sampling encoding position is as follows: Figure 2 As shown, specifically:

[0096]

[0097] Similarly, sampling along the y-direction is as follows:

[0098]

[0099] The difference in sampling encoding position in polar coordinate sampling is as follows: Figure 3 As shown, specifically:

[0100]

[0101] Among them, L s This is the sampling step size.

[0102] pass The phase gradient distribution M(x,y) can be obtained by solving, thus yielding the fringe density distribution as D(x,y) = M(x,y) / 2π. This is achieved through a determined sampling search method and sampling step size L. s And the fringe curve shape can be used to estimate the actual sampling location difference δ. s (x, y), according to the principle of the dynamic programming (DP) algorithm, with a threshold of t, the distance difference between the encoded curve and the actual curve caused by the DP algorithm should satisfy:

[0103] δ DP ≤t;

[0104] Therefore, based on the set sampling traversal method, sampling step size, and DP compression threshold, the corresponding encoded wavefront error can be estimated as follows:

[0105]

[0106] Where m is the set diffraction order and λ is the wavelength of light.

[0107] Implementation Method 4. This implementation method further defines the efficient compression coding method for computational holograms described in Implementation Method 1. The range-limited thinning process in step 5 specifically includes:

[0108] Set a thinning threshold, connect the first and last sampling points of each stripe line to obtain the corresponding connecting lines, calculate the maximum distance from each sampling point on each stripe line to the corresponding connecting line, and determine whether the maximum distance from each sampling point on each stripe line to the corresponding connecting line is less than the thinning threshold. If it is, delete all sampling points between the two connected sampling points on the corresponding stripe line. If not, divide the corresponding stripe line into two segments at the sampling point corresponding to the maximum distance, and re-determine until the maximum distance from each sampling point on each stripe line to the corresponding connecting line is less than the thinning threshold. Then, end the determination and use the remaining sampling points as the sparse point set of the stripe line.

[0109] The thinning algorithm used is the Douglas-Puk (DP) algorithm, a classic trajectory data compression algorithm that can perform distance-limited thinning. This algorithm can be used to thin each stripe. The compression process is as follows: Figure 4 As shown. First, set a thinning threshold t, connect the first and last two points A and B of the data curve, and then calculate the distance from all points on the curve to the line connecting points A and B. Calculate the maximum distance. If it is less than the set threshold, all points in between can be removed. Otherwise, the curve is divided into two segments at the maximum distance. Repeat the above operation for each segment of the curve until all distance values ​​are less than t.

[0110] After processing by the above algorithm, a data curve with a positional deviation from the original data curve not exceeding a set threshold t can be obtained by thinning.

[0111] Implementation Method 5. The coding wavefront error verification method of the present invention is implemented based on any one of the methods described in Implementation Methods 1 to 4, specifically as follows:

[0112] Based on the phase distribution function of CGH, it is possible to construct the encoded wavefront error function for any point on each sampling edge (the line segment interval formed by each adjacent sampling point), which is the error caused by insufficient accuracy of polygon fitting of the real curve:

[0113]

[0114] in, Let be the phase distribution function of CGH, (x,y) be a point on the sampling edge, and L be the fringe order;

[0115] The maximum value of the encoded wavefront error function on each sampling edge is taken as the encoded wavefront error of the corresponding sampling edge. By traversing all sampling edges, the encoded wavefront error distribution of the entire holographic region is obtained.

[0116] For example, for adjacent sampling points A i (x i ,y i A) i-1 (x i-1 ,y i-1 The sampling edges formed by the line segment interval A are solved by solving the line segment interval A. i A i-1 The maximum value of this function can be used as an evaluation of the coding wavefront error corresponding to the sampling edge, thereby enabling the solution of the coding wavefront error distribution of the entire holographic region.

[0117] Implementation Method Six. The encoding processing file generation method described in this implementation method is based on the methods described in Implementation Methods One to Four, specifically as follows:

[0118] After encoding and compression, the sampled points from the sparse point set of the stripe lines are used to construct stripe polygons and written into the GDSII file to obtain the encoded processing file.

[0119] The commonly used layout file format for CGH lithography is GDSII, where the basic element is a polygon, meaning the etched stripe areas are composed of polygons. After all areas are encoded, the stripe polygons can be constructed using the sampling points corresponding to each stripe order. Adjacent stripe orders are used to construct first-order stripes, which are then written into the GDSII file.

[0120] Implementation Method Seven. This implementation method takes the design of a CGH (Conducting Ghost) for a large-aperture aspherical mirror as an example to illustrate in detail the specific calculation process using the methods described in Implementation Methods One through Six:

[0121] 1) Based on the CGH phase functions of the designed main hologram, aligned hologram, and marked hologram, calculate the fringe distribution diagrams, as shown below. Figure 5 , Figure 6 , Figure 7 As shown, it can be determined that the main hologram and the marker hologram are sampled using rectangular coordinates, while the alignment hologram is sampled using polar coordinates.

[0122] 2) For each holographic region to be encoded, given a fixed sampling step size and DP threshold, the corresponding encoded wavefront error can be calculated by combining the phase parameters. Appropriate sampling parameters are designed to ensure that the wavefront error remains within a reasonable range. Taking the main holographic region as an example, its diffraction order is 1st, and its minimum linewidth is 3.12 μm. Using rectangular coordinate sampling along the x-direction, with a designed sampling step size of 1.6 μm and a DP threshold of 1 nm, the encoded wavefront error introduced by Newton's iterative sampling can be calculated to be 4.8 × 10⁻⁶. -7 λ, the coding wavefront error introduced by the DP algorithm compression is 1.6 × 10⁻⁶. -4 Therefore, the total coding wavefront error is 1.6 × 10⁻⁶. -4 λ has met the requirements for normal use;

[0123] 3) Based on Newton's iterative solution equation, sampling calculations are performed on the stripe lines of each order according to the set sampling step size and the determined sampling search method to obtain the corresponding set of sampling data points. At this time, the size of the encoded file is 5720.9MB.

[0124] 4) For the discretely sampled data set, thinning was performed using a DP threshold of t = 1 nm. The compressed master holographic encoded file was 457.6 MB, representing a 12.5-fold reduction in file size. The overall computation time for sampling and compression was only 7.65 hours, achieving fast encoding computation. Different DP thresholds were also set for thinning, and the compressed file size and corresponding encoded wavefront errors were shown below. Figure 8As shown, with an error of 10 -2 At a value in the order of λ, the file size can be compressed by up to 139 times, down to 41.2MB, thus achieving a balance and controllability between encoded file size and encoding precision;

[0125] 5) Taking the compression result with a DP threshold of 1nm as an example, the encoded wavefront error is evaluated and verified according to the formula in Implementation Method 5 for the compressed sparse data point set. The encoded wavefront error distribution can be obtained as follows: Figure 9 As shown, the maximum coding wavefront error is 1.6 × 10⁻⁶. -4 λ is very close to the estimated value in step 2;

[0126] 6) After all the holographic regions are encoded, the corresponding stripe line data point sets are connected to form stripe polygons, and finally written into the GDSII file using the relevant software toolkit.

[0127] 7) Using the coded processing file, laser direct writing processing of the CGH was performed, and the aspherical mirror under inspection was tested. The test results are as follows: Figure 10 As shown, this demonstrates that it can meet the application requirements for high-precision detection of optical surfaces.

[0128] The above provides a detailed description of the efficient compression encoding method, verification method, and file generation method for computational holograms proposed in this invention. Specific examples have been used to illustrate the principles and implementation methods of this invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this invention. Therefore, the content of this specification should not be construed as a limitation of this invention.

Claims

1. A method for efficient compression coding of computational holograms, characterized in that, Includes the following steps: Step 1: Obtain the phase distribution function of CGH; Step 2: Based on the phase distribution function of CGH, obtain a schematic diagram of the stripe distribution; Step 3: For stripes of different shapes in the stripe distribution diagram, determine the corresponding sampling search method, perform wavefront error pre-evaluation analysis before encoding, and determine the sampling parameters. Step 4: Sample the stripe lines using the sampling search method and sampling parameters described in Step 3, and obtain discrete sampling points through Newton's iteration calculation. Step 5: Use the DP algorithm to perform distance-limited thinning processing on the discrete sampling points to obtain a sparse point set of stripe lines, thus completing the encoding compression.

2. The efficient compression coding method for computational holograms according to claim 1, characterized in that, The phase distribution function based on CGH in step 2 yields a schematic diagram of the stripe distribution, specifically as follows: By setting a low-density display factor to compress the phase distribution function of the CGH, a schematic diagram of the stripe distribution is obtained.

3. The efficient compression coding method for computational holograms according to claim 1, characterized in that, In step 3, for different morphologies of stripes in the stripe distribution diagram, the corresponding sampling search method is determined respectively, specifically as follows: The different types of stripes include stripes with no rotation in the x-direction, no rotation in the y-direction, and stripes with closed rotation. For stripes without rotation in the x-direction, sampling is performed along the x-direction; for stripes without rotation in the y-direction, sampling is performed along the y-direction; and for stripes with rotational closure, sampling is performed using polar coordinates.

4. The efficient compression coding method for computational holograms according to claim 1, characterized in that, Step 3, which involves pre-evaluating and analyzing the wavefront error before encoding to determine the sampling parameters, specifically includes: Based on the phase distribution function of CGH, the fringe curvature radius distribution and fringe slope distribution are obtained. The sampling step size is then set to obtain the sampling position difference δ. s (x,y), obtain the distance difference δ between the encoded curve and the actual curve caused by the DP algorithm. DP The difference between the encoded curve and the actual curve caused by the DP algorithm should satisfy the following: d DP ≤t; Where t is the threshold of the DP algorithm; The conditions that the coding wavefront error should meet are: Where m is the set diffraction order, λ is the light wavelength, and D(x,y) is the fringe density distribution; Based on the wavefront error pre-evaluation analysis before encoding described above, sampling parameters can be determined according to the encoding accuracy requirements. These sampling parameters include the sampling step size L. s And the threshold t of the DP algorithm.

5. The efficient compression coding method for computational holograms according to claim 3, characterized in that, The sampling along the x-direction specifically refers to: Based on the phase distribution function of CGH, the highest value L of the fringe line order is obtained. max The lowest value L of the fringe order min Set the sampling step size for the interval [L] min ,L max For each order of the stripe line in the image, iterate through the horizontal coordinate x and solve for the corresponding vertical coordinate y to obtain the discrete sampling points sampled along the x-direction; The sampling along the y-direction specifically refers to: Using the same method as sampling along the x-direction, traverse the y-coordinates and solve for the corresponding x-coordinates to obtain the discrete sampling points along the y-direction.

6. The efficient compression coding method for computational holograms according to claim 3, characterized in that, The aforementioned sampling using polar coordinates specifically refers to: Determine the poles in polar coordinates, set the sampling step size, and for the interval [L]... min ,L max For each order of the stripe line in the equation, the polar angle is traversed in the interval [0, 2π], and the corresponding polar radius is solved to obtain the discrete sampling points in polar coordinates.

7. The efficient compression coding method for computational holograms according to claim 1, characterized in that, The distance-limited thinning process in step 5 is specifically as follows: Set a thinning threshold, connect the first and last sampling points of each stripe line to obtain the corresponding connecting lines, calculate the maximum distance from the sampling point on each stripe line to the corresponding connecting line, and determine whether to delete the sampling point based on the maximum distance. The sampling points that are not deleted are taken as the sparse point set of the stripe line.

8. The efficient compression coding method for computational holograms according to claim 7, characterized in that, The method of determining whether to delete a sampling point based on the maximum distance is as follows: Determine whether the maximum distance from the sampling point on each stripe to the corresponding connecting line is less than the thinning threshold. If so, delete all sampling points between the two connecting sampling points on the corresponding stripe. If not, divide the corresponding stripe into two segments at the sampling point corresponding to the maximum distance and re-determine until the maximum distance from the sampling point on each stripe to the corresponding connecting line is less than the thinning threshold, and then end the determination.

9. A method for verifying encoded wavefront errors, wherein the method is implemented based on the method described in any one of claims 1-8, characterized in that, Specifically: Based on the phase distribution function of CGH, the encoded wavefront error function at any point on each sampling edge is constructed: in, Let be the phase distribution function of CGH, (x,y) be a point on the sampling edge, and L be the fringe order; The maximum value of the encoded wavefront error function on each sampling edge is taken as the encoded wavefront error of the corresponding sampling edge. By traversing all sampling edges, the encoded wavefront error distribution of the entire holographic region is obtained.

10. A method for generating encoded processing files, wherein the method is implemented based on the method described in any one of claims 1-8, characterized in that, Specifically: After encoding and compression, the sampled points from the sparse point set of the stripe lines are used to construct stripe polygons and written into the GDSII file to obtain the encoded processing file.

Images