Method for measuring and correcting axial position error of lens array in integrated imaging system
By using sub-pixel search and image element scaling methods in an integrated imaging system, the problems of measuring and correcting the axial position error of the lens array are solved, and high-quality stereoscopic image reconstruction is achieved.
Patent Information
- Application Number
- CN202410143793.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-02-01
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-02-01
AI Technical Summary
In integrated imaging systems, the axial position error of the lens array is difficult to measure and correct, resulting in degradation and distortion of the quality of the reconstruction image, especially the inevitable errors during manual and mechanical assembly of large lens arrays.
By using the reference subpixel and reference voxel as standards, the ideal subpixel coordinates of the same name under the ideal lens are calculated, and the subpixel is searched as the step size to find the actual subpixel with the smallest geometric center distance of the imaging point, and calculate the actual lens Axial position error and image element scaling makes the reconstruction voxel coincide with the reference voxel to achieve correction.
High-precision axial position error measurement and correction are achieved, the image reconstruction quality of the integrated imaging system is improved, and the accuracy and clarity of the reconstruction image are ensured.
Smart Images

Figure CN117956128B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to integrated imaging technology, in particular to a method for measuring and correcting axial position errors of lens arrays in an integrated imaging system. Background Art
[0002] Integrated imaging is a true 3D stereo display technology that can reconstruct the light field distribution of a 3D object in space, presenting a 3D image with full parallax information and continuous viewpoints, without stereoscopic viewing fatigue, and has become a research hotspot. Integrated imaging includes two stages: recording and displaying. Figure 1 As shown. In the recording stage, the lens array records the light field information of the three-dimensional object to form a micro-image array; in the display stage, the micro-image array is projected through the lens array to reconstruct the image of the three-dimensional object in space. According to the principle of reversibility of the optical path, a natural three-dimensional image can only be correctly reconstructed when the recording and display devices are consistent in structural parameters. Therefore, the optical modulation characteristics of the lens arrays in the two stages should be the same. Even small errors, such as misalignment and deformation of the lens array, will lead to deterioration and distortion of the reconstructed three-dimensional image quality. For large lens arrays, lateral position errors and axial position errors are inevitable during the manual and mechanical assembly process, which will cause the propagation direction of the reconstructed light to shift, resulting in a deviation in the intersection point of the reconstructed light and a drift in the position of the voxel, forming a drifting voxel, as shown in the attached figure. Figure 2 As shown, the quality of the reconstructed image is affected. Therefore, it is very important to measure and correct the errors in the lens array.
[0003] In practical applications, a high-precision substrate can be used to fix each lens to eliminate the influence of lateral position error. As for axial position error, the axial position error of large lens arrays is still difficult to solve due to the unavoidable deformation of the lens array caused by manual installation error and gravity. Summary of the invention
[0004] The present invention provides a method for measuring and correcting the axial position error of a lens array in an integrated imaging display system, the method comprising two processes: axial position error measurement and micro-image array correction. In the axial position error measurement process, the coordinate position of the ideal same-name sub-pixel 10 corresponding to the ideal lens 9 is calculated based on the reference sub-pixel 15 and the reference voxel 7; a search is performed around the ideal same-name sub-pixel 10 with a sub-pixel as a step length, and the sub-pixel that satisfies the minimum geometric center distance between the imaging point at the center depth plane and the imaging point of the reference sub-pixel 15 is the actual same-name sub-pixel 12, and the axial position error of the actual lens 8 is calculated based on the coordinate position of the actual same-name sub-pixel 12 and the ideal same-name sub-pixel 10. In the micro-image array correction process, the image element scaling factor is calculated based on the positional relationship between the actual same-name sub-pixel 12 and the ideal same-name sub-pixel 10 obtained by the search, and the axial position error of the lens array, and the image element is scaled to make the reconstructed voxel in space coincide with the reference voxel, and finally achieve the purpose of correcting the axial position error of the lens array.
[0005] The axial position error measurement process is as follows Figure 3 As shown, the horizontal direction of the display screen is the X axis, the direction perpendicular to the display screen is the Z axis, and the first pixel in the upper left corner of the display screen is the coordinate origin to establish a Cartesian coordinate system. A lens is selected as the reference lens 14, and the coordinates are (X lens-ref ,Z lens-ref ), and corresponds to the central sub-pixel coordinate (X subpx-ref ,0) as the reference sub-pixel 15, and the image point thereof in the central depth plane is set as the reference voxel 7, with the coordinates (X vx-ref ,Z vx-ref According to the position of the reference sub-pixel 15 under the reference lens 14, the position coordinates X(x) of a group of ideal sub-pixels 10 with the same name under the ideal lens array condition can be obtained. subpx-1 ,x subpx-2 ,...,x subpx-i ,...), where x subpx-1 represents the ideal sub-pixel position coordinates of the same name under the first lens, x subpx-i represents the position coordinates of the ideal sub-pixel with the same name under the i-th lens. The position coordinates of the ideal sub-pixel with the same name 10 can be obtained by the following formula.
[0006]
[0007] Where X lens-i , Z lens-ref Respectively represent the lateral position coordinate and axial position coordinate of the i-th lens. In the image element where the ideal sub-pixel 10 with the same name is located, a search is performed with a sub-pixel as a step length, as shown in the attached figure. Figure 4As shown in the figure, the distance between the geometric center of the image point in the central depth plane and the geometric center of the reference voxel 7 is compared respectively. Figure 5 The figure shows the distance between the drift voxel 6 and the reference voxel 7 before searching. Figure 6 FIG. 1 is a schematic diagram showing the distance between the voxel 13 and the reference voxel 7 during the search process. Figure 7 The figure shows the overlap of the actual voxel 14 and the reference voxel 7 after the search is completed. The sub-pixel corresponding to the actual voxel is the actual sub-pixel 12 with the same name. Therefore, an actual sub-pixel 12 with the same name can be found near each ideal sub-pixel 10 with the same name, and finally a group of actual sub-pixels with the same name are obtained, and the horizontal position coordinates are X'(x' subpx-1 ,x' subpx-2 ,...,x' subpx-i ,...). Substituting the position coordinates of the actual same-name sub-pixel 12 and the ideal same-name sub-pixel 10 into the following formula, the axial position error of the lens can be calculated.
[0008]
[0009] Where l c represents the distance between the central depth plane and the display screen, and g is the distance between the ideal lens array and the display screen.
[0010] In the process of searching for actual sub-pixels with the same name, the search step is one sub-pixel, so the position deviation of the actual sub-pixels with the same name obtained by searching does not exceed half a sub-pixel. The measurement accuracy δ of the actual axial position of the lens array can be expressed by the following formula.
[0011]
[0012] Preferably, the measurement accuracy is inversely proportional to the distance between the measured lens and the reference lens. In order to improve the measurement accuracy, two rounds of measurement can be performed. In the first round, the lens at the left end of the actual lens array is selected as the reference lens, and the lens at the right end of the actual lens array is measured; in the second round, the lens at the right end of the actual lens array is selected as the reference lens, and the lens at the left end of the actual lens array is measured. Both measurement processes ensure that the distance between the reference lens and the measured lens is large to achieve extremely high measurement accuracy.
[0013] The micro-image array correction principle is as shown in the attached Figure 8 As shown, the actual sub-pixel 12 of the same name has a certain offset relative to the ideal sub-pixel 10 of the same name, so that the image point can coincide with the reference voxel. By scaling the image element, the actual voxel 14 can be made to coincide with the reference voxel 7. The image element scaling factor k i The following formula must be satisfied:
[0014]
[0015] Where Δx' i and Δx i are the distances from the actual sub-pixel 12 and the ideal sub-pixel 10 to the lens optical axis, Z vx-i To reconstruct the distance between the voxel and the display screen, by scaling the image element, the generated corrected micro-image array can still correctly modulate light through the actual lens 8 to obtain a correct 3D image without voxel drift.
[0016] Preferably, the change of the scaling factor is extremely small within the limited depth of field of the integrated imaging. In order to simplify the correction process and improve the correction efficiency, the central depth plane can be selected as the voxel reconstruction position, and the scaling factor is constantized. The constantized scaling factor k cdp It can be expressed by the following formula.
[0017]
[0018] The present invention proposes a method for measuring and correcting the axial position error of a lens array in an integrated imaging system. In the process of measuring the axial position error of the lens array, the method converts the axial position error into the geometric center distance between the image point of the sub-pixel through the actual lens at the central depth plane and the reference voxel, amplifies the axial position error for easy measurement, and searches one by one with sub-pixels as the step size, so as to conveniently, accurately and quickly locate the position of the actual sub-pixels with the same name, thereby achieving extremely high axial position error measurement accuracy; in the process of correcting the micro-image array, the voxel offset caused by the axial position error is compensated by scaling the image element, and the central depth plane is selected to obtain a constant scaling factor, thereby simplifying the process of processing the image element. The method proposed by the present invention can accurately measure the axial position error of the lens array in the integrated imaging display system, correct the axial position error, and achieve high-quality stereoscopic image reconstruction. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] Figure 1 Schematic diagram of the integrated imaging recording and display stage.
[0020] Figure 2 Schematic diagram of voxel drift caused by ideal sub-pixels of the same name being modulated by actual lenses.
[0021] Figure 3 Schematic diagram of axial position error measurement.
[0022] Figure 4 A schematic diagram for searching for actual sub-pixel positions with the same name.
[0023] Figure 5 Schematic diagram of the relationship between the reconstructed voxels and the reference voxels before search.
[0024] Figure 6Schematic diagram of the relationship between the reconstructed voxels and the reference voxels during the search process.
[0025] Figure 7 Schematic diagram of the relationship between the reconstructed voxel and the reference voxel after the search is completed.
[0026] Figure 8 Schematic diagram of image element scaling.
[0027] Fig. 9 This is the result map of aliased voxels.
[0028] Fig.10 This is the result map of overlapping voxels.
[0029] Fig.11 This is the measurement result of the axial position error of the lens array.
[0030] Fig.12 This is the resolution target display result diagram before axial position error correction.
[0031] Fig.13 This is the resolution target display result diagram after axial position error correction.
[0032] The diagram numbers in the above drawings are:
[0033] 1 Three-dimensional object, 2 lens array, 3 micro-image array, 4 reconstructed three-dimensional image, 5 central depth plane, 6 drift voxel, 7 reference voxel, 8 actual lens, 9 ideal lens, 10 ideal same-name sub-pixel, 11 sub-pixel in the search process, 12 actual same-name sub-pixel, 13 voxel in the search process, 14 actual voxel, 15 reference lens, 16 reference sub-pixel, 17 local enlarged image of the resolution target image before correction, 18 local enlarged image of the resolution target image after correction.
[0034] It should be understood that the above drawings are only schematic and are not drawn to scale. DETAILED DESCRIPTION
[0035] A typical embodiment of the method for measuring and correcting the axial position error in the integrated imaging display system using the present invention is described in detail below, and the present invention is further described in detail. It is necessary to point out here that the following embodiments are only used to further illustrate the present invention and cannot be understood as limiting the protection scope of the present invention. Personnel skilled in the art in this field may make some non-essential improvements and adjustments to the present invention based on the above-mentioned content of the present invention, which still fall within the protection scope of the present invention.
[0036] The present invention proposes a method for measuring and correcting the axial position error of a lens array in an integrated imaging display system, and the method includes two processes: axial position error measurement and micro-image array correction. In the axial position error measurement process, the coordinate position of the ideal same-name sub-pixel corresponding to the ideal lens is calculated with reference sub-pixels and reference voxels as standards; a search is performed around the ideal same-name sub-pixel with a sub-pixel as a step size, and the sub-pixel that satisfies the minimum geometric center distance between the imaging point at the center depth plane and the imaging point of the reference sub-pixel is the actual same-name sub-pixel, and the axial position error of the actual lens array is calculated according to the coordinate position of the actual same-name sub-pixel and the ideal same-name sub-pixel. In the micro-image array correction process, the image element scaling factor is calculated according to the position relationship between the actual same-name sub-pixel and the ideal same-name sub-pixel obtained by the search, and the axial position error of the lens array, and the image element is scaled to make the reconstructed voxel in space coincide with the reference voxel, and finally achieve the purpose of correcting the axial position error of the lens array.
[0037] The axial position error measurement process is as follows Figure 3 As shown in the figure, the horizontal direction of the display screen is the X-axis, the direction perpendicular to the display screen is the Z-axis, and the first pixel in the upper left corner of the display screen is the coordinate origin to establish a Cartesian coordinate system. The sub-pixel size of the display screen is 0.03mm, the lens diameter in the ideal lens array is 11mm, the lens spacing is 13mm, the lens focal length is 11mm, the number of lenses is 50×35, the distance between the ideal lens array and the display screen is 11.8mm, and the distance of the central depth plane is 173.8mm. In the first round of measurement, the leftmost lens is selected as the reference lens with coordinates (6.5mm, 11.8mm), and the central sub-pixel coordinates (6.5mm, 0) of its corresponding image element are used as the reference sub-pixel. The image point in the central depth plane is set as the reference voxel with coordinates (6.5mm, 173.8mm), and the lens on the right is measured. According to the reference sub-pixel position under the reference lens, the position coordinates of a group of ideal sub-pixels with the same name under the ideal lens array condition can be obtained using formula (1) X(6.5mm, 20.44mm, 34.39mm, ..., x subpx-i ,...), x subpx-i Represents the ideal sub-pixel position coordinates of the same name under the i-th lens.
[0038]
[0039] In the image element where the ideal sub-pixel with the same name is located, the search is performed with the sub-pixel as the step size, as shown in the attached figure. Figure 4 As shown in the figure, the distance between the geometric center of the image point in the central depth plane and the geometric center of the reference voxel is compared, as shown in the figure. Fig. 9 The figure shows the aliased voxel result of drift voxel and reference voxel before searching. Fig.10After the search is completed, the actual voxel and the reference voxel form a coincident voxel result map. The sub-pixel corresponding to the actual voxel is the actual sub-pixel with the same name. Therefore, an actual sub-pixel with the same name can be found near each ideal sub-pixel with the same name, and finally a group of actual sub-pixels with the same name are obtained, and the horizontal position coordinates are X'(6.5mm, 20.49mm, 33.3mm,..., x' subpx-i ,...). In the second round of measurement, the rightmost lens is selected as the reference lens to measure the left lens. The measurement steps are the same as above to improve the measurement accuracy. Substituting the actual and ideal sub-pixels of the same name into formula (2) can calculate the axial position error. Under the condition that the reference lens and the lens to be measured are 4 times the lens spacing, the difference between the obtained measurement result and the actual value is less than 0.05mm, which is consistent with the analysis of theoretical formula (3). The obtained axial position error result is shown in the attached figure. Fig.11 shown.
[0040]
[0041]
[0042] The micro-image array correction principle is as shown in the attached Figure 8 As shown in the figure, the actual sub-pixels with the same name have a certain offset relative to the ideal sub-pixels with the same name, so that the image point can coincide with the reference voxel. By scaling the image element, the actual voxel can be made to coincide with the reference voxel. The image element scaling factor k i The following formula must be satisfied:
[0043]
[0044] Where Δx' i and Δx i are the distances from the actual sub-pixel with the same name and the ideal sub-pixel with the same name to the optical axis of the lens, Z vx-i To reconstruct the distance between the voxel and the display screen, by scaling the image element, the generated corrected micro-image array can still correctly modulate light through the actual lens to obtain a correct 3D image without voxel drift.
[0045] Preferably, the change of the scaling factor is extremely small within the limited depth of field of the integrated imaging. In order to simplify the correction process and improve the correction efficiency, the central depth plane l can be selected. c = 174 mm is the voxel reconstruction position, and the scaling factor is constantized. The constantized scaling factor k cdp It can be expressed by formula (5). Using the resolution target as a virtual 3D scene, the image is reconstructed through the above integrated imaging system. The resolution target display result before correction is shown in the attached figure. Fig.12As shown, element 4 in group 1 is difficult to identify; the image element is scaled, and the resolution target result after axial position error correction is shown in the attached figure. Fig.13 As shown, element No. 4 in group 1 can be clearly identified. After correction using the method described above, the display quality of the image is significantly improved.
[0046]
[0047] The present invention proposes a method for measuring and correcting the axial position error in integrated imaging. In the process of measuring the axial position error of the lens array, the method converts the axial position error into the geometric center distance between the imaging point of the sub-pixel through the actual lens at the central depth plane and the reference voxel, amplifies the axial position error for easy measurement, and searches one by one with sub-pixels as the step size, so as to conveniently, accurately and quickly locate the position of the actual sub-pixels with the same name, thereby achieving extremely high axial position error measurement accuracy; in the process of correcting the micro-image array, the voxel offset caused by the axial position error is compensated by scaling the image element, and the central depth plane is selected to obtain a constant scaling factor, which simplifies the process of processing the image element. The method proposed by the present invention can accurately measure the axial position error of the lens array in the integrated imaging display system, correct the axial position error, and achieve high-quality stereoscopic image reconstruction.
Claims
1. A method for measuring and correcting the axial position error of a lens array in an integrated imaging display system, characterized in that: The method comprises two processes: measuring the axial position error of the lens array and correcting the micro-image array. In the process of measuring the axial position error of the lens array, the horizontal direction of the display screen is taken as x Axis, perpendicular to the display screen z Axis, take the first pixel in the upper left corner of the display screen as the coordinate origin and establish a Cartesian coordinate system; Select a lens in the actual lens array as the reference lens, with coordinates ( x lens-ref , z lens-ref ), the remaining lenses in the actual lens array are the lenses to be tested, and the central sub-pixel of the image element corresponding to the reference lens is used as the reference sub-pixel, and the coordinates are ( x subpx-ref , 0), the image point of the reference sub-pixel in the central depth plane is set as the reference voxel, and the coordinates are ( x vx-ref , z vx-ref );According to the reference sub-pixel position under the reference lens, the position coordinates of a group of ideal sub-pixels with the same name under the ideal lens array condition are obtained X ( x subpx-1 , x subpx-2 , ..., x subpx-i ,...),in x subpx-1 represents the ideal sub-pixel position coordinates of the same name under the first ideal lens, x subpx-i Indicates i The ideal position coordinates of the same-name sub-pixels under an ideal lens are obtained by the following formula: In the formula x lens-i Indicates i The lateral position coordinates of an ideal lens; searching in the image element where the ideal sub-pixel with the same name is located with the sub-pixel as the step length, respectively comparing the distance between the geometric center of the image point of the searched sub-pixel in the central depth plane and the geometric center of the reference voxel, when the geometric center of the image point of the searched sub-pixel in the central depth plane coincides with the geometric center of the reference voxel, the searched sub-pixel is recorded as the actual sub-pixel with the same name, and the image point of the actual sub-pixel with the same name in the central depth plane is recorded as the actual voxel; Therefore, an actual sub-pixel with the same name is found near each ideal sub-pixel with the same name, and finally a group of actual sub-pixels with the same name are obtained, and the horizontal position coordinates are X' ( x' subpx-1 , x' subpx-2 ,..., x' subpx-i ,...), substitute the actual and ideal sub-pixels with the same name into the following formula to calculate the axial position error of the actual lens array: In the formula l c Indicates the distance between the center depth plane and the display screen, g is the distance between the ideal lens array and the display screen; in the process of searching for actual sub-pixels with the same name, the search step is one sub-pixel, so the position offset of the actual sub-pixels with the same name obtained by searching does not exceed half a sub-pixel; In the micro-image array correction process, the ideal sub-pixel with the same name is offset from the actual sub-pixel with the same name. In order to make the ideal sub-pixel with the same name coincide with the actual sub-pixel with the same name, the image element where the ideal sub-pixel with the same name is located is scaled to achieve the coincidence of the actual voxel with the reference voxel. The scaling factor of the image element is k i The following formula must be satisfied: Among them, Δ x' i and Δ x i are the distances from the actual sub-pixel with the same name and the ideal sub-pixel with the same name to the optical axis, where the optical axis is the optical axis of the actual lens or the optical axis of the ideal lens, z vx-i To reconstruct the distance between the voxel and the display, z' lens-i It represents the axial position of the actual lens. By scaling the image element where the ideal sub-pixel with the same name is located, the light emitted by the corrected micro-image array can be correctly modulated after passing through the actual lens array, thereby obtaining a correct 3D image without voxel drift.
2. The method for measuring and correcting the axial position error of the lens array in the integrated imaging display system according to claim 1, characterized in that: In the process of measuring the axial position error of the lens array, the measurement accuracy of the axial position error of the actual lens array is inversely proportional to the distance between the lens to be measured and the reference lens. In order to improve the measurement accuracy of the axial position error of the actual lens array, two rounds of measurement are performed: in the first round, the lens at the left end of the actual lens array is selected as the reference lens, and the lens at the right end of the actual lens array is measured; in the second round, the lens at the right end of the actual lens array is selected as the reference lens, and the lens at the left end of the actual lens array is measured to achieve accurate measurement.
3. The method for measuring and correcting the axial position error of the lens array in the integrated imaging display system according to claim 1, characterized in that: In the micro-image array correction process, in order to simplify the correction process and improve the correction efficiency, the central depth plane is selected as the voxel reconstruction position, and the scaling factor of the image scaling is controlled. k i After constantization, the scaling factor after constantization is expressed as follows:
