Method and device for parallel single-pixel computer-generated holographic imaging
By segmenting the images in single-pixel holographic imaging technology into multiple sub-regions, and combining diffraction gratings with parallel common path interference, the checks and balances between imaging quality and speed are solved, efficient wavefront detection is achieved, crosstalk is eliminated, and data acquisition efficiency is improved.
Patent Information
- Application Number
- CN202510403081.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-01
- Publication Date
- 2025-07-11
AI Technical Summary
There is a crosstalk problem in the existing single-pixel holographic imaging technology, and the problem of crosstalk between imaging quality and imaging speed in direct parallel interference measurements, which cannot effectively restore complete wavefront information.
The parallel single-pixel computing holographic imaging method is used to segment the image into multiple sub-regions, each sub-region is sampled using the same modulation mode, and the spectral components are separated by designing a diffraction grating and parallel common path interference, and parallel detection is performed by combining a multiplexed diffraction grating.
While ensuring imaging quality, the imaging speed is significantly improved, the parallel detection of multi-region wavefront information is realized, the wavefront crosstalk is eliminated, and data acquisition efficiency is improved.
Smart Images

Figure CN120295078A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of single-pixel computational holographic imaging. Background Art
[0002] Single-pixel imaging is an emerging computational imaging technology that uses a series of modulation patterns with spatial resolution to modulate the wavefront or scene to be measured. The intensity information is recorded by a single-pixel detector without spatial resolution, and then the image is reconstructed based on the correlation between the modulation patterns and the intensity values. When single-pixel imaging is combined with interferometry, single-pixel detection uses complex coefficients corresponding to a series of modulation patterns and can further reconstruct spatial wavefront information. Benefiting from the advantages of high sensitivity and wide spectrum of single-pixel detectors, single-pixel detection for holographic imaging is very important in the fields of three-dimensional imaging, biomedical microscopy, and quantum imaging.
[0003] Currently, there are mainly two methods for single-pixel computational holographic imaging based on interferometry: the double-path method and the common-path method. The former requires an additional optical path to be used as the reference light, and this configuration has disadvantages such as complex optical paths and being vulnerable to environmental interference. The latter uses different reference strategies to simultaneously introduce the signal light and the reference light into the same optical path for interference, realizing the common-path interference single-pixel computational holographic imaging technology with a simple device and strong anti-interference ability. However, during the imaging process, a large number of modulation patterns need to be projected. As the imaging resolution increases, the number of required modulation patterns increases significantly, which limits the imaging speed. Although downsampling measurement can reduce the number of modulation patterns, it loses the imaging quality to a certain extent.
[0004] To balance the imaging quality and the imaging speed, methods for accelerating acquisition also include parallel single-pixel computational holographic imaging, in which the image is divided into multiple sub-regions, and each sub-region is sampled using the same modulation pattern. However, when directly performing parallel common-path interferometric measurement, after Fourier transform, due to the spectral information of multiple sub-regions being aliased, crosstalk occurs in the reconstructed amplitude and phase, and the complete wavefront cannot be restored. Summary of the Invention
[0005] The technical problems to be solved by the present invention are: how to solve the trade-off problem between imaging quality and imaging speed in single-pixel holographic imaging, how to solve the crosstalk problem in direct parallel interferometric detection, and how to design the spectral distribution to achieve faster wavefront detection while ensuring the imaging quality.
[0006] The technical solution adopted by the present invention is: a method for parallel single-pixel computational holographic imaging, which is carried out according to the following steps Step 1: Relay the target spatial wavefront to the spatial light modulator (3) through a 4-f system. A modulation basis mode is loaded on the spatial light modulator (3). According to the idea of parallel common-path interference imaging, the target spatial wavefront (image) is divided into multiple sub-target regions on the spatial light modulator (3). A diffraction grating needs to be reasonably designed for each sub-target region to achieve parallel common-path interference measurement. According to the design of parallel common-path interference, determine the detector arrangement at the detection end. If a single-pixel detector is used, the modulated light carrying the target information reflected by the spatial light modulator (3) is split by a beam splitter. Each beam of light split by the beam splitter carries the wavefront information of all regions. Different sub-regions correspond to different spectral components (beams). Each beam of light is independently detected for the signal intensity value by a single-pixel detector after passing through a lens and a pinhole on a straight line. If an array detector is used, the modulated light field reflected by the spatial light modulator (3) carrying multiple sub-target information is detected by the array detector (20) for the signal intensity values corresponding to the sub-regions of each sub-target. Step 2: Select the phase modulation mode loaded on the spatial light modulator (3). The phase modulation mode is one of the common-path multi-step phase-shift interference method and the common-path off-axis interference method. When it is the common-path multi-step phase-shift interference method, several phase-shifting masks and reference partitioning strategies are loaded on the spatial light modulator (3). Among them, the phase-shifting masks are used to perform multi-step phase-shift operations on the reference beam, and the reference partitioning strategy is used to partition the incident wavefront into a reference light part and a signal light part. The unmodulated wavefront and the modulated wavefront reflected from the spatial light modulator (3) undergo common-path interference. When it is the common-path off-axis interference method, an inclined phase grating and a reference partitioning strategy are loaded on the spatial light modulator (3). Among them, the inclined phase grating is used to introduce a small angle between the signal beam and the reference beam to generate a fixed phase difference, and the reference partitioning strategy is used to partition the incident wavefront into a reference light part and a signal light part. The unmodulated wavefront and the modulated wavefront reflected from the spatial light modulator (3) undergo common-path interference. Step 3: Determine the diffraction grating loaded on the spatial light modulator (3). By dividing the wavefront to be measured into multiple sub-regions, different characteristic diffraction gratings are loaded in the corresponding sub-regions of the spatial light modulator (3). The period and direction of the diffraction grating for each sub-region are designed to form a multiplexed diffraction grating. Among them, the period of the diffraction grating affects the diffraction angle, thereby affecting the dispersion degree of the diffraction order, resulting in different distances of movement of the frequency components of the sub-regions in the Fourier spectrum. The direction of the diffraction grating refers to the arrangement direction of the grating lines, which determines the diffraction direction of the diffraction order, causing the spectral components of the sub-regions in the Fourier spectrum to move along the direction perpendicular to the grating lines. By designing the period and direction of the diffraction grating, the optimal multiplexed diffraction grating is introduced into the parallel common-path interference to separate the spectral components of each sub-region concentrated at the center of the Fourier plane, avoiding overlap or crosstalk between spectral points. Step 4: Configure the initial parameters of the detection system. For the wavefront imaging resolution M×N to be measured, the number of sub-regions K of the wavefront to be measured, and the single-pixel imaging sampling rate φ. When the phase modulation mode is the common-path multi-step phase-shift interference method, the number of phase-shift steps needs to be configured. When the phase modulation mode is the common-path off-axis interference method, the period number of the tilted phase grating needs to be configured, that is, the deviation angle between the signal light and the reference light. Step 5: Load the combination of the phase modulation mode and the multiplexed diffraction grating on the spatial light modulator (3). When it is the common-path multi-step phase-shift interference method, the φ×M×N / K generated modulation modes are simultaneously loaded into each sub-region of the spatial light modulator (3), and the multiplexed diffraction grating is loaded onto the spatial light modulator (3). When it is the common-path off-axis interference method, the M×N / K generated modulation modes are simultaneously loaded into each sub-region of the spatial light modulator (3), and the tilted phase grating is loaded onto the spatial light modulator (3) to generate a fixed phase difference for interference, and the multiplexed diffraction grating is loaded onto the spatial light modulator (3). If the detection end is a single-pixel detector, the modulated light with target information reflected by the spatial light modulator (3) passes through a beam splitter (the beam splitter is one or more. One beam splitter can split into two 1:1 light beams. Here, according to the number of sub-region divisions, a certain number of beam splitters and detectors with the same number as the region number are used for parallel detection. A single-pixel detector independently detects the information of one region) to split out K light beams (K is the number of sub-regions). Each light beam split by the beam splitter passes through a lens and a pinhole in a straight line and is detected by a single-pixel detector for the signal intensity value. If the detection end is an array detector, the modulated light field with multiple sub-target information reflected by the spatial light modulator (3) has the signal intensity values of the sub-regions corresponding to each sub-target detected in parallel by the array detector. Step 6: Reconstruct the target spatial wavefront. When the phase modulation mode is the common-path multi-step phase-shift interference method, the complex-valued coefficients of the sub-region target spatial wavefront are obtained according to the phase-shift technology (in the form of a + ib. When measuring an object, if you want to obtain the amplitude-phase information of the object, interference is required. When using the common-path multi-step phase-shift interference, for example, the four-step phase-shift interference method requires 4 different phase differences. For each modulation pattern, it will be projected four times corresponding to four different phase differences. The detector will record the intensity values of the four patterns, and then according to the four-step phase-shift calculation formula, use the four intensity values to obtain the complex-valued coefficients corresponding to the modulation pattern. Using the corresponding relationship between the complex-valued coefficients and the modulation pattern, the target wavefront can be restored), and the single-pixel reconstruction algorithm is used to restore the sub-region target spatial wavefront, and multiple sub-regions are stitched together to restore the complete wavefront. When the phase modulation mode is the common-path off-axis interference method, the interference hologram of the sub-region target spatial wavefront is restored according to the single-pixel reconstruction algorithm, and then the sub-region target spatial wavefront is restored from the interference hologram according to the Fourier fringe analysis method or the Hilbert transform, and multiple sub-regions are stitched together to restore the complete wavefront.
[0007] The modulation mode in the first step is one of the Hadamard transform basis mode, the discrete cosine transform basis mode, and the Fourier transform basis mode.
[0008] The reference partitioning strategy in the second step is one of the checkerboard reference pattern, the internal and external reference pattern, and the pattern self-encoding reference pattern.
[0009] The common-path multi-step phase-shift interference method in the second step is one of two-step phase shift, three-step phase shift, and four-step phase shift.
[0010] The diffraction grating in the third step is an amplitude grating or a phase grating. The amplitude grating realizes diffraction by periodically modulating the amplitude of light, and the phase grating realizes diffraction by periodically modulating the phase of light. By designing the period and direction of the diffraction grating and introducing the multiplexed diffraction grating into the parallel common-path interference, the frequency components of the sub-region can be shifted by a corresponding distance in a specific direction, and the aliased wavefront information can be separated.
[0011] When a single-pixel detector is used, a beam splitter is used to split out K beams, and the signal intensity value of each beam is independently detected by the single-pixel detector after passing through a lens and a pinhole in a straight line; when an array detector is used, the signal intensity values of the corresponding spectral components in each sub-region are collected in parallel by the array detector (20).
[0012] An apparatus for parallel single-pixel computational holographic imaging, comprising a first lens (1), a second lens (2), and a spatial light modulator (3), wherein the first lens (1) and the second lens (2) form a 4-f system, and the central axis of the 4-f system is perpendicular to the mirror surface of the spatial light modulator (3); when a single-pixel detector is used, it further comprises a plurality of beam splitters, a plurality of beam-splitting lenses, pinholes having the same number as the beam-splitting lenses, and single-pixel detectors having the same number as the beam-splitting lenses. Each beam-splitting lens corresponds to a pinhole and a single-pixel detector, and they are in a straight line. Each single-pixel detector receives the detection signal intensity value of the corresponding beam; when an array detector is used, it further comprises a seventh lens (19) and an array detector (20), and the reflected light of the spatial light modulator (3) is received by the array detector (20) after passing through the seventh lens (19).
[0013] The aperture of the pinhole satisfies the formula r ≤ 1.22λ * f / d, where r is the aperture of the pinhole, λ is the wavelength of the beam, f is the focal length of the beam-splitting lens, and d is the number of micromirrors used on the spatial light modulator (3) * the size of each micromirror, that is, the diameter of the modulated light field with target information reflected by the spatial light modulator (3).
[0014] To solve the trade-off problem between imaging quality and imaging speed in single-pixel holographic imaging, the image is segmented into multiple sub-regions using the parallel concept, and each sub-region is sampled simultaneously using the same modulation pattern. However, when directly performing parallel common-path interferometric measurement, the spectral information of multiple sub-regions will be aliased, and the complete wavefront information cannot be restored. The diffraction grating modulates light waves using its periodic structure, thereby generating a diffraction effect. The period of the diffraction grating affects the diffraction angle, thereby affecting the dispersion degree of the diffraction order. The direction of the diffraction grating refers to the arrangement direction of the grating lines, which determines the diffraction direction of the diffraction order. This patent proposes combining the diffraction grating with parallel common-path interference. By carefully designing the period and direction of the diffraction grating, the optimal multiplexed diffraction grating is introduced into parallel common-path interference to separate the spectral components of each sub-region concentrated at the center of the Fourier plane, avoiding overlap or crosstalk between spectral points. When the detection end uses a single-pixel detector, after splitting a beam of light into multiple beams, each beam of light simultaneously carries the information of all sub-regions. According to the position of the spectral components of each sub-region in the spectral distribution, the arrangement of the single-pixel detectors is designed. A single-pixel detector independently collects the wavefront information of one region, which can improve the speed several times while ensuring the imaging quality.
[0015] The beneficial effects of the present invention are as follows: The present invention fully utilizes the diffraction principle of the diffraction grating to solve the crosstalk problem in direct parallel interferometric measurement. By combining the multiplexed diffraction grating with parallel common-path interference, parallel single-pixel computational holographic imaging is realized. On the device for parallel single-pixel computational holographic imaging, diffraction gratings with different periods and directions are carefully designed on each sub-region to form the optimal multiplexed diffraction grating, separating the aliased wavefront information of the sub-regions and eliminating wavefront information crosstalk. According to the spectral distribution, the corresponding arrangement of the detectors at the detection end is designed to achieve parallel detection of the wavefront information of multiple regions. Compared with the existing single-pixel holographic imaging technology, wavefront detection with a faster speed can be realized while ensuring the imaging quality; by combining the multiplexed diffraction grating to simultaneously detect multiple sub-regions of the target wavefront, the data acquisition efficiency is improved by parallel processing while eliminating wavefront crosstalk; the imaging device and method are very flexible, and the phase-shifting interference method with better imaging quality can be selected according to the application scenario, or the off-axis interference method with further improved speed can be selected on the premise of ensuring a certain imaging quality; spatial wavefront detection can be realized for most of the existing single-pixel imaging with different modulation basis modes. Brief Description of the Drawings
[0016] Figure 1 Schematic diagram of the device for parallel single-pixel computational holographic imaging with a single-pixel detector at the detection end; Figure 2 Schematic diagram of the device for parallel single-pixel computational holographic imaging with an array detector at the detection end; Figure 3Amplitude and phase distribution diagrams of the target spatial wavefront simulated for Examples 1 and 2. Among them, (a) is the amplitude distribution diagram, which is a matrix of all ones, and (b) is the phase distribution diagram, which is a grayscale image of vegetables; Figure 4 Amplitude and phase distribution diagrams of the target spatial wavefront achieved by using the four-step phase-shifting method combined with direct four-region parallel detection. (a) is the Fourier spectrum diagram after Fourier transform, where the frequency components of the four sub-regions are concentrated at the center of the Fourier plane; (b) and (c) are the amplitude and phase distribution diagrams of the target spatial wavefront reconstructed without combining the multiplexed diffraction grating; Figure 5 Amplitude and phase distribution diagrams of the target spatial wavefront achieved by using the four-step phase-shifting method combined with a multiplexed diffraction grating with a poorly designed direction for four-region parallel detection. (a) is the distribution of the frequency components of the four sub-regions in the Fourier spectrum diagram after Fourier transform; (b) and (c) are the amplitude and phase distribution diagrams of the target spatial wavefront where the direction of the diffraction grating of the second sub-region is not designed well, resulting in the overlap of the frequency components of the first and second sub-regions and crosstalk in the reconstructed amplitude and phase of the first and second sub-regions; Figure 6 Amplitude and phase distribution diagrams of the target spatial wavefront achieved by using the four-step phase-shifting method combined with a multiplexed diffraction grating with a poorly designed period for four-region parallel detection. (a) is the distribution of the frequency components of the four sub-regions in the Fourier spectrum diagram after Fourier transform; (b) and (c) are the amplitude and phase distribution diagrams of the target spatial wavefront where the period of the diffraction grating of the second sub-region is not designed well, resulting in poor reconstruction effect of the second sub-region; Figure 7 Amplitude and phase distribution diagrams of the high-resolution target spatial wavefront achieved by using the four-step phase-shifting method combined with the optimal multiplexed diffraction grating for four-region parallel detection; among them, (a) is the distribution of the frequency components of the four sub-regions in the Fourier spectrum diagram after Fourier transform; (b) is the reconstructed amplitude distribution diagram, and (c) is the reconstructed phase distribution diagram; Figure 8 Off-axis hologram of the target spatial wavefront reconstructed by using the off-axis method combined with the optimal multiplexed diffraction grating for four-region parallel detection and amplitude and phase distribution diagrams reconstructed by using the Fourier fringe analysis method; among them, (a) is the distribution of the frequency components of the four sub-regions in the Fourier spectrum diagram after Fourier transform; (b) is the interference hologram for restoring the target spatial wavefront, and (c) and (d) are the amplitude and phase distribution diagrams reconstructed from the hologram by using the Fourier fringe analysis method or the Hilbert transform based on the interference hologram; Among them, 1. the first lens, 2. the second lens, 3. the spatial light modulator, 4. the first beam splitter, 5. the second beam splitter, 6. the third beam splitter, 7. the third lens, 8. the fourth lens, 9. the fifth lens, 10. the sixth lens, 11. the first pinhole, 12. the second pinhole, 13. the third pinhole, 14. the fourth pinhole, 15. the first single-pixel detector, 16. the second single-pixel detector, 17. the third single-pixel detector, 18. the fourth single-pixel detector, 19. the seventh lens, 20. the array detector. Detailed implementation manners
[0017] Embodiment 1: A method for parallel single-pixel computational holographic imaging is carried out according to the following steps Step 1: Relay the target spatial wavefront to the spatial light modulator 3 through a 4-f system. A Hadamard transform basis mode is loaded on the spatial light modulator 3. According to the idea of parallel common-path interference imaging, the target spatial wavefront is divided into 4 sub-target regions on the spatial light modulator 3. Diffraction gratings need to be reasonably designed for each sub-target region to achieve parallel common-path interference measurement. Four single-pixel detectors are used. The modulated light with target information reflected by the spatial light modulator 3 is split by a beam splitter. Each beam of light split by the beam splitter carries the wavefront information of all regions. Different sub-regions correspond to different spectral components. Each beam of light is independently detected for the signal intensity value by a single-pixel detector after passing through a lens and a pinhole on a straight line. Step 2: Select the phase modulation mode loaded on the spatial light modulator 3. The phase modulation mode is a common-path four-step phase-shifting interference method. A number of masks for phase shifting and a checkerboard reference partitioning strategy are loaded on the spatial light modulator 3. Among them, the masks for phase shifting are used to perform four-step phase-shifting operations on the reference beam, and the checkerboard reference partitioning is used to partition the incident wavefront into a reference light part and a signal light part. The unmodulated wavefront and the modulated wavefront reflected from the spatial light modulator 3 undergo common-path interference. Step 3: Determine the amplitude-type diffraction grating loaded on the spatial light modulator 3. The amplitude-type grating realizes diffraction by periodically modulating the amplitude of light. By dividing the wavefront to be measured into 4 sub-regions, diffraction gratings with different characteristics are loaded on the corresponding sub-regions of the spatial light modulator 3. The periods and directions of the diffraction gratings in each sub-region are designed to form a multiplexed diffraction grating. Among them, the period of the diffraction grating affects the diffraction angle, thereby affecting the dispersion degree of diffraction orders, causing the frequency components of the sub-regions to move different distances in the Fourier spectrum. The direction of the diffraction grating refers to the arrangement direction of the grating lines, which determines the diffraction direction of the diffraction orders, making the spectral components of the sub-regions in the Fourier spectrum move along the vertical direction of the grating lines. By designing the periods and directions of the diffraction gratings, the optimal multiplexed diffraction grating is introduced into the parallel common-path interference to separate the spectral components of each sub-region concentrated at the center of the Fourier plane, avoiding overlap or crosstalk between spectral points; Step 4: Configure the initial parameters of the detection system. When the imaging resolution of the wavefront to be measured is 128×128, the number of sub-regions of the wavefront to be measured is 4, the single-pixel imaging sampling rate is 1, and the phase modulation mode is the common-path four-step phase-shifting interference method, configure the number of phase-shifting steps to 4; Step 5: Load a combination of the phase modulation mode and the multiplexed diffraction grating on the spatial light modulator 3; when using the common-path four-step phase-shifting interference method, simultaneously load the generated 1×128×128 / 4 modulation modes to each sub-region of the spatial light modulator 3, and load the multiplexed diffraction grating to the spatial light modulator 3; when the detection end is a single-pixel detector, the modulated light with target information reflected by the spatial light modulator 3 is split into 4 beams by 3 beam splitters. According to the spectral distribution, arrange 4 single-pixel detectors, and determine the aperture of the pinhole (using the formula r≤1.22λ*f / d, where r is the aperture of the pinhole, λ is the wavelength of the light beam, f is the focal length of the beam-splitting lens, and d is the number of micromirrors used on the spatial light modulator 3 * the size of each micromirror, that is, the diameter of the modulated light field with target information reflected by the spatial light modulator 3). Each beam of light split by the beam splitter passes through a beam-splitting lens and a pinhole in a straight line and is independently detected for the signal intensity value by a single-pixel detector; Step 6: Reconstruct the target spatial wavefront; when using the common-path four-step phase-shifting interference method, obtain the complex-valued coefficients of the sub-region target spatial wavefront based on the phase-shifting technology, and use the single-pixel reconstruction algorithm to restore the sub-region target spatial wavefront, and splice multiple sub-regions to restore the complete wavefront; As Figure 3 shown, it is the amplitude and phase distribution diagrams of the target spatial wavefront simulated in this embodiment. ((a) is the amplitude distribution diagram, which is a matrix of all ones, and (b) is the phase distribution diagram, which is a grayscale image of a vegetable). As Figure 4As shown, the amplitude and phase distribution diagrams of the target spatial wavefront achieved by using the four-step phase-shifting method in combination with direct four-region parallel detection. (a) is the Fourier spectrum diagram after Fourier transform, where the frequency components of the four sub-regions are concentrated at the center of the Fourier plane; (b) and (c) are the amplitude and phase distribution diagrams of the target spatial wavefront reconstructed without combining the multiplexed diffraction grating. Figure 5 And Figure 6 The amplitude and phase distribution diagrams of the target spatial wavefront achieved by using the four-step phase-shifting method in combination with a non-carefully designed multiplexed diffraction grating for four-region parallel detection. When the period and direction of the diffraction grating are not carefully designed, complete high-quality wavefront information cannot be restored. In one embodiment, Figure 5 Is the reconstruction result diagram without carefully designing the direction of the sub-region diffraction grating. Without carefully designing the direction of the diffraction grating in the second sub-region, the spectral points of the first and second sub-regions in (a) overlap, resulting in crosstalk in the reconstructed amplitudes (b) and phases (c) of the first and second sub-regions; Figure 6 Is the reconstruction result diagram without carefully designing the period of the sub-region diffraction grating. Without carefully designing the period of the diffraction grating in the second sub-region, the reconstructed amplitudes (b) and phases (c) of the first and second sub-regions have poor effects; Figure 7 Are the amplitude and phase distribution diagrams of the high-resolution target spatial wavefront achieved by using the four-step phase-shifting method in combination with the optimal multiplexed diffraction grating for four-region parallel detection. ( Figures 3 - 7 The obtained imaging results are not limited to the single-pixel detector detection system and the common-path four-step phase-shifting interference method) Embodiment 2: A method for parallel single-pixel computational holographic imaging is carried out according to the following steps Step 1: Relay the target spatial wavefront to the spatial light modulator 3 through a 4-f system. The Fourier transform basis mode is loaded on the spatial light modulator 3. According to the idea of parallel common-path interference imaging, the target spatial wavefront is divided into 4 sub-target regions on the spatial light modulator 3. Each sub-target region needs to be reasonably designed with a diffraction grating to achieve parallel common-path interference measurement; when using an array detector, the spatial light modulator 3 reflects the modulated light field with multiple sub-target information, and the signal intensity values of the sub-regions corresponding to each sub-target are detected by the array detector 20; Step 2: Select the phase modulation mode loaded on the spatial light modulator 3. When the phase modulation mode is the common-path off-axis interference method, an inclined phase grating and a mode self-encoding reference partition strategy are loaded on the spatial light modulator 3. The inclined phase grating is used to introduce a small angle between the signal beam and the reference beam to generate a fixed phase difference, and the mode self-encoding reference partition strategy is used to partition the incident wavefront into a reference light part and a signal light part, and the unmodulated wavefront and the modulated wavefront reflected from the spatial light modulator 3 undergo common-path interference; Step 3: Determine the phase-type diffraction grating loaded on the spatial light modulator 3. The phase-type grating realizes diffraction by periodically modulating the phase of light. By dividing the wavefront to be measured into 4 sub-regions, diffraction gratings with different characteristics are loaded on the corresponding sub-regions of the spatial light modulator 3. The period and direction of the diffraction grating for each sub-region are designed to form a multiplexed diffraction grating. Among them, the period of the diffraction grating affects the diffraction angle, thereby affecting the dispersion degree of diffraction orders, resulting in different displacements of the frequency components of the sub-regions in the Fourier spectrum. The direction of the diffraction grating refers to the arrangement direction of the grating lines, which determines the diffraction direction of the diffraction orders, causing the spectral components of the sub-regions in the Fourier spectrum to move along the vertical direction of the grating lines. By designing the period and direction of the diffraction grating, the optimal multiplexed diffraction grating is introduced into the parallel common-path interference to separate the spectral components of each sub-region concentrated at the center of the Fourier plane accordingly, avoiding overlap or crosstalk between spectral points; Step 4: Configure the initial parameters of the detection system. The imaging resolution of the wavefront to be measured is 128×128, the number of sub-regions of the wavefront to be measured is 4, the single-pixel imaging sampling rate is 1. When the phase modulation mode is the common-path off-axis interference method, configure the number of periods of the tilted phase grating, that is, the deviation angle between the signal light and the reference light; Step 5: Load the combination of the phase modulation mode and the multiplexed diffraction grating on the spatial light modulator 3. When using the common-path off-axis interference method, sequentially load the 128×128 / 4 modulation modes generated by single-pixel imaging onto each sub-region of the spatial light modulator 3, and load the tilted phase grating onto the spatial light modulator 3 to generate a fixed phase difference for interference. The diffraction grating for separating diffraction orders is loaded on different sub-regions of the spatial light modulator 3, and the array detector 20 parallelly collects the signal intensity values of the spectral points corresponding to each mode in each sub-region; Step 6: Reconstruct the target spatial wavefront. When the phase modulation mode is the common-path off-axis interference method, restore the interference hologram of the target spatial wavefront of the sub-region according to the single-pixel reconstruction algorithm, and then restore the target spatial wavefront of the sub-region from the interference hologram according to the Fourier fringe analysis method or the Hilbert transform, and splice multiple sub-regions to restore the complete wavefront.
[0018] Embodiment 3: An apparatus for parallel single-pixel computational holographic imaging (an apparatus corresponding to Embodiment 1), such as Figure 1As shown in the figure, the detection end is a single-pixel detector, including a first lens 1, a second lens 2, a spatial light modulator 3, a first beam splitter 4, a second beam splitter 5, a third beam splitter 6, a third lens 7, a fourth lens 8, a fifth lens 9, a sixth lens 10, a first pinhole 11, a second pinhole 12, a third pinhole 13, a fourth pinhole 14, a first single-pixel detector 15, a second single-pixel detector 16, a third single-pixel detector 17, and a fourth single-pixel detector 18. Among them, the third lens 7, the fourth lens 8, the fifth lens 9, and the sixth lens 10 are beam-splitting lenses. The first lens 1 and the second lens 2 form a 4-f system. The central axis of the 4-f system is perpendicular to the mirror surface of the spatial light modulator 3. Four beams of light are split by three beam splitters. Each beam-splitting lens, the pinhole corresponding to the beam-splitting lens, and the single-pixel detector corresponding to the beam-splitting lens are on a straight line. And the four outgoing light rays after passing through the beam splitters are received by the single-pixel detectors after passing through the corresponding beam-splitting lenses and pinholes.
[0019] In this embodiment, the phase modulation mode is a common-path four-step phase-shifting interference method. The wavefront to be measured is relayed to the spatial light modulator 3 through the 4-f system composed of the first lens 1 and the second lens 2. After the spatial light modulator 3 loads the modulation mode and the multiplexed diffraction grating and modulates the target irradiated thereon, the modulated light with target information is reflected onto the beam splitter. The three beam splitters equally divide the laser into four beams of light. After the modulated light with sub-region target information is focused by the corresponding lens, it passes through the pinhole and is received by the corresponding single-pixel detector to detect the intensity value.
[0020] The spatial light modulator 3 is used to carry the designed phase distribution and play an expected modulation role on the transmitted wavefront. The phase modulation mode is a common-path four-step phase-shifting interference method. The spatial light modulator 3 is used to load the four-step phase-shifting operation and the checkerboard reference partitioning strategy to partition the input wavefront and generate phase-shifting interference. The aperture of the pinhole satisfies the formula r ≤ 1.22λ * f / d, where r is the aperture of the pinhole, λ is the wavelength of the light beam, f is the focal length of the beam-splitting lens, and d is the number of micromirrors used on the spatial light modulator 3 * the size of each micromirror, that is, the diameter of the modulated light field with target information reflected by the spatial light modulator 3.
[0021] Embodiment 4: An apparatus for parallel single-pixel computational holographic imaging (the apparatus corresponding to Embodiment 2), as Figure 2 shown, the detection end is an array detector, including a first lens 1, a second lens 2, a spatial light modulator 3, a seventh lens 19, and an array detector. The first lens 1 and the second lens 2 form a 4-f system. The central axis of the 4-f system is perpendicular to the mirror surface of the spatial light modulator 3. The reflected light of the spatial light modulator 3 passes through the seventh lens 19 and is received by the array detector 20.
[0022] The phase modulation mode of this device is a common-path off-axis interference method. The spatial light modulator 3 is used to load a tilted phase grating and a mode self-encoding reference partition strategy to partition the input wavefront and generate off-axis interference.
[0023] The wavefront to be measured is relayed to the spatial light modulator 3 through a 4-f system composed of the first lens 1 and the second lens 2. After the spatial light modulator 3 loads the phase modulation mode and the multiplexed diffraction grating and modulates the target irradiated thereon, the modulated light with target information is focused by the seventh lens 19 and then received by the array detector 20 through a pinhole to obtain the detection intensity value of the sub-region.
Claims
1. A method for parallel single-pixel computational holographic imaging, characterized in that: Proceed as follows Step 1: Relay the target spatial wavefront to the spatial light modulator (3) through a 4-f system. Load the modulation basis mode on the spatial light modulator (3). According to the idea of parallel common-path interference imaging, divide the target spatial wavefront into multiple sub-target regions on the spatial light modulator (3). Each sub-target region needs to be reasonably designed with a diffraction grating to achieve parallel common-path interference measurement. According to the design of parallel common-path interference, determine the detector arrangement at the detection end. If a single-pixel detector is used, the modulated light carrying the target information reflected by the spatial light modulator (3) is split by a beam splitter. Each beam of light split by the beam splitter carries the wavefront information of all regions. Different sub-regions correspond to different spectral components. Each beam of light passes through a lens and a pinhole on a straight line and is independently detected for the signal intensity value by a single-pixel detector. If an array detector is used, the modulated light field reflected by the spatial light modulator (3) carrying multiple sub-target information is detected for the signal intensity values of the sub-regions corresponding to each sub-target by the array detector (20). Step 2: Select the phase modulation mode loaded on the spatial light modulator (3). The phase modulation mode is one of the common-path multi-step phase-shift interference method and the common-path off-axis interference method. When it is the common-path multi-step phase-shift interference method, load several masks for phase shifting and a reference partitioning strategy on the spatial light modulator (3). Among them, the mask for phase shifting is used to perform multi-step phase-shifting operations on the reference beam, and the reference partitioning strategy is used to partition the incident wavefront into a reference light part and a signal light part. The unmodulated wavefront and the modulated wavefront reflected from the spatial light modulator (3) undergo common-path interference. When it is the common-path off-axis interference method, load an inclined phase grating and a reference partitioning strategy on the spatial light modulator (3). Among them, the inclined phase grating is used to introduce a small angle between the signal beam and the reference beam to generate a fixed phase difference, and the reference partitioning strategy is used to partition the incident wavefront into a reference light part and a signal light part. The unmodulated wavefront and the modulated wavefront reflected from the spatial light modulator (3) undergo common-path interference. Step 3: Determine the diffraction grating loaded on the spatial light modulator (3). By dividing the wavefront to be measured into multiple sub-regions, load diffraction gratings with different characteristics on the corresponding sub-regions of the spatial light modulator (3). Design the period and direction of the diffraction grating for each sub-region to form a multiplexed diffraction grating. Among them, the period of the diffraction grating affects the diffraction angle, thereby affecting the dispersion degree of the diffraction order, causing the frequency components of the sub-region to move different distances in the Fourier spectrum. The direction of the diffraction grating refers to the arrangement direction of the grating lines, which determines the diffraction direction of the diffraction order, causing the spectral components of the sub-region in the Fourier spectrum to move along the direction perpendicular to the grating lines. Step 4: Configure the initial parameters of the detection system. For the wavefront imaging resolution M×N to be measured, the number of sub-regions K of the wavefront to be measured, and the single-pixel imaging sampling rate φ. When the phase modulation mode is the common-path multi-step phase-shift interference method, the number of phase-shift steps needs to be configured. When the phase modulation mode is the common-path off-axis interference method, the period number of the tilted phase grating needs to be configured, that is, the deviation angle between the signal light and the reference light. Step 5: Load a combination of the phase modulation mode and the multiplexed diffraction grating on the spatial light modulator (3). When it is the common-path multi-step phase-shift interference method, the φ×M×N / K generated modulation modes are simultaneously loaded into each sub-region of the spatial light modulator (3), and the multiplexed diffraction grating is loaded onto the spatial light modulator (3). When it is the common-path off-axis interference method, the M×N / K generated modulation modes are simultaneously loaded into each sub-region of the spatial light modulator (3), and the tilted phase grating is loaded onto the spatial light modulator (3) to generate a fixed phase difference for interference, and the multiplexed diffraction grating is loaded onto the spatial light modulator (3). If the detection end is a single-pixel detector, the modulated light with target information reflected by the spatial light modulator (3) is split into K beams by a beam splitter. Each beam split by the beam splitter passes through a lens and a pinhole in a straight line and is detected by a single-pixel detector for the signal intensity value. If the detection end is an array detector, the modulated light field reflected by the spatial light modulator (3) with multiple sub-target information is detected by the array detector in parallel for the signal intensity values of the sub-regions corresponding to each sub-target. Step 6: Reconstruct the target spatial wavefront. When the phase modulation mode is the common-path multi-step phase-shift interference method, the complex-valued coefficients of the sub-region target spatial wavefront are obtained based on the phase-shift technology, and the single-pixel reconstruction algorithm is used to restore the sub-region target spatial wavefront, and multiple sub-regions are stitched together to restore the complete wavefront. When the phase modulation mode is the common-path off-axis interference method, the interference hologram of the sub-region target spatial wavefront is restored based on the single-pixel reconstruction algorithm, and then the sub-region target spatial wavefront is restored from the interference hologram according to the Fourier fringe analysis method or the Hilbert transform, and multiple sub-regions are stitched together to restore the complete wavefront.
2. The method for parallel single-pixel computational holographic imaging according to claim 1, wherein: The modulation mode in the above Step 1 is one of the Hadamard transform basis mode, the discrete cosine transform basis mode, and the Fourier transform basis mode.
3. A method for parallel single-pixel computational holographic imaging according to claim 1, characterized in that: The reference partition strategy in the above Step 2 is one of the checkerboard reference pattern, the inner and outer reference pattern, and the pattern self-encoding reference pattern.
4. A method for parallel single-pixel computational holographic imaging according to claim 1, characterized in that: The common-path multi-step phase-shift interference method in the above Step 2 is one of the two-step phase shift, the three-step phase shift, and the four-step phase shift.
5. A method for parallel single-pixel computational holographic imaging according to claim 1, characterized in that: The diffraction grating in the above Step 3 is an amplitude grating or a phase grating. The amplitude grating realizes diffraction by periodically modulating the amplitude of light, and the phase grating realizes diffraction by periodically modulating the phase of light. By designing the period and direction of the diffraction grating, the multiplexed diffraction grating is introduced into the parallel common-path interference, and the frequency components of the sub-regions can be moved a corresponding distance in a specific direction to separate the aliased wavefront information.
6. A method for parallel single-pixel computational holographic imaging according to claim 1, characterized in that: When using a single-pixel detector, a beam splitter is used to split out K light beams. After each light beam passes through a lens and a pinhole that are in a straight line, the single-pixel detector independently detects the signal intensity value; when using an array detector, the array detector (20) simultaneously acquires the signal intensity values of the corresponding spectral components in each sub-region.
7. An apparatus used in the method according to claim 6, characterized in that: It includes a first lens (1), a second lens (2), and a spatial light modulator (3). The first lens (1) and the second lens (2) form a 4-f system, and the central axis of the 4-f system is perpendicular to the mirror surface of the spatial light modulator (3); when using a single-pixel detector, it further includes a plurality of beam splitters, a plurality of beam-splitting lenses, the same number of pinholes as the beam-splitting lenses, and the same number of single-pixel detectors as the beam-splitting lenses. Each beam-splitting lens corresponds to a pinhole and a single-pixel detector, and they are in a straight line. Each single-pixel detector receives the detected signal intensity value of the corresponding light beam; when using an array detector, it further includes a seventh lens (19) and an array detector (20). The reflected light of the spatial light modulator (3) is received by the array detector (20) after passing through the seventh lens (19).
8. The device according to claim 7, characterized in that: The aperture of the pinhole satisfies the formula r ≤ 1.22λ * f / d, where r is the aperture of the pinhole, λ is the wavelength of the light beam, f is the focal length of the beam-splitting lens, and d is the number of micromirrors used on the spatial light modulator (3) * the size of each micromirror, that is, the diameter of the modulated light field with target information reflected by the spatial light modulator (3).