A high-speed high-resolution compressed sensing imaging system and an imaging method thereof

Abstract

The application discloses a high-speed and high-resolution compressed sensing imaging system and an imaging method. Light of an object is transmitted and reflected by a polarization beam splitter after being emitted to the polarization beam splitter after a first imaging module. The transmitted light is reflected back to the polarization beam splitter after being encoded and modulated by a spatial light modulator, and then is incident on a target surface of a camera to form a coded imaging image through a second imaging module. The reflected light is transmitted to a hollow ridge reflection prism through a quarter-wave plate, is reflected back to the quarter-wave plate by the hollow ridge reflection prism, and then is incident on the target surface of the camera to form a normal imaging image through the quarter-wave plate, the polarization beam splitter and the second imaging module in sequence. Then, the coded imaging data and the normal imaging data are matched, and then the alternating direction multiplier method of a point spread function of the first imaging module and the second imaging module is used for decoding and reconstruction. Compared with a traditional imaging method, the application can break through the imaging frame rate limit of a camera, and has the advantages of high imaging resolution and small data quantity.

Description

Technical Field

[0001] This invention relates to the field of optical imaging, and more specifically to a high-speed, high-resolution compressed sensing imaging system and its imaging method. Background Technology

[0002] Optical imaging is a crucial way for humans to acquire information. However, with ever-increasing imaging demands, traditional imaging techniques can no longer meet researchers' needs in terms of imaging speed and resolution. Compressed sensing imaging technology, based on the mathematical assumption of signal sparsity, utilizes coded random sampling and post-processing algorithms to reconstruct images. Compared to traditional imaging methods, it breaks through the Nyquist sampling theorem, enabling accurate image reconstruction with minimal sampling, significantly improving imaging speed, and ultimately surpassing the frame rate limit of cameras. In recent years, researchers have gradually applied compressed sensing imaging technology to various imaging fields, such as aperture-coded snapshot spectral imaging (CASSI) in hyperspectral imaging and compressed ultrafast imaging (CUP) in ultrafast imaging.

[0003] To achieve compressed sensing imaging, improvements to traditional imaging systems are generally required in terms of hardware coding and algorithmic reconstruction. In terms of coded imaging, researchers typically use digital micromirror devices or spatial light modulators to load binary mask patterns to encode the imaging beam. However, because the modulation of the mask pattern prevents some pixels from transmitting light, this results in a loss of imaging light energy and information, further leading to insufficient reconstruction resolution and loss of detail.

[0004] In the field of decoding and reconstruction, researchers have utilized various inverse problem-solving frameworks to reconstruct compressed sensing images, such as the Two-Step Iterative Convergence Algorithm (TwIST), the Alternating Direction Multiplier Method (ADMM), and the Generalized Alternating Projection Method (GAP). Among these, ADMM is widely used due to its flexible structure, its ability to be combined with various priors, and its high reconstruction quality. However, because traditional ADMM does not consider the blurring effect of the optical point spread function on compressed sensing images, the final reconstructed image may suffer from low resolution. Summary of the Invention

[0005] To address the problems existing in the background art, the present invention provides a high-speed, high-resolution compressed sensing imaging system and its imaging method.

[0006] The technical solution adopted in this invention is:

[0007] The present invention includes a primary imaging module, a polarizing beam splitter, a quarter-wave plate, a hollow roof reflecting prism, a spatial light modulator, a secondary imaging module, and a camera. The light from the object is incident on the primary imaging module, and after passing through the primary imaging module, it is emitted to the polarizing beam splitter where it is transmitted and reflected. The transmitted light through the polarizing beam splitter is used as an encoded imaging beam, which is encoded and modulated by the spatial light modulator and then reflected back to the polarizing beam splitter. After passing through the secondary imaging module, it is incident on the target surface of the camera to form an encoded imaging image.

[0008] The reflected light from the polarizing beam splitter is transmitted as a normal imaging beam through a quarter-wave plate to a hollow roof reflecting prism. The beam is then reflected back to the quarter-wave plate by the hollow roof reflecting prism. Subsequently, the beam is transmitted through the quarter-wave plate, the polarizing beam splitter, and the secondary imaging module before being incident on the target surface of the camera to form a normal imaging image. The coded imaging image and the normal imaging image are located in different areas of the camera target surface, and the two are stitched together to form a composite image.

[0009] The light emitted from the object being tested is a beam of light carrying information about the spatial structure of the object.

[0010] It also includes a decoding and reconstruction unit, which is electrically connected to the camera and is used to decode and reconstruct the coded imaging image and ordinary imaging image acquired by the camera according to the mask coding information of the preset spatial light modulator to obtain the reconstructed image of the object under test in different time states during the camera exposure time.

[0011] Both the primary imaging module and the secondary imaging module are lenses or lens groups.

[0012] The object under test and the spatial light modulator form a pair of conjugate surfaces, and the spatial light modulator and the target surface of the camera form a pair of conjugate surfaces.

[0013] The position of the hollow roof ridge reflecting prism is set to ensure that the ordinary imaging image and the coded imaging image formed on the camera target surface do not overlap.

[0014] The imaging method of a high-speed, high-resolution compressed sensing imaging system includes the following steps:

[0015] S1. A composite image is acquired using a high-speed, high-resolution compressed sensing imaging system;

[0016] S2. Extract the coded imaging image and the ordinary imaging image from the composite image;

[0017] S3. Combining the mask coding information of the preset spatial light modulator and the optical point spread functions of the first-level and second-level imaging modules, the alternating direction multiplier method is used to reconstruct the image of the object under test from the coded imaging image and the ordinary imaging image, resulting in several reconstructed images of the object under test in different time states during the camera exposure time. The coded imaging image and the ordinary imaging image do not overlap.

[0018] The number of reconstructed images of the object under test in different time states obtained in step S3 is the same as the number of masks replaced by the spatial light modulator during the camera exposure time.

[0019] The reconstructed images of the object under test at different time states are set according to the following formula:

[0020]

[0021] O(x,y)=[O1(x,y),O2(x,y),…,O N (x,y)]

[0022] Where O(x,y) represents the light emitted by the object, I1 and I2 represent the coded image and the ordinary image, respectively, x and y represent the horizontal and vertical coordinates of the image, and PSF1 and PSF2 represent the optical point spread functions of the first-level and second-level imaging modules, respectively. Represents the convolution operation, O n M represents the light emitted from an object at the nth moment during a single camera exposure. n Let represent the mask pattern loaded by the spatial light modulator at time n during a single exposure of the camera, N represent the total number of mask transformations by the spatial light modulator during a single exposure of the camera, dx represent the pixel offset caused by the hollow roof prism, and R[O(x,y)] represent the prior constraints on the image.

[0023] The beneficial effects of this invention are:

[0024] In terms of coded imaging, this invention uses a polarizing beam splitter to divide the imaging beam into two parts: a coded imaging beam and a normal imaging beam. After being modulated by a spatial light modulator, the coded imaging beam and the light from the normal imaging beam are simultaneously imaged on the camera. In the face of the problem of pixel light blockage caused by the mask modulation of the existing imaging optical path, the introduction of a normal imaging optical path reduces light energy loss and light information loss, laying the foundation for high-resolution reconstruction.

[0025] This invention proposes an image reconstruction method with higher image resolution. It integrates the optical point spread function of the first-level imaging module and the second-level imaging module into the traditional alternating direction multiplier decoding framework, resulting in a higher resolution reconstructed image compared with traditional decoding methods.

[0026] This invention uses compressed sensing encoding imaging and decoding reconstruction as its basic principles to achieve high-speed imaging that exceeds the camera's frame rate. Attached Figure Description

[0027] Figure 1 This is a schematic diagram of the system workflow in this embodiment;

[0028] Figure 2 This is a schematic diagram of the optical system in this embodiment;

[0029] Figure 3 This is a schematic diagram of the camera exposure timing and spatial light modulator mask loading timing in this embodiment;

[0030] Figure 4 This is a diagram showing the simulated imaging results and the solution results of this embodiment;

[0031] Figure 5 This is a schematic diagram of the optical system structure of the present invention applied to light sheet microscopy imaging;

[0032] Figure 6 This is an XY plane illustration of the simulated imaging results and solution results of the present invention applied to light sheet microscopy;

[0033] Figure 7 This is an XZ plane diagram showing the simulated imaging results and solution results of the present invention applied to light sheet microscopy.

[0034] In the diagram: 1. Primary imaging module; 2. Polarizing beam splitter; 3. Quarter-wave plate; 4. Hollow roof reflecting prism; 5. Spatial light modulator; 6. Secondary imaging module; 7. Camera; 9. Laser emitting device. Detailed Implementation

[0035] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0036] like Figure 1 and Figure 2 As shown, this embodiment includes a laser emitting device 9, a primary imaging module 1, a polarizing beam splitter 2, a quarter-wave plate 3, a hollow roof reflecting prism 4, a spatial light modulator 5, a secondary imaging module 6, and a camera 7.

[0037] The light emitted from the object under test is incident on the first-level imaging module 1 as the object's light. After passing through the first-level imaging module 1, it is emitted to the polarizing beam splitter 2, where it is transmitted and reflected at right angles. The transmitted light from the polarizing beam splitter 2 is used as an encoded imaging beam. After being encoded and modulated by the spatial light modulator 5, it is reflected back to the polarizing beam splitter 2 and then reflected at right angles to the second-level imaging module 6. After passing through the second-level imaging module 6, it is incident on the target surface of the camera 7 to form an encoded imaging image. The right-angle reflected light from the polarizing beam splitter 2 is used as an ordinary imaging beam. It is transmitted through the quarter-wave plate 3 to the hollow roof reflecting prism 4 and reflected back to the quarter-wave plate 3. Then, it is transmitted through the quarter-wave plate 3 and the polarizing beam splitter 2 in sequence and then incident on the second-level imaging module 6. After passing through the second-level imaging module 6, it is incident on the target surface of the camera 7 to form an ordinary imaging image. The encoded imaging image and the ordinary imaging image are located in different areas of the target surface of the camera (7). The two are stitched together to form a composite image. That is, camera 7 outputs only one image, and the encoded image and the ordinary image are images of different regions of the same image output by camera 7.

[0038] The light emitted from the object under test is a beam of light carrying information about the spatial structure of the object. Specifically, this includes, but is not limited to, light reflected / transmitted / scattered by the object under test or light emitted by the object itself.

[0039] The target surface of camera 7 is specifically the imaging plane where the photosensitive detector inside the camera is located, also called the photosensitive surface, imaging surface, or sensor surface.

[0040] Normally, the intensity of the coded imaging beam is equal to that of the ordinary imaging beam.

[0041] Specifically, the light used to encode the imaging beam is P-ray, such as... Figure 2 As shown by the solid line, this portion of the light continues to propagate through the polarizing beam splitter 2 to the spatial light modulator 5, where it is modulated by the mask pattern and its polarization state changes to S-wave. This portion of the beam is then reflected 180° by the spatial light modulator and re-enters the polarizing beam splitter 2. The polarizing beam splitter 2 reflects the coded imaging beam 90° before it enters the secondary imaging module 6. The beam passing through the secondary imaging module 6 is finally imaged on the target surface of the camera 7.

[0042] The light in a typical imaging beam is S-ray, such as... Figure 2 As shown by the dashed line, this portion of the light is reflected 90° by the polarizing beam splitter 2 and first passes through the quarter-wave plate 3. It is then reflected sequentially on the two reflecting surfaces of the hollow roof ridge reflecting prism 4 and passes through the quarter-wave plate 3 again, becoming a P-beam. This beam continues to propagate through the polarizing beam splitter 2 to the secondary imaging module 6, and finally forms an image on the target surface of the camera 7.

[0043] The light from the object is incident on the first-level imaging module 1. The direction of the light incident on the first-level imaging module 1 is taken as the incident direction. The object under test, the first-level imaging module 1, the polarizing beam splitter 2, and the spatial light modulator 5 are arranged sequentially at intervals along the incident direction. The object under test and the spatial light modulator 5 form a pair of conjugate surfaces. The hollow roof reflecting prism 4, the quarter-wave plate 3, the polarizing beam splitter 2, the second-level imaging module 6, and the camera 7 are arranged sequentially at intervals along a direction perpendicular to the incident direction. The target surface of the spatial light modulator 5 and the camera 7 form a pair of conjugate surfaces.

[0044] It also includes a decoding and reconstruction unit, which is electrically connected to the camera 7. The decoding and reconstruction unit is used to decode and reconstruct the composite image acquired by the camera according to the preset mask coding information of the spatial light modulator 5 to obtain the reconstructed image of the object under test in different time states during the camera exposure time.

[0045] In macroscopic scenarios, both the primary imaging module 1 and the secondary imaging module 6 are lenses or lens groups.

[0046] The object under test and the spatial light modulator 5 form a pair of conjugate surfaces, and the spatial light modulator 5 and the target surface of the camera 7 form a pair of conjugate surfaces.

[0047] A conjugate surface is a pair of interrelated planes, where changes on one surface of an object are mapped onto the other. According to the principle of light reversibility, if a light source is placed at an object point, an image will be formed at an image point, and vice versa. These two corresponding points are called a pair of conjugate points, which can form conjugate lines, and thus conjugate surfaces.

[0048] Hollow roof prism 4 is mounted on a displacement stage, which allows the position of hollow roof prism 5 to be adjusted so that the light from the ordinary imaging beam is imaged precisely on the target surface of camera 7. The position of hollow roof prism 4 is set to ensure that the formed ordinary imaging image and the coded imaging image do not overlap on the target surface of camera 7; that is, the ordinary imaging image and the coded imaging image are located in different sub-regions of the camera output image and are independent of each other. By adjusting the position of hollow roof prism 4, the position of the ordinary imaging image on the target surface of camera 7 can be adjusted.

[0049] The hollow roof prism 4 has two reflecting surfaces arranged at 90°. The ordinary imaging beam is reflected sequentially from these two surfaces, causing the beam's propagation direction to reverse by 180° and its lateral position to shift. Therefore, the image of the ordinary imaging beam on the target surface of the camera 7 is exactly misaligned with the image of the coded imaging beam. Figure 2 As shown.

[0050] A high-speed, high-resolution compressed sensing imaging method includes the following steps:

[0051] S1. A composite image is acquired using a high-speed, high-resolution compressed sensing imaging system;

[0052] S2. Extract the coded imaging image and the ordinary imaging image from the composite image and perform pixel matching;

[0053] S3. Combining the mask coding information of the preset spatial light modulator 5 and the optical point spread function of the first-level imaging module 1 and the second-level imaging module 6, the alternating direction multiplier method is used to reconstruct the image of the object under test from the coded imaging image and the ordinary imaging image, so as to obtain the reconstructed image of the object under test in different time states during the camera exposure time.

[0054] The number of reconstructed images of the object under test in different time states during the camera exposure time is the same as the number of masks replaced by the spatial light modulator 5 during the camera exposure time.

[0055] Encoded images and ordinary images are located in different regions and do not overlap in the composite image acquired by the camera.

[0056] This embodiment includes two parts: an coded imaging method and a decoding reconstruction method. In the coded imaging part, the light emitted by the object first passes through various optical devices of the coded imaging unit and is finally imaged and recorded on the target surface of the camera 7. Then, in the decoding reconstruction part, the recorded raw imaging data is reconstructed into multiple high-resolution images through computer processing steps in the decoding reconstruction module.

[0057] The decoding and reconstruction method mainly includes: region matching of coded imaging data and ordinary imaging data, and data reconstruction using the alternating direction multiplier method. The coded imaging data matching step involves cropping coded and ordinary imaging data corresponding to the same object position from the acquired image data and performing pixel matching between the two images. The matched image data then enters the alternating direction multiplier method reconstruction step. Simultaneously, the optical point spread function of the first-level imaging module 1 and the second-level imaging module 6, and the mask pattern loaded on the spatial light modulator 5 are also input into the alternating direction multiplier method reconstruction step. After the reconstruction steps, a single acquired image data is reconstructed into multiple high-resolution images.

[0058] In this embodiment, the decoding and reconstruction unit first selects I1(x,y) and I2(x+dx,y) corresponding to the same region for pixel matching, and then uses the alternating direction multiplier method to fuse the optical point spread functions PSF1 and PSF2 of the first-level imaging module 1 and the second-level imaging module 6 to reconstruct O(x,y).

[0059] The spatial light modulator 5's encoding modulation is specifically as follows: the spatial light modulator 5 and the camera 7 are activated simultaneously. When the camera begins exposure, the spatial light modulator 5 starts loading the first preset mask according to its preset mask encoding information. When the camera exposure ends, the spatial light modulator 5 completes the loading of all preset masks according to its preset mask encoding information. The mask patterns loaded by the spatial light modulator 5 include, but are not limited to, binarized random mask patterns, such as... Figure 2 As shown.

[0060] Specifically, such as Figure 3 During the imaging process shown, the spatial light modulator 5 changes the mask multiple times during a single exposure of camera 7. The spatial light modulator 5 begins loading the first mask precisely at the moment camera 7 starts its exposure. Camera 7 achieves precise synchronization with the spatial light modulator 5 through three different timing control methods. These precise synchronization timing control methods include: Method 1: The light modulator 5 sends a signal each time it begins loading the first mask, and camera 7 starts its exposure upon receiving the signal; Method 2: Camera 7 sends a signal when it starts its exposure, and the spatial light modulator 5 starts loading the mask upon receiving the signal; Method 3: An external signal source sends a signal, and camera 7 and the spatial light modulator 5 start their respective exposure and mask loading processes upon receiving the signal.

[0061] In all three synchronization methods, the exposure time of camera 7 needs to be preset to be exactly equal to the time it takes for spatial light modulator 5 to load multiple masks.

[0062] The reconstructed images of the object under test at different time states are set according to the following formula:

[0063]

[0064] O(x,y)=[O1(x,y),O2(x,y),…,O N (x,y)]

[0065] Where O(x,y) represents the light from the object, I1 and I2 represent the images formed by the coded imaging beam and the ordinary imaging beam, respectively, x and y represent the horizontal and vertical coordinates on the image, and PSF1 and PSF2 represent the optical point spread functions of the first-level imaging module 1 and the second-level imaging module 6, respectively. Represents the convolution operation, O n M represents the light emitted by the object at the nth moment during a single exposure of camera 7. n dx represents the mask pattern loaded by the spatial light modulator 5 at the nth moment during one exposure of camera 7, N represents the total number of mask transformations by the spatial light modulator 5 during one exposure of camera 7, and dx represents the pixel offset caused by the hollow roof prism 4.

[0066] R[O(x,y)] represents the prior constraints on the image. The prior methods that can be used include, but are not limited to, total variation prior, L1 norm prior, Hessian continuity prior, weighted kernel norm minimization prior, etc.

[0067] Let the light emitted by the object be O. n (x,y), where x,y represent the horizontal and vertical coordinates on the image, and the first-level imaging module 1 pairs O n The imaging process of (x,y) can be used This indicates that, for a coded imaging beam, mask modulation on the spatial light modulator can be described as the product of the mask and the fluorescence, i.e. The imaging process of the secondary imaging module 2 can then be represented as follows: In a single acquisition by camera 7, the mask changes multiple times, so the image formed by the coded imaging beam can be described as:

[0068]

[0069] For a typical imaging beam, the hollow roof prism 4 acts as a lateral displacement of the beam; therefore, the image formed by the typical imaging beam can be described as follows:

[0070]

[0071] In the specific reconstruction steps, intermediate variables u and v need to be set, and these three variables are updated in each iteration.

[0072]

[0073] u i+1 =u i +ρ(O i+1 -v i+1 )

[0074] Here, ρ is an adjustable parameter, and i represents the number of updates.

[0075] like Figure 4 The diagram shown is a display of the simulated imaging results and the solution results of this embodiment, that is, the situation where the focal lengths of the primary imaging module 1 and the secondary imaging module 6 of the device are changed to suit normal macroscopic imaging. It should be noted that in this simulation, the exposure time of the camera 7 corresponds to 8 transformations of the mask pattern, but the present invention does not limit the number of transformations in the specific implementation process. Figure 4 a represents ordinary imaging data. Figure 4 b represents coded imaging data. Figure 4 c represents multiple resulting images decoded from ordinary imaging data and coded imaging data. The decoded and reconstructed images demonstrate the dynamic changes of the imaged person.

[0076] like Figure 5 The diagram shown is an optical structure schematic of one implementation method of the present invention applied to light sheet microscopy imaging, and... Figure 2 In contrast, this embodiment requires the addition of a laser emitting device 9 when applied to light sheet microscopy. The laser emitting device 9 includes a laser source for emitting laser light, a laser beam expander, a cylindrical lens, and an excitation objective. After the laser source emits laser light, it passes sequentially through the laser beam expander, the cylindrical lens, and the excitation objective to excite the imaging object. The laser beam expander is used to enlarge the diameter of the laser beam, the cylindrical lens shapes the circular laser beam into a sheet-like beam, and the excitation objective focuses the sheet-like beam onto the imaging object.

[0077] During the coded imaging process, such as Figure 3 As shown, during one exposure of camera 7, spatial light modulator 5 changes the mask multiple times, and the spatial light modulator 5 begins loading the first mask precisely at the moment each exposure of camera 7 begins. Simultaneously, the displacement stage continuously moves the imaging object, and each exposure of camera 7 records an image of the object at a different depth. It should be noted that in this simulation, one exposure time of camera 7 corresponds to eight mask pattern changes, but in specific implementations of this invention, the number of changes is not limited.

[0078] like Figure 6 The figure shown is an XY plane diagram illustrating the simulated imaging results and the solution results in one implementation method of the present invention applied to light sheet microscopy. Figure 6 The sample in the sample is ocular vascular fluorescence. Figure 6 a represents ordinary imaging data. Figure 6 b represents coded imaging data. Figure 6 c represents multiple resulting images decomposed from ordinary imaging data and coded imaging data. For example... Figure 6 As shown in the area marked by the dashed circle, the decoded and reconstructed images more clearly demonstrate the extension and changes of blood vessels.

[0079] like Figure 7 The image shown is an XZ plane view of the simulated imaging results and the calculated results in one implementation method of light sheet microscopy in this embodiment. This image is... Figure 6 The image shows the XZ plane effect. Figure 7 'a' represents the XZ plane effect of ordinary imaging data. Figure 7 b represents the XZ plane effect of the encoded imaging data. Figure 7 c represents the XZ plane effect of the resulting image calculated from ordinary imaging data and coded imaging data. A comparison of the three shows that, along the Z-axis, the decoded and reconstructed image has richer details and higher resolution.

[0080] This invention can be applied to situations where the object under test is moving at high speed or where the surface information of the object under test is scanned rapidly.

[0081] This invention comprises two units: coded imaging and decoded reconstruction. In the coded imaging section, light emitted from the object passes through a primary imaging module and is split into a coded imaging beam and a normal imaging beam by a polarizing beam splitter. The coded imaging beam is encoded, modulated, and reflected by a spatial light modulator before re-entering the polarizing beam splitter. The normal imaging beam is reflected by a hollow roof reflecting prism before also re-entering the polarizing beam splitter. The coded imaging beam and the normal imaging beam exiting the polarizing beam splitter then pass through a secondary imaging module and simultaneously image at different positions on the camera target surface. In the decoded reconstruction section, the coded imaging data and normal imaging data are first matched. Then, the alternating direction multiplier method, incorporating the point spread function from both the primary and secondary imaging modules, is used for decoded reconstruction.

[0082] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A high-speed, high-resolution compressed sensing imaging system, characterized in that: The system includes a primary imaging module (1), a polarizing beam splitter (2), a quarter-wave plate (3), a hollow roof reflecting prism (4), a spatial light modulator (5), a secondary imaging module (6), and a camera (7). Light emitted from the object under test is incident on the primary imaging module (1) as the object's light. After passing through the primary imaging module (1), the light is emitted to the polarizing beam splitter (2) where it is transmitted and reflected. The transmitted light through the polarizing beam splitter (2) is used as an encoded imaging beam. After being encoded and modulated by the spatial light modulator (5), the light is reflected back to the polarizing beam splitter (2) and then reflected to the secondary imaging module (6). After passing through the secondary imaging module (6), the light is incident on the target surface of the camera (7) to form an encoded imaging image. The reflected light from the polarizing beam splitter (2) is transmitted through the quarter-wave plate (3) to the hollow roof reflecting prism (4), and is reflected back to the quarter-wave plate (3) by the hollow roof reflecting prism (4). Then, it is transmitted through the quarter-wave plate (3), the polarizing beam splitter (2), and the secondary imaging module (6) in sequence before being incident on the target surface of the camera (7) to form a normal imaging image. The coded imaging image and the normal imaging image are located in different areas of the target surface of the camera (7), and the two are stitched together to form a composite image. The light emitted from the object being tested is a beam of light carrying information about the spatial structure of the object being tested. The position setting of the hollow roof ridge reflecting prism (4) satisfies that the ordinary imaging image and the coded imaging image formed on the target surface of the camera (7) do not overlap.

2. The high-speed, high-resolution compressed sensing imaging system according to claim 1, characterized in that: It also includes a decoding and reconstruction unit, which is electrically connected to the camera (7) and is used to decode and reconstruct the coded imaging image and ordinary imaging image acquired by the camera according to the mask encoding information of the preset spatial light modulator (5) to obtain the reconstructed image of the object under test in different time states during the exposure time of the camera (7).

3. The high-speed, high-resolution compressed sensing imaging system according to claim 1, characterized in that: Both the primary imaging module (1) and the secondary imaging module (6) are lenses or lens groups.

4. The high-speed, high-resolution compressed sensing imaging system according to claim 1, characterized in that: The object under test and the spatial light modulator (5) form a pair of conjugate surfaces, and the spatial light modulator (5) and the target surface of the camera (7) form a pair of conjugate surfaces.

5. An imaging method employing the high-speed, high-resolution compressed sensing imaging system as described in any one of claims 1-4, characterized in that, The method includes the following steps: S1. A composite image is acquired using a high-speed, high-resolution compressed sensing imaging system; S2. Extract the coded imaging image and the ordinary imaging image from the composite image; S3. Combining the mask coding information of the preset spatial light modulator (5) and the optical point spread function of the first-level imaging module (1) and the second-level imaging module (6), the alternating direction multiplier method is used to reconstruct the image of the object under test from the coded imaging image and the ordinary imaging image, and a number of reconstructed images of the object under test in different time states during the exposure time of the camera (7) are obtained. The reconstructed images of the object under test at different time states are specifically set according to the following formula: in, Represents the light of an object, These represent coded imaging images and ordinary imaging images, respectively. These represent the horizontal and vertical coordinates on the image, respectively. Let represent the optical point spread functions of the first-level imaging module (1) and the second-level imaging module (6), respectively. This represents the convolution operation. This represents prior constraints on the image. Let n represent the light emitted by the object at the nth moment during a single exposure of the camera (7). Let N represent the mask pattern loaded by the spatial light modulator (5) at the nth moment during one exposure of the camera (7), where N represents the total number of mask transformations by the spatial light modulator (5) during one exposure of the camera (7). This indicates the pixel offset caused by the hollow roof prism (4).

6. The high-speed, high-resolution compressed sensing imaging method according to claim 5, characterized in that: The coded imaging image and the ordinary imaging image do not overlap.

7. The high-speed, high-resolution compressed sensing imaging method according to claim 5, characterized in that: The number of reconstructed images of the object under test in different time states during the exposure time of the camera (7) obtained in step S3 is the same as the number of masks replaced by the spatial light modulator (5) during the exposure time of the camera (7).

Images