Single-pixel compressed holographic object imaging system and method

Abstract

The invention discloses a single-pixel compression holographic object imaging system and method, and the method comprises the steps: obtaining holographic interference pattern compression data of a reference light wave and an object light wave which are interfered, pass through a spatial light modulator and reach a single-pixel detector, and the spatial light modulator is provided with modulation parameters; inputting the holographic interference pattern compressed data into the trained neural network model to obtain a holographic image containing the object of interest; the imaging system comprises a light source, a first beam splitter, a first reflecting mirror, a first beam expander, a second reflecting mirror, a second beam expander, a second spectroscope, a spatial light modulator, a single-pixel detector and a processor, light emitted by the light source is split into two paths of light through the first beam splitter, and the two paths of light pass through the reflecting mirrors and the beam expanders respectively and then are combined through the second beam expander to form two paths of light; and the light reaches the single-pixel detector through the spatial light modulator. The embodiment of the invention can reduce data volume, facilitates data transmission and processing, and can be widely applied to the technical field of holographic imaging.

Description

Technical Field

[0001] The present invention relates to the field of holographic imaging technology, and in particular to a single-pixel compressed holographic object imaging system and method using deep learning technology. Background Art

[0002] In the actual work and life needs, people's demand for information is increasing. In work applications, more depth information is needed on the original two-dimensional picture to understand an object in order to meet more task requirements and carry out work. In life, people's demand for seeing more realistic virtual things is growing. Based on this, holographic images came into being. In traditional holographic technology, it mainly relies on the interference and diffraction principles of light. The image of the object is captured by recording the amplitude and phase information of the light waves emitted or reflected by the object. This usually requires the use of lasers as light sources, because lasers provide coherent light waves, which are necessary to produce interference patterns. The production of holograms involves interfering the object light waves with the reference light waves on the photosensitive plate to form interference fringes, which can then be illuminated by appropriate light sources to reproduce the image of the original object. Because traditional holographic technology is accompanied by a large cost for single imaging and is not easy to transmit data, modern holographic technologies such as digital holography have come into the public eye, but digital holography involves the same large amount of data collection and processing, which has high requirements for computing resources. Summary of the invention

[0003] In view of this, in order to solve one of the above problems, an embodiment of the present invention aims to provide a single-pixel compressed holographic object imaging system and method to reduce the amount of data and facilitate data transmission and processing.

[0004] On the one hand, an embodiment of the present invention provides a single-pixel compressed holographic object imaging method, comprising:

[0005] Obtaining holographic interference pattern compression data after the reference light wave and the object light wave interfere and then pass through a spatial light modulator to reach a single pixel detector; the spatial light modulator is provided with modulation parameters;

[0006] The holographic interference pattern compressed data is input into a trained neural network model to obtain a holographic image containing an object of interest.

[0007] Optionally, the neural network model is trained by the following method:

[0008] Acquire sample data; the sample data includes an original holographic interference sample image and a holographic sample image containing a sample of an object of interest;

[0009] Inputting the original holographic interference sample image into a binary convolution layer to obtain compressed sample data;

[0010] Inputting the compressed sample data into a neural network model to obtain a holographic image containing a sample of an object of interest;

[0011] The convolution kernel parameters of the binary convolution layer and the model parameters of the neural network model are updated according to the holographic image containing the sample of the object of interest and the holographic sample image containing the object of interest until the preset requirements are met.

[0012] Optionally, the convolution kernel parameters of the binary convolution layer after training are used as modulation parameters of the spatial light modulator.

[0013] Optionally, the sample data includes an experimental data set and / or a simulation data set.

[0014] Optionally, the simulation data set is obtained by the following method:

[0015] Performing hologram simulation on an image containing a sample of an object of interest to obtain an original holographic interference sample image;

[0016] Reconstructing the original holographic interference sample image to restore the image scene information, and filtering out the background and uninteresting objects to obtain a holographic sample image containing a sample of the object of interest;

[0017] A simulation data set is formed according to the plurality of original holographic interference sample images and the plurality of holographic sample images containing samples of the object of interest.

[0018] Optionally, the experimental data set is obtained by the following method:

[0019] Acquire an original holographic interference sample image generated by interference between a reference sample light wave and a sample object light wave containing a sample of an object of interest;

[0020] Performing digital holographic processing on the original holographic interference sample image to restore image scene information, and filtering out background and uninteresting objects to obtain a holographic sample image containing samples of the object of interest;

[0021] An experimental data set is formed according to the plurality of original holographic interference sample images and the plurality of holographic sample images containing samples of the object of interest.

[0022] On the other hand, an embodiment of the present invention provides a single-pixel compressed holographic object imaging system, comprising:

[0023] The first module is used to obtain the holographic interference pattern compression data of the reference light wave and the object light wave after passing through the spatial light modulator and arriving at the single pixel detector; the spatial light modulator is provided with modulation parameters;

[0024] The second module is used to input the holographic interference pattern compression data into the trained neural network model to obtain a holographic image containing the object of interest.

[0025] On the other hand, an embodiment of the present invention provides a single-pixel compressed holographic object imaging device, comprising:

[0026] at least one processor;

[0027] at least one memory for storing at least one program;

[0028] When the at least one program is executed by the at least one processor, the at least one processor implements the above method.

[0029] On the other hand, an embodiment of the present invention provides a computer-readable storage medium, in which a program executable by a processor is stored. When the program executable by the processor is executed by the processor, it is used to perform the above method.

[0030] On the other hand, an embodiment of the present invention provides a single-pixel compressed holographic object imaging system, including a light source, a first beam splitter, a first reflector, a first beam expander, a second reflector, a second beam expander, a second beam splitter, a spatial light modulator, a single-pixel detector and a processor. The light emitted by the light source is divided into a first light path and a second light path by the first beam splitter. The first light path passes through the first reflector and the first beam expander in sequence and reaches the second beam expander. The second light path passes through the second reflector, the second beam expander and the target object in sequence and reaches the second beam expander. After the first light path and the second light path are combined by the second beam expander, they pass through the spatial light modulator to reach the single-pixel detector. The spatial light modulator is set with modulation parameters. The processor is used to implement the above method.

[0031] Implementation of the embodiments of the present invention includes the following beneficial effects: In the embodiment, firstly, compressed data of the holographic interference pattern is obtained after the reference light wave and the object light wave interfere with each other and then pass through the spatial light modulator to reach the single-pixel detector. The spatial light modulator is provided with modulation parameters. The holographic interference pattern is compressed with the help of the modulation parameters of the spatial light modulator. Then, the compressed data of the holographic interference pattern is input into the trained neural network model. The parameters in the neural network model complete the screening and extraction of the holographic objects of the specific holographic interference pattern, that is, the identification of the holographic object of interest. When imaging, only the three-dimensional information of the object of interest is contained, and irrelevant background and non-object of interest information are filtered out, so as to obtain a holographic image containing only the object of interest, and end-to-end compression and reconstruction of the object of interest are completed. Only a small amount of data obtained by compressing the holographic interference pattern is needed to restore the information of the object of interest. This not only gets rid of the difficulty of large amounts of data being difficult to transmit and high cost, but also can process a large amount of optical information at a single time, ensuring the quality and integrity of the image, and facilitating wide deployment in practical applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 is a structural schematic diagram of a single-pixel compressed holographic object imaging system provided by an embodiment of the present invention;

[0033] Figure 2 is a schematic flow chart of the steps of a single-pixel compressed holographic object imaging method provided by an embodiment of the present invention;

[0034] Figure 3 is a schematic diagram of a set of experimental data provided by an embodiment of the present invention;

[0035] Figure 4 is a schematic diagram of a set of prediction data provided by an embodiment of the present invention;

[0036] Figure 5 is a structural block diagram of a single-pixel compressed holographic object imaging system provided by an embodiment of the present invention;

[0037] Figure 6 It is a structural block diagram of a single-pixel compressed holographic object imaging device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0038] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The step numbers in the following embodiments are only provided for the convenience of explanation and description, and the order between the steps is not limited in any way. The execution order of each step in the embodiment can be adaptively adjusted according to the understanding of those skilled in the art.

[0039] See also Figure 1The embodiment of the present invention provides a single-pixel compressed holographic object imaging system, comprising a light source (1-1), a first beam splitter (1-2), a first reflector (1-3), a first beam expander (1-4), a second reflector (1-6), a second beam expander (1-7), a second beam splitter (1-5), a spatial light modulator (1-9), a single-pixel detector (1-10) and a processor (1-11). Light emitted by the light source (1-1) is divided into a first light path and a second light path by the first beam splitter (1-2). The first light path is sequentially divided into a first light path and a second light path by the first reflector (1-3) and the first beam expander (1-11). -4) and then arrive at the second beam expander (1-5); the second light path passes through the second reflector (1-6), the second beam expander (1-7) and the target object (1-8) in sequence and then arrives at the second beam expander (1-5); the first light path and the second light path are combined at the second beam expander (1-5) and then pass through the spatial light modulator (1-9) and arrive at the single-pixel detector (1-10); the spatial light modulator (1-10) is provided with modulation parameters; and the processor (1-11) is used for compressing data of a holographic interference pattern acquired by the single-pixel detector (1-10) to obtain a holographic image containing only the object of interest.

[0040] like Figure 2 As shown, an embodiment of the present invention provides a single-pixel compressed holographic object imaging method, comprising:

[0041] S100, obtaining holographic interference pattern compression data after the reference light wave and the object light wave interfere with each other and then pass through the spatial light modulator to reach the single pixel detector; the spatial light modulator is set with modulation parameters;

[0042] S200, inputting the compressed data of the holographic interference pattern into the trained neural network model to obtain a holographic image containing the object of interest.

[0043] The object light wave refers to the light wave generated after the light passes through the target object, and the reference light wave refers to the light wave generated after the same light does not pass through the target object.

[0044] Specifically, after the reference light wave and the object light wave are combined and interfered, they pass through a spatial light modulator, which compresses the combined light wave. The compressed combined light wave forms a light intensity value in a single-pixel detector, and holographic interference pattern compression data is obtained. The holographic interference pattern compression data is input into a trained neural network model to obtain a holographic image containing the object of interest.

[0045] It should be noted that the spatial light modulation is determined according to the actual application, and this embodiment does not make any specific restrictions. For example, the spatial light modulator is implemented by using a DMD (Digital Micromirror Device). The structure of the neural network model is determined according to the actual application, and this embodiment does not make any specific restrictions. For example, the neural network model is a U-Net network model.

[0046] Optionally, the neural network model is trained by:

[0047] S010, acquiring sample data; the sample data includes an original holographic interference sample image and a holographic sample image containing a sample of an object of interest;

[0048] S020, inputting the original holographic interference sample image into the binary convolution layer to obtain compressed sample data;

[0049] S030, inputting the compressed sample data into the neural network model to obtain a holographic image containing the sample of the object of interest;

[0050] S040. Update the convolution kernel parameters of the binary convolution layer and the model parameters of the neural network model according to the holographic image containing the sample of the object of interest and the holographic sample image containing the object of interest until the preset requirements are met.

[0051] It should be noted that during the training process, whether the preset requirements for the end of training are met is determined according to the actual application, and this embodiment does not impose any specific restrictions. For example, the training is ended when the number of training times is met; or the difference between the result of the holographic image containing the sample of the object of interest after the neural network output and the holographic sample image containing the object of interest meets the accuracy requirements, and the training is ended.

[0052] In a specific embodiment, the neural network model is implemented by selecting a U-Net network. The U-Net network is a convolutional neural network (CNN) for image segmentation, which can be used for image segmentation tasks. The structure of U-Net is a symmetrical encoder-decoder structure, in which the encoder adopts a stacked convolutional layer structure, and each encoder stage includes a convolution layer and a pooling layer. The decoder is symmetrical with the encoder, and each decoder stage includes an upsampling operation (deconvolution or interpolation) and a convolution layer. In the decoder stage, a skip connection is also introduced to merge the feature map of the encoder with the feature map of the decoder to improve the accuracy of the model. The advantage of U-Net is that it can process relatively small data sets, has high segmentation accuracy, and can be applied to different image segmentation tasks.

[0053] On the basis of the U-Net network, a binary convolution layer is added to match the DMD lens (spatial light modulator) of the single-pixel system. The binary convolution kernel parameters are limited to a 64*64 matrix that can only produce two values ​​of 0 and 1, corresponding to the two states of the DMD micromirror 12° and -12° respectively. Set different compression rates. For example, for a 64*64 image, if the compression rate is 5%, the number of convolution kernels is 64*64*5%=205. All images in the training set and test set are set to 64*64 size. After the original holographic interference sample image in the training set is input into the network, it first passes through the binary convolution layer. Each input pattern generates a corresponding number of values ​​after passing through the binary convolution kernel. The corresponding number of values ​​is then trained through the U-Net network. After the training is completed, all parameters of the training network and the kernel parameters of the binary convolution layer are saved.

[0054] Optionally, the convolution kernel parameters of the binary convolution layer after training are used as modulation parameters of the spatial light modulator.

[0055] After the training is completed, the convolution kernel parameters of the binary convolution layer are used as the modulation parameters of the spatial light modulator, that is, the convolution kernel parameters of the binary convolution layer are loaded into the DMD lens (spatial light modulator).

[0056] Optionally, the sample data includes an experimental data set and / or a simulation data set.

[0057] The experimental data set and the simulation data set complement each other. When the experimental data set is insufficient, the simulation data set can be supplemented. The experimental data set and the simulation data set jointly construct sample data, which can improve the accuracy of neural network model training.

[0058] Optionally, the simulation data set is obtained by:

[0059] Performing hologram simulation on an image containing a sample of an object of interest to obtain an original holographic interference sample image;

[0060] The original holographic interference sample image is reconstructed to restore the image scene information, and the background and uninteresting objects are filtered out to obtain a holographic sample image containing the object of interest sample;

[0061] A simulation data set is formed according to a number of original holographic interference sample images and a number of holographic sample images containing samples of an object of interest.

[0062] Specifically, for an image containing a sample of an object of interest, such as Figure 3 As shown in (a), the hologram simulation is performed in Matlab software under the same conditions as the experiment to produce the original holographic interference sample image, as shown in Figure 3As shown in (b), the original holographic interference sample image is reconstructed to restore the image scene information. The background and uninteresting objects are filtered out for the restored scene information, and only the object information of interest is retained to obtain a holographic sample image containing only the object sample of interest, as shown in Figure 3 As shown in (c), the original holographic interference sample image is used as the label image and the original image input to the neural network model to construct a simulation data set.

[0063] Optionally, the experimental dataset is obtained by:

[0064] Acquire an original holographic interference sample image generated by interference between a reference sample light wave and a sample object light wave containing a sample of an object of interest;

[0065] Performing digital holographic processing on the original holographic interference sample image to restore the image scene information, and filtering out the background and uninteresting objects to obtain a holographic sample image containing the object of interest sample;

[0066] An experimental data set is formed according to a number of original holographic interference sample images and a number of holographic sample images containing samples of objects of interest.

[0067] In a laboratory environment, reference sample light waves and sample object light waves are used to record scene information containing objects of interest through CCD (Charge Coupled Device)\COMS (Complementary Metal-Oxide-Semiconductor) to obtain the original holographic interference sample image. Digital holographic technology is used to restore the scene information of the object of interest in a computer. For the restored scene information, the background and objects of no interest are filtered out, and only the information of the object of interest is retained. A holographic sample image containing only the sample of the object of interest is obtained as the label image. The original holographic interference sample image recorded by CCD is used as the original image input to the neural network model to construct an experimental data set.

[0068] In a specific embodiment, see Figure 4 , Figure 4 (a) is an image containing the target object and scene information. Figure 4 (b) is the light intensity value recorded by a single pixel detector. Figure 4 The holographic interference pattern recorded in (b) is input into the trained neural network model, and the following is obtained: Figure 4 (c) shows a holographic image containing only the object of interest.

[0069] Implementation of the embodiments of the present invention includes the following beneficial effects: In the embodiment, firstly, the holographic interference pattern compression data of the reference light wave and the object light wave after passing through the spatial light modulator and arriving at the single-pixel detector is obtained, the spatial light modulator is provided with modulation parameters, and the holographic interference pattern is compressed with the help of the modulation parameters of the spatial light modulator, and then the holographic interference pattern compression data is input into the trained neural network model, and the parameters in the neural network model complete the screening and extraction of the holographic objects of the specific holographic interference pattern, that is, the identification of the holographic object of interest, and the imaging only contains the three-dimensional information of the object of interest, and the irrelevant background and non-object of interest information are filtered out, so as to obtain a holographic image containing only the object of interest, and the end-to-end compression and reconstruction of the object of interest is completed, and only a small amount of data obtained by the holographic interference pattern compression is needed to restore the information of the object of interest, which not only gets rid of the difficulty of large amounts of data being difficult to transmit and high cost, but also can process a large amount of optical information at a single time, ensure the quality and integrity of the image, and facilitate wide deployment in practical applications.

[0070] See also Figure 5 , an embodiment of the present invention provides a single-pixel compressed holographic object imaging system, comprising:

[0071] The first module is used to obtain the holographic interference pattern compression data of the reference light wave and the object light wave after passing through the spatial light modulator and arriving at the single pixel detector; the spatial light modulator is set with modulation parameters;

[0072] The second module is used to input the compressed data of the holographic interference pattern into the trained neural network model to obtain a holographic image containing the object of interest.

[0073] It can be seen that the contents of the above method embodiments are all applicable to the present system embodiments, the functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0074] See also Figure 6 , an embodiment of the present invention provides a single-pixel compressed holographic object imaging device, comprising:

[0075] at least one processor;

[0076] at least one memory for storing at least one program;

[0077] When at least one program is executed by at least one processor, the at least one processor implements the above method.

[0078] Among them, the memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. The memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a remote memory remotely arranged relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0079] It can be seen that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0080] In addition, the embodiment of the present application also discloses a computer program product or a computer program, which is stored in a computer-readable storage medium. The processor of the computer device can read the computer program from the computer-readable storage medium, and the processor executes the computer program, so that the computer device performs the above method.

[0081] The embodiment of the present invention further provides a computer-readable storage medium, which stores a program executable by a processor, and the program executable by the processor is used to implement the above method when executed by the processor. Similarly, the contents of the above method embodiment are all applicable to the present storage medium embodiment, and the functions specifically implemented by the present storage medium embodiment are the same as those of the above method embodiment, and the beneficial effects achieved are also the same as those achieved by the above method embodiment.

[0082] It is understood that all or some steps and systems in the disclosed method above can be implemented as software, firmware, hardware and appropriate combinations thereof. Some physical components or all physical components can be implemented as software executed by a processor, such as a central processing unit, a digital signal processor or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, and the computer-readable medium can include a computer storage medium (or a non-transitory medium) and a communication medium (or a temporary medium). As known to those of ordinary skill in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cassette, magnetic tape, disk storage or other magnetic storage device, or any other medium that can be used to store desired information and can be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0083] See also Figure 1 The embodiment of the present invention provides a single-pixel compressed holographic object imaging system, comprising a light source (1-1), a first beam splitter (1-2), a first reflector (1-3), a first beam expander (1-4), a second reflector (1-6), a second beam expander (1-7), a second beam splitter (1-5), a spatial light modulator (1-9), a single-pixel detector (1-10) and a processor (1-11). Light emitted by the light source (1-1) is split into a first light path and a second light path by the first beam splitter (1-2). The first light path is sequentially passed through the first reflector (1-1). -3) and the first beam expander (1-4) before reaching the second beam expander (1-5); the second light path passes through the second reflector (1-6), the second beam expander (1-7) and the target object (1-8) in sequence before reaching the second beam expander (1-5); the first light path and the second light path are combined in the second beam expander (1-5) and then pass through the spatial light modulator (1-9) before reaching the single-pixel detector (1-10); the spatial light modulator (1-10) is provided with modulation parameters; and the processor (1-11) is used to implement the above-mentioned single-pixel compressed holographic imaging method based on deep learning.

[0084] It should be noted that the light source is determined according to the actual application and is not specifically limited in this embodiment. For example, the light source is a laser. It is understandable that the processor can be an intelligent terminal, a laptop computer, a desktop computer or a mobile phone.

[0085] It can be seen that the contents of the above method embodiments are all applicable to the present system embodiments, the functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0086] The above is a specific description of the preferred implementation of the present invention, but the invention is not limited to the embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A single-pixel compressed holographic object imaging method, characterized in that: include: Obtaining holographic interference pattern compression data of the reference light wave and the object light wave after passing through the spatial light modulator and arriving at the single-pixel detector; The spatial light modulator is provided with modulation parameters; The holographic interference pattern compressed data is input into a trained neural network model to obtain a holographic image containing an object of interest.

2. The method according to claim 1, characterized in that: The neural network model is trained by the following method: Acquire sample data; the sample data includes an original holographic interference sample image and a holographic sample image containing a sample of an object of interest; Inputting the original holographic interference sample image into a binary convolution layer to obtain compressed sample data; Inputting the compressed sample data into a neural network model to obtain a holographic image containing a sample of an object of interest; The convolution kernel parameters of the binary convolution layer and the model parameters of the neural network model are updated according to the holographic image containing the sample of the object of interest and the holographic sample image containing the object of interest until the preset requirements are met.

3. The method according to claim 2, characterized in that The convolution kernel parameters of the binary convolution layer after training are used as the modulation parameters of the spatial light modulator.

4. The method according to claim 2, characterized in that: The sample data includes an experimental data set and / or a simulation data set.

5. The method according to claim 4, characterized in that The simulation data set is obtained by the following method: Performing hologram simulation on an image containing a sample of an object of interest to obtain an original holographic interference sample image; Reconstructing the original holographic interference sample image to restore the image scene information, and filtering out the background and uninteresting objects to obtain a holographic sample image containing a sample of the object of interest; A simulation data set is formed according to the plurality of original holographic interference sample images and the plurality of holographic sample images containing samples of the object of interest.

6. The method according to claim 4, characterized in that The experimental data set is obtained by the following method: Acquire an original holographic interference sample image generated by interference between a reference sample light wave and a sample object light wave containing a sample of an object of interest; Performing digital holographic processing on the original holographic interference sample image to restore image scene information, and filtering out background and uninteresting objects to obtain a holographic sample image containing samples of the object of interest; An experimental data set is formed according to the plurality of original holographic interference sample images and the plurality of holographic sample images containing samples of the object of interest.

7. A single-pixel compressed holographic object imaging system, characterized in that: include: The first module is used to obtain the compressed data of the holographic interference pattern after the reference light wave and the object light wave interfere with each other and then pass through the spatial light modulator to reach the single-pixel detector; The spatial light modulator is provided with modulation parameters; The second module is used to input the holographic interference pattern compression data into the trained neural network model to obtain a holographic image containing the object of interest.

8. A single-pixel compressed holographic object imaging device, characterized in that: include: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a program executable by a processor, characterized in that: The processor-executable program is used to perform the method according to any one of claims 1 to 6 when executed by the processor.

10. A single-pixel compressed holographic object imaging system, characterized in that: The invention comprises a light source, a first beam splitter, a first reflector, a first beam expander, a second reflector, a second beam expander, a second beam splitter, a spatial light modulator, a single-pixel detector and a processor, wherein the light emitted by the light source is divided into a first light path and a second light path by the first beam splitter, the first light path passes through the first reflector and the first beam expander in sequence and then reaches the second beam expander, the second light path passes through the second reflector, the second beam expander and the target object in sequence and then reaches the second beam expander, the first light path and the second light path are combined by the second beam expander and then pass through the spatial light modulator to reach the single-pixel detector, the spatial light modulator is provided with modulation parameters, and the processor is used to implement the method as described in any one of claims 1 to 6.

Images