A binary Fourier single-pixel imaging method, system, device and medium

Abstract

The application discloses a kind of binary Fourier single-pixel imaging method, system, device and medium, method includes: based on error diffusion template, by m kind of pixel scanning order to Fourier base pattern is error diffusion, obtains m group binary Fourier base pattern;M group binary Fourier base pattern is projected respectively to processing, and corresponding m two-dimensional target image is reconstructed to obtain;Select n two-dimensional target image to carry out fusion processing, and obtain target image.The application obtains multiple groups of binary Fourier base pattern by different scanning order, reconstructs by the projection of binary Fourier base pattern, finally, the reconstructed two-dimensional target image is fused, effectively reduce the quantization noise and environmental noise of image, while the high-frequency information of image is preferably retained, so that the imaging quality is greatly improved, while realizing the function of good noise reduction, without losing image detail part due to excessive smoothing, can be widely applied in optical imaging technology field.

Description

Technical Field

[0001] This invention relates to the field of optical imaging technology, and in particular to a binarized Fourier single-pixel imaging method, system, device, and medium. Background Technology

[0002] Fourier single-pixel imaging is a single-pixel imaging method that uses a Fourier basis pattern as a modulation pattern. It modulates the target scene by projecting the Fourier basis pattern. A single-pixel detector collects light intensity information, demodulates this information, and finally reconstructs a two-dimensional image of the target scene. However, the Fourier basis pattern is multi-grayscale, while high-speed spatial light modulators can only display binary patterns. Therefore, it is difficult to display Fourier basis patterns at high speed on high-speed spatial light modulators.

[0003] In 1976, Robert W. Floyd et al. proposed an error diffusion template (Floyd-Steinberg error diffusion template) for binarizing multi-grayscale images. In 2017, Zhang et al. used this error diffusion template to binarize Fourier basis patterns, enabling the projection of Fourier basis patterns using a high-speed spatial light modulator (DMD), successfully achieving fast Fourier single-pixel imaging. However, this method has a significant drawback: it introduces considerable quantization noise during binarization, typically manifested as noticeable mesh patterns in the reconstructed image. This quantization noise affects the visual quality of the reconstructed image and obscures its high-frequency characteristics, severely impacting image quality.

[0004] Resolution is one of the many evaluation criteria for image quality. In past Fourier single-pixel imaging, the resolution of reconstructed images often fell short of the maximum resolution of spatial light modulators. This limitation stems from the significant quantization noise introduced during the binarization of the Fourier basis pattern using the Floyd-Steinberg error diffusion template. To suppress this noise, researchers typically upsample the Fourier basis pattern before error diffusion, thus reducing quantization noise. However, this also leads to a decrease in image resolution. Therefore, this upsampling method reduces the impact of quantization noise on the reconstructed image by sacrificing spatial resolution. While this method can suppress quantization noise to some extent, it also significantly reduces image resolution, resulting in a lower overall image quality. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide a binarized Fourier single-pixel imaging method, system, device, and medium, which can effectively reduce quantization noise and environmental noise, and improve imaging effect and quality.

[0006] On one hand, embodiments of the present invention provide a binarized Fourier single-pixel imaging method, comprising:

[0007] Based on the error diffusion template, the Fourier basis pattern is subjected to error diffusion through m pixel scanning sequences to obtain m sets of binary Fourier basis patterns.

[0008] Projection processing is performed on each of the m sets of binarized Fourier basis patterns to reconstruct the corresponding m two-dimensional target images;

[0009] Select n two-dimensional target images and perform fusion processing to obtain the target image; where n is less than or equal to m.

[0010] Optionally, the step of performing error diffusion on the Fourier basis pattern using m pixel scanning sequences based on the error diffusion template to obtain m sets of binary Fourier basis patterns includes:

[0011] The pixel values ​​of each pixel in the Fourier basis pattern are quantized to obtain the quantization error.

[0012] Based on the quantization error, weight allocation is performed on the unbindified pixels of the pixel using the error diffusion template to complete the error diffusion of the pixel.

[0013] Based on the error diffusion of all the pixels, the error diffusion of the Fourier basis pattern is completed, resulting in a set of binarized Fourier basis patterns.

[0014] The Fourier base pattern is flipped and / or rotated, and then the step of quantizing the pixel values ​​of each pixel in the Fourier base pattern to obtain the quantization error is returned, until m sets of binary Fourier base patterns are obtained.

[0015] Optionally, the step of projecting the m sets of binarized Fourier basis patterns to reconstruct the corresponding m two-dimensional target images includes:

[0016] The target scene is obtained by projecting m sets of binary Fourier basis patterns onto each other using a spatial light modulator.

[0017] The modulated voltage signal of the target scene is collected by a single-pixel detector;

[0018] The Fourier spectrum of the target scene is obtained by solving the phase-shift algorithm on the modulated voltage signal.

[0019] The inverse Fourier spectrum is then transformed to obtain a two-dimensional target image;

[0020] The number of the modulated voltage signal, the Fourier spectrum, and the two-dimensional target image are all m.

[0021] Optionally, the step of solving the phase-shift algorithm on the modulated voltage signal to obtain the Fourier spectrum of the target scene includes:

[0022] The modulation voltage signal is solved by an N-step phase shift algorithm to obtain the Fourier coefficients;

[0023] Based on the Fourier coefficients, the Fourier spectrum of the target scene is filled in;

[0024] Where N is greater than or equal to 2, the formula for the N-step phase shift algorithm is:

[0025]

[0026] In the formula, C(f) x ,f y ) represents the Fourier coefficients, N represents the number of phase shift steps, and D * f represents the modulated voltage signal. x f represents the spatial frequency in the x-direction. x The frequency in the y-direction is represented by , and j represents the imaginary unit.

[0027] Optionally, the step of performing an inverse transform on the Fourier spectrum to obtain a two-dimensional target image includes:

[0028] Based on the inverse transform formula, the Fourier spectrum is inversely transformed to obtain a two-dimensional target image;

[0029] The inverse transformation formula is as follows:

[0030] I(x,y)=F -1 {C(f x ,f y )}

[0031] In the formula, I(x,y) represents a two-dimensional target image; F -1 Represents the inverse Fourier transform; C(f) x ,f y ) represents the Fourier coefficients, and the Fourier spectrum is obtained by filling in the Fourier coefficients.

[0032] Optionally, the step of selecting n two-dimensional target images for fusion processing to obtain a target image includes:

[0033] Based on the n two-dimensional target images, one of them is determined as the reference image, and the rest are images to be registered.

[0034] Based on the reference image, the translation amount of the image to be registered relative to the reference image is obtained;

[0035] Based on the translation amount, register n of the two-dimensional target images;

[0036] The registered n two-dimensional target images are fused to obtain the target image.

[0037] Optionally, fusing the registered n two-dimensional target images to obtain the target image includes:

[0038] Based on the fusion formula, combined with the weight estimation parameters and image prior weights, the registered n two-dimensional target images are fused to obtain the target image;

[0039] The fusion formula is as follows:

[0040]

[0041] In the formula, x represents the target image, t represents the number of iterations, and x t Let p() represent the target image after t iterations, p() represent the probability, y represent the two-dimensional target image, and β represent the probability. t α represents the weight estimation parameter. t Let F represent the prior weights of the image, and let F represent the total energy function.

[0042] On the other hand, embodiments of the present invention provide a binarized Fourier single-pixel imaging system, including:

[0043] The first module is used to perform error diffusion on the Fourier basis pattern based on the error diffusion template and through m pixel scanning sequences to obtain m sets of binary Fourier basis patterns.

[0044] The second module is used to project the m sets of binary Fourier basis patterns respectively to reconstruct the corresponding m two-dimensional target images;

[0045] The third module is used to select n two-dimensional target images for fusion processing to obtain a target image; where n is less than or equal to m.

[0046] On the other hand, embodiments of the present invention provide a binarized Fourier single-pixel imaging device, including a processor and a memory;

[0047] The memory is used to store programs;

[0048] The processor executes the program to implement the method described above.

[0049] On the other hand, embodiments of the present invention provide a computer-readable storage medium storing a program that is executed by a processor to implement the methods described above.

[0050] This invention also discloses a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device can read the computer instructions from the computer-readable storage medium and execute the computer instructions, causing the computer device to perform the aforementioned method.

[0051] This invention first uses an error diffusion template to perform error diffusion on Fourier basis patterns through m different pixel scanning sequences, resulting in m sets of binarized Fourier basis patterns. Each of these m sets of binarized Fourier basis patterns is then projected to reconstruct m corresponding two-dimensional target images. Finally, n of these two-dimensional target images are fused to obtain the target image; where n is less than or equal to m. This invention obtains multiple sets of corresponding binarized Fourier basis patterns through different scanning sequences, and then, through projection and reconstruction of these patterns, finally fuses the reconstructed two-dimensional target images. This effectively reduces quantization noise and environmental noise while better preserving high-frequency information, significantly improving image quality. It achieves good noise reduction without losing image details due to excessive smoothing. This invention effectively reduces quantization noise and environmental noise, improving imaging effect and quality. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0053] Figure 1 A schematic flowchart of a binarized Fourier single-pixel imaging method provided in an embodiment of the present invention;

[0054] Figure 2 A schematic diagram illustrating two error propagation directions provided in an embodiment of the present invention;

[0055] Figure 3 A schematic diagram of an original target image provided for an embodiment of the present invention;

[0056] Figure 4 This is a schematic diagram of another original target image provided for an embodiment of the present invention;

[0057] Figure 5 This is a schematic diagram of a target image obtained by fusing two original target images according to an embodiment of the present invention;

[0058] Figure 6This is a schematic diagram of a binarized Fourier single-pixel imaging system provided in an embodiment of the present invention;

[0059] Figure 7 This is a schematic diagram of a binarized Fourier single-pixel imaging device provided in an embodiment of the present invention. Detailed Implementation

[0060] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0061] On the one hand, refer to Figure 1 The present invention provides a binarized Fourier single-pixel imaging method, comprising:

[0062] S100. Based on the error diffusion template, the Fourier base pattern is subjected to error diffusion through m pixel scanning sequences to obtain m sets of binary Fourier base patterns.

[0063] It should be noted that in some embodiments, the pixel values ​​of each pixel in the Fourier basis pattern are quantized to obtain the quantization error; based on the quantization error, the weights of the unbinarized pixels are assigned according to the error diffusion template to complete the error diffusion of the pixels; based on the error diffusion of all pixels, the error diffusion of the Fourier basis pattern is completed to obtain a set of binarized Fourier basis patterns; the Fourier basis pattern is flipped and / or rotated, and then the step of quantizing the pixel values ​​of each pixel in the Fourier basis pattern to obtain the quantization error is returned, until m sets of binarized Fourier basis patterns are obtained.

[0064] The process involves several steps: First, the grayscale value of a pixel on the Fourier base pattern is quantized to be the smallest integer closest to it. The difference between the original value and the quantized value is calculated to obtain the quantization error. This error is then multiplied by different weights, and the grayscale values ​​of adjacent unbinded pixels are added to this weighted error to complete the error diffusion at that point. This process is repeated for all pixels in the Fourier base pattern to complete the error diffusion. The error diffusion is performed point-by-point, row by row, from the top-left pixel to the bottom-right pixel. Depending on whether the Fourier base pattern is flipped horizontally, it is rotated counterclockwise by p degrees, and then error diffusion is performed. Finally, the Fourier base pattern is adjusted back to its original orientation. By achieving different pixel scanning orders, the error diffusion order can be changed, resulting in m sets of Fourier base patterns. Here, p = 0°, 90°, 180°, 270°.

[0065] Specifically, using an error diffusion template, the pixel value at a given point is quantized to the smallest integer closest to it. In this embodiment, the quantization is either 0 or 1. The error between the original value and the quantized value is distributed to the unbinded pixels near that pixel according to the weight of the error diffusion template, and then superimposed on the corresponding pixel value. In this embodiment, since error diffusion is performed row by row from the top-left pixel to the bottom-right pixel, the scanning order of the pixels can be changed by flipping or rotating the Fourier basis pattern, thus achieving error diffusion in different directions and obtaining multiple sets of binary Fourier basis patterns. In this embodiment, two sets of binary Fourier basis patterns are mainly used. The two error diffusion directions used in some specific embodiments are as follows: Figure 2 As shown. In other embodiments, multiple sets of binarized Fourier basis patterns may also be used.

[0066] S200. Project the m sets of binary Fourier basis patterns respectively to reconstruct the corresponding m two-dimensional target images;

[0067] It should be noted that in some embodiments, the target scene is obtained by projecting m sets of binary Fourier basis patterns through a spatial light modulator; the modulated voltage signal of the target scene is collected by a single-pixel detector; the Fourier spectrum of the target scene is obtained by solving the phase shift algorithm on the modulated voltage signal; and the two-dimensional target image is obtained by inverse transforming the Fourier spectrum; wherein, the number of modulated voltage signals, Fourier spectra and two-dimensional target images are all m.

[0068] Specifically, the target scene is modulated by projecting m sets of binary Fourier basis patterns through a spatial light modulator. The m sets of modulated voltage signals are collected by a single-pixel detector to obtain the voltage value of each Fourier basis pattern in the m sets. The Fourier spectrum of the m target images is obtained by using an N (N≥2)-step phase shift algorithm. The m two-dimensional target images can be obtained by performing an inverse transform on the Fourier spectrum.

[0069] Specifically, passive lighting is used for projection processing. In other embodiments, active lighting can also be used. The aforementioned binary Fourier basis pattern is displayed using a spatial light modulator. In some embodiments, a DMD is selected as the spatial light modulator. In addition, other spatial light modulators can be used in other embodiments. The DMD displays the binary Fourier basis pattern to modulate the target scene, resulting in a modulated target scene. In some specific embodiments, the scene is modulated using two sets of binary Fourier basis patterns in the S100 specific embodiment. In addition, M (M≥3) sets of binary Fourier basis patterns can also be used in the embodiments. Then, the reflected light on the DMD is collected using a single-pixel detector to obtain a modulated voltage signal. In some embodiments, the single-pixel detector is a PMT. In addition, other single-pixel detectors can be used in other embodiments. The voltage signal is processed to obtain the Fourier spectrum of the target scene. The Fourier spectrum is inversely transformed to obtain the original target image of the target scene (i.e., a two-dimensional target image, wherein, in some specific embodiments, the two two-dimensional target images obtained are as follows). Figure 3 and 4 (As shown). In some embodiments, the voltage value (i.e., the modulated voltage signal) of each binarized Fourier basis pattern is obtained by quantizing the voltage signal. The Fourier coefficients of the corresponding image are obtained using a three-step phase-shift algorithm based on the voltage value. The Fourier coefficients are then filled into the spectrum to obtain the Fourier spectrum. An inverse transform is performed on the Fourier spectrum to obtain the original target image. In other embodiments, the Fourier spectrum can also be obtained using an N (N≥2)-step phase-shift algorithm, but the corresponding N-step phase-shifted Fourier basis pattern must be projected.

[0070] In some alternative embodiments, S200 can be implemented through the following steps:

[0071] S210. The binarized Fourier basis pattern is transmitted to the memory of the DMD module using a data interface. In this embodiment, it is transmitted to the RAM of the DMD. In this embodiment, other storage media may also be used. The DMD control program reads the binarized Fourier basis pattern and displays it at a certain rate. In this embodiment, the refresh rate of the DMD is 2000 Hz. In this embodiment, the refresh rate of the DMD may also be any refresh rate not exceeding its maximum refresh rate. The target scene is imaged on the DMD, realizing the modulation of the target scene.

[0072] S220: The DMD reflects structured light containing target scene information. This reflected light is collected by a single-pixel detector (PMT). The PMT is connected to a signal acquisition card, which samples the signal to obtain continuous voltage signals. The number of sampled values ​​for each binarized Fourier substrate pattern is obtained by calculating the ratio of the data acquisition card's sampling rate to the DMD's projection rate. In this embodiment, the data acquisition card's sampling rate is 500,000 Hz, and the DMD's projection rate is 2,000 Hz, resulting in 250 sampling points for each binarized Fourier substrate pattern. In this embodiment, the rates of the acquisition card and the DMD can be different. Each sampled point of the binarized Fourier substrate pattern is processed to obtain a voltage value. In this embodiment, the values ​​of the 250 sampled points are averaged to obtain a voltage value. Other processing methods can also be selected in this embodiment. This process is repeated to obtain the voltage value for each binarized Fourier substrate pattern.

[0073] S230. In this embodiment, the obtained voltage value is used to calculate the Fourier spectrum using a three-step phase-shift algorithm. An inverse transform is then performed on the Fourier spectrum to obtain the original target image. Alternatively, an N (N≥2)-step phase-shift algorithm can be used to obtain the Fourier coefficients, but this requires projecting a Fourier basis pattern corresponding to the N-step phase shift. The formula for obtaining the Fourier coefficients using the N-step phase-shift algorithm is shown below:

[0074]

[0075] Where N is the number of phase shift steps; f x f y These refer to the spatial frequencies corresponding to the x and y directions, respectively; D * This refers to the light response value caused by background illumination at the position of a single pixel detector, i.e., the modulation voltage signal; C(f x ,f y ) represents the Fourier coefficients; j is the imaginary unit.

[0076] The obtained Fourier coefficients are filled into the Fourier spectrum, and an inverse transform is performed on the obtained Fourier spectrum to obtain the corresponding original target image. The formula for the inverse transform is:

[0077] I(x,y)=F -1 {C(f x ,f y )}

[0078] Where I(x,y) represents the original target image, i.e., the two-dimensional target image; F -1 Represents the inverse Fourier transform; C(f) x ,f y ) represents the Fourier coefficients, and the Fourier spectrum is obtained by filling in the Fourier coefficients.

[0079] S300. Select n two-dimensional target images and perform fusion processing to obtain the target image;

[0080] It should be noted that n is less than or equal to m; in some embodiments, based on n two-dimensional target images, one of them is determined as the reference image and the rest are images to be registered; according to the reference image, the translation amount of the image to be registered relative to the reference image is obtained; according to the translation amount, the n two-dimensional target images are registered; the registered n two-dimensional target images are fused to obtain the target image.

[0081] The process begins with registering the two-dimensional target images. This is done by selecting one of the n two-dimensional target images as a reference, calculating the translation of the other two-dimensional target images relative to this image, and then performing corresponding translations on the other two-dimensional target images to achieve image registration. Finally, a fusion algorithm is used to fuse the n images, thereby obtaining a single target image with full resolution, high quality, and a wide field of view.

[0082] Specifically, the obtained original target images (i.e., the registered two-dimensional target images) are fused to obtain a full-resolution, full-field-of-view, high-quality target image. In some embodiments, two of the above-mentioned original target images are fused, wherein, in some specific embodiments, based on Figure 3 and 4 The target image obtained by fusing two-dimensional target images is as follows: Figure 5 As shown. In other embodiments, P (P≥2) original target images can also be used for fusion. In some embodiments, the two original target images are first registered, using one as a reference, the translation amount of the other image relative to the first image is calculated, and then the non-reference image is translated accordingly to achieve registration of the two images. After registration, the two images are fused using a fusion algorithm. In some embodiments, an iterative reweighted minimization fusion algorithm is used. In other embodiments, other fusion algorithms can also be used. Finally, a full-resolution, full-field-of-view, and high-quality target scene image is obtained.

[0083] In some alternative embodiments, S300 can be implemented through the following steps:

[0084] S310. In this embodiment, an iterative reweighted minimization fusion algorithm is selected to fuse the original target images. Other fusion algorithms may also be used in this embodiment. First, a zero-mean normalization vector is calculated using one image as a reference image. Construct an image pyramid. Then downsample the image to create a pyramid with a size only slightly smaller than the original. Figure 1A low-resolution image (the same operation is performed on the reference image and the original image). The correlation between the reference image and the other image is calculated n times (n can be equal to 30) until it is maximized. The translation amount output at this point is the translation amount required for registration. The non-reference image is then translated according to this translation amount to achieve registration of the two images.

[0085] S320. In this embodiment, the two registered images are fused. As shown in the following formula, the weight estimation parameter β is calculated. t and image prior weight α t It can reconstruct the final target image.

[0086]

[0087] Where x represents the target image, t represents the number of iterations, and x t Let p() represent the target image after t iterations, p() represent the probability, y represent the two-dimensional target image, and β represent the probability. t α represents the weight estimation parameter. t Let F represent the prior weights of the image, and let F represent the total energy function.

[0088] In addition, some embodiments reconstruct the obtained images using full sampling. However, due to the sparsity of natural images, undersampling reconstruction can be performed. Although this will result in the loss of some high-frequency information and the introduction of ringing artifacts, it can still achieve good reconstruction results while preserving most of the details and improving reconstruction efficiency.

[0089] Specifically, in the embodiments, undersampling can greatly shorten the data acquisition time and simplify the calculation process of solving the Fourier coefficients. This allows for a perfect combination of fast Fourier single-pixel imaging and the method of this invention, enabling the rapid acquisition of high-quality reconstructed images while preserving basic image details and without excessive loss of image resolution. This promotes the further development and application of Fourier single-pixel imaging technology and even single-pixel imaging.

[0090] In summary, this invention significantly reduces quantization noise and environmental noise in images while fully utilizing the resolution of the spatial light modulator, and at the same time better preserves high-frequency information, resulting in a substantial improvement in imaging quality. Furthermore, the technical solution of this invention can greatly shorten the time required for image reconstruction through undersampling, while retaining most of the image details and high resolution, thus improving imaging efficiency. Compared with traditional methods, the technology of this application can achieve full-resolution, full-field-of-view, and high-quality binarized Fourier single-pixel imaging.

[0091] On the other hand, refer to Figure 6An embodiment of the present invention provides a binarized Fourier single-pixel imaging system 400, comprising: a first module 410, used to perform error diffusion on a Fourier substrate pattern based on an error diffusion template and through m pixel scanning sequences to obtain m sets of binarized Fourier substrate patterns; a second module 420, used to perform projection processing on the m sets of binarized Fourier substrate patterns respectively to reconstruct m corresponding two-dimensional target images; and a third module 430, used to select n of the two-dimensional target images for fusion processing to obtain a target image; wherein n is less than or equal to m.

[0092] The content of the method embodiments of the present invention is applicable to the system embodiments. The specific functions implemented in the system embodiments are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above methods.

[0093] Reference Figure 7 Another aspect of the present invention provides a binarized Fourier single-pixel imaging device 500, including a processor 510 and a memory 520.

[0094] The memory is used to store programs;

[0095] The processor executes the program to implement the method described above.

[0096] The content of the method embodiments of the present invention is applicable to the embodiments of the present electronic device. The specific functions implemented by the embodiments of the present electronic device 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 methods.

[0097] Another aspect of this invention provides a computer-readable storage medium storing a program that is executed by a processor to implement the methods described above.

[0098] The content of the method embodiments of the present invention is applicable to the computer-readable storage medium embodiments. The specific functions implemented by the computer-readable storage medium 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 methods.

[0099] This invention also discloses a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device can read the computer instructions from the computer-readable storage medium and execute the computer instructions, causing the computer device to perform the aforementioned method.

[0100] In some alternative embodiments, the functions / operations mentioned in the block diagrams may not occur in the order shown in the operation diagrams. For example, depending on the functions / operations involved, two consecutively shown blocks may actually be executed substantially simultaneously, or the blocks may sometimes be executed in reverse order. Furthermore, the embodiments presented and described in the flowcharts of this invention are provided by way of example to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is altered and sub-operations described as part of a larger operation are executed independently.

[0101] Furthermore, although the invention has been described in the context of functional modules, it should be understood that, unless otherwise stated, one or more of the described functions and / or features may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the invention. Rather, given the properties, functions, and internal relationships of the various functional modules in the apparatus disclosed herein, the actual implementation of the module will be understood within the scope of conventional skill of an engineer. Therefore, those skilled in the art can implement the invention as set forth in the claims using ordinary techniques without excessive experimentation. It is also understood that the specific concepts disclosed are merely illustrative and not intended to limit the scope of the invention, which is determined by the full scope of the appended claims and their equivalents.

[0102] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0103] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution means, apparatus, or device (such as a computer-based device, a processor-including device, or other means that can fetch and execute instructions from, or in conjunction with, an instruction execution means, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution means, apparatus, or device.

[0104] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0105] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution device. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0106] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0107] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

[0108] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of the present invention.

Claims

1. A binary Fourier single-pixel imaging method, characterized in that, The method comprises the following steps: error diffusion based on a template, error diffusion of a Fourier basis pattern through m pixel scanning sequences, to obtain m sets of binary Fourier basis patterns; projecting and processing m sets of the binary Fourier basis patterns respectively, to reconstruct m corresponding two-dimensional target images; selecting n two-dimensional target images for fusion processing, to obtain a target image; wherein n is less than or equal to m; wherein the error diffusion based on a template, error diffusion of a Fourier basis pattern through m pixel scanning sequences, to obtain m sets of binary Fourier basis patterns, comprises: quantizing pixel values of each pixel point of the Fourier basis pattern, to obtain quantization errors; based on the quantization errors, weight distribution of non-binary pixels of the pixel points based on an error diffusion template, to complete error diffusion of the pixel points; based on error diffusion of all the pixel points, completing error diffusion of the Fourier basis pattern, to obtain a set of binary Fourier basis patterns; flipping and / or rotating the Fourier basis pattern, and then returning to the step of quantizing pixel values of each pixel point of the Fourier basis pattern, to obtain quantization errors, until m sets of binary Fourier basis patterns are obtained. 2.The binary Fourier on-pixel imaging method of claim 1, wherein, the projecting and processing m sets of the binary Fourier basis patterns respectively, to reconstruct m corresponding two-dimensional target images, comprises: projecting and processing m sets of the binary Fourier basis patterns respectively through a spatial light modulator, to obtain a modulated target scene; collecting modulated voltage signals of the target scene through a single-pixel detector; solving the modulated voltage signals through a phase shift algorithm, to obtain a Fourier spectrum of the target scene; inverse transforming the Fourier spectrum, to obtain a two-dimensional target image; wherein the number of the modulated voltage signals, the Fourier spectrum and the two-dimensional target image is m.

3. The binary Fourier compressive imaging method according to claim 2, wherein, the solving the modulated voltage signals through a phase shift algorithm, to obtain a Fourier spectrum of the target scene, comprises: solving the modulated voltage signals through an N-step phase shift algorithm, to obtain Fourier coefficients; based on the Fourier coefficients, filling to obtain a Fourier spectrum of the target scene; wherein N is greater than or equal to 2, and the formula of the N-step phase shift algorithm is: wherein denotes the Fourier coefficient, denotes the number of phase shifts, denotes the modulated voltage signal, denotes the spatial frequency in the x-direction, denotes the spatial frequency in the y-direction, denotes the imaginary unit.

4. The binary Fourier compressive imaging method according to claim 2, wherein, the inverse transforming the Fourier spectrum, to obtain a two-dimensional target image, comprises: inverse transforming the Fourier spectrum based on an inverse transform formula, to obtain a two-dimensional target image; wherein the inverse transform formula is: wherein denotes a two-dimensional target image; denotes an inverse Fourier transform; denotes a Fourier coefficient, the Fourier spectrum being filled by the Fourier coefficients.

5. The binary Fourier compressive imaging method according to claim 1, wherein, the selecting n two-dimensional target images for fusion processing, to obtain a target image, comprises: based on n two-dimensional target images, determining one of them as a reference image, and the rest as to-be-registered images; according to the reference image, obtaining a translation amount of the to-be-registered images relative to the reference image; according to the translation amount, registering n two-dimensional target images; fusing the registered n two-dimensional target images, to obtain a target image.

6. The binary Fourier compressive imaging method according to claim 5, wherein, the fusing the registered n two-dimensional target images, to obtain a target image, comprises: Based on the fusion formula, the n target images are fused to obtain a target image by combining the weight estimation parameter and the image prior weight. The fusion formula is: wherein denotes the target image, denotes the iteration number, denotes the iteration the target image after the iteration, denotes the probability, denotes the two-dimensional target image, denotes the weight estimation parameter, denotes the image prior weight, denotes the total energy function.

7. A binary Fourier single-pixel imaging system, characterized in that, The system is applied to the binary Fourier single-pixel imaging method of claim 1, and the system comprises: A first module is configured to perform error diffusion on the Fourier base pattern by using m pixel scanning sequences based on an error diffusion template to obtain m groups of binary Fourier base patterns. A second module is configured to perform projection processing on the m groups of binary Fourier base patterns to reconstruct m target images. A third module is configured to select n target images for fusion processing to obtain a target image, wherein n is less than or equal to m.

8. A binary Fourier single-pixel imaging device, comprising a processor and a memory. The memory is configured to store a program. The processor executes the program to implement the method of any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The storage medium stores a program, and the program is executed by the processor to implement the method of any one of claims 1 to 6.

Images