Image reconstruction method and image reconstruction device based on single pixel imaging

Abstract

The present invention provides an image reconstruction method based on single-pixel imaging, comprising: S11: generating a random number matrix and its pseudo-inverse matrix; S12: modulating a light beam based on the random number matrix and its pseudo-inverse matrix to obtain modulated light; S13: collecting the modulated light after it passes through a measured object to obtain a bucket detection signal; and S14: reconstructing an image of the measured object based on the random number matrix, the pseudo-inverse matrix, and the bucket detection signal. The present invention provides a practical reconstruction algorithm for single-pixel imaging, which is time-efficient, resource-efficient, and provides excellent reconstruction results.

Description

Technical Field

[0001] The present disclosure relates to the field of computer imaging technology, and in particular to an image reconstruction method based on single-pixel imaging, an image reconstruction device based on single-pixel imaging, and a computer-readable storage medium. Background Art

[0002] Single-pixel imaging is a method of acquiring two-dimensional images using a single-pixel sensor without spatial resolution. Its core is to use active light modulation to collect reflected light intensity using a single-pixel sensor, obtain global information about the target image, and reconstruct the image based on the correlation between the encoded projection pattern and the light intensity signal. Compared to traditional imaging, single-pixel imaging does not require a spatial array sensor, is highly resistant to interference from scattering media, and has a larger photosensitivity area, higher quantum efficiency, lower dark noise, and faster response speed. This means that single-pixel sensors have significant advantages in non-visible light and low-light energy scenarios.

[0003] In single-pixel imaging, various algorithms are used to reconstruct images using a measurement matrix constructed from random speckle patterns and the measured bucket detection signals. Common reconstruction algorithms include classical correlation calculation, compressed sensing, Hadamard transform-based scanning single-pixel imaging, Fourier transform-based scanning single-pixel imaging, and pseudo-inverse single-pixel imaging.

[0004] Some of the above reconstruction algorithms have poor reconstruction effects, such as the classical correlation calculation algorithm; some reconstructions are time-consuming and resource-intensive, such as the compressed sensing algorithm and the pseudo-inverse single-pixel imaging algorithm.

[0005] The contents of the background technology section merely disclose the technologies known to the inventors and do not necessarily represent the existing technologies in this field. Summary of the Invention

[0006] In view of one or more existing deficiencies, the present invention relates to an image reconstruction method based on single-pixel imaging, comprising:

[0007] S11: Generate a random number matrix and its pseudo-inverse matrix;

[0008] S12: Based on the random number matrix and its pseudo-inverse matrix, modulate the light beam to obtain modulated light;

[0009] S13: Collecting the modulated light after passing through the measured object to obtain the barrel detection signal; and

[0010] S14: Reconstructing an image of the object under test based on the random number matrix, the pseudo-inverse matrix and the bucket detection signal.

[0011] According to one aspect of the present invention, step S12 includes: generating a basis matrix based on the random number matrix and its pseudo-inverse matrix, wherein the column vectors of the basis matrix are approximately orthogonal basis vectors.

[0012] According to one aspect of the present invention, the step S12 further includes: when the active illumination mode is adopted, the light beam is modulated in sequence by using each column vector in the basis matrix and projected onto the object to be measured.

[0013] According to one aspect of the present invention, step S13 includes: performing base scanning data acquisition on the modulated light after being reflected or transmitted by the object to be measured, obtaining the bucket detection signal and converting it into a spectrum matrix of the object to be measured.

[0014] According to one aspect of the present invention, the step S12 further includes: when adopting the passive illumination mode, using each column vector in the basis matrix to sequentially modulate the light beam from the object to be measured.

[0015] According to one aspect of the present invention, step S13 includes: performing base scanning data acquisition on the modulated light obtained after modulating the light beam from the object to be measured, obtaining the bucket detection signal and converting it into a spectrum matrix of the object to be measured.

[0016] According to one aspect of the present invention, step S14 includes: performing an inverse transformation based on the random number matrix, the pseudo-inverse matrix and the spectrum matrix to generate a sampling matrix, and reconstructing a two-dimensional image of the object under test.

[0017] According to one aspect of the present invention, step S13 includes collecting the modulated light after passing through the object to be measured using a single-pixel detector.

[0018] According to one aspect of the present invention, the method further includes: collecting modulated light after passing through the object to be measured from different viewing angles using a plurality of single-pixel detectors, and synthesizing a three-dimensional image of the object to be measured.

[0019] The present invention also relates to an image reconstruction device based on single-pixel imaging, comprising:

[0020] an active lighting module, the active lighting module comprising a light source and a light modulator, the light modulator being configured to modulate a light beam emitted by the light source into modulated light based on a random number matrix and a pseudo-inverse matrix thereof, and project the modulated light onto the object to be measured;

[0021] an acquisition module, the acquisition module comprising at least one single-pixel detector, the at least one single-pixel detector being configured to perform base scanning data acquisition on the modulated light after passing through the object to be measured at at least one viewing angle to obtain a barrel detection signal; and

[0022] An imaging module is coupled to the light modulator and the single-pixel detector and is configured to perform an inverse transformation based on a random number matrix, the pseudo-inverse matrix and the bucket detection signal to generate a sampling matrix and reconstruct a two-dimensional or three-dimensional image of the object being measured.

[0023] According to one aspect of the present invention, the active lighting module further includes:

[0024] The collimating lens group is arranged between the light source and the light modulator, and is used for collimating and correcting the light beam emitted by the light source.

[0025] According to one aspect of the present invention, the active lighting module further includes:

[0026] The first imaging lens group is arranged downstream of the optical path of the light modulator, and is used to image the modulated light onto the object to be measured.

[0027] According to one aspect of the present invention, the acquisition module further includes:

[0028] The second imaging lens group is arranged upstream of the optical path of the single-pixel detector and is used to focus the modulated light after passing through the object to be measured.

[0029] The present invention also relates to a computer-readable storage medium, comprising computer-executable instructions stored thereon, wherein the executable instructions implement the image reconstruction method as described above when executed by a processor.

[0030] This method uses a random matrix and its pseudo-inverse to construct a measurement basis for data acquisition. It then uses this matrix and its pseudo-inverse to reconstruct the object's image from the barrel detection signal. This algorithm is extremely time-efficient and resource-efficient, achieving excellent reconstruction results and providing a practical reconstruction algorithm for single-pixel imaging. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The accompanying drawings, which constitute part of this disclosure, are used to provide a further understanding of the disclosure. The exemplary embodiments of the disclosure and their descriptions are used to explain the disclosure and do not constitute an improper limitation of the disclosure. In the accompanying drawings:

[0032] Figure 1 A flowchart of an image reconstruction method based on single-pixel imaging according to an embodiment of the present invention is shown;

[0033] Figure 2 A schematic diagram of an image reconstruction device using an active illumination mode according to an embodiment of the present invention is shown;

[0034] Figure 3a A schematic diagram of an image reconstruction device using a passive illumination mode according to an embodiment of the present invention is shown;

[0035] Figure 3bA schematic diagram of an image reconstruction device using a passive illumination mode according to another embodiment of the present invention is shown;

[0036] Figure 4 A schematic diagram of image reconstruction based on single-pixel imaging according to an embodiment of the present invention is shown;

[0037] Figure 5 A comparison of an image reconstruction rendering of an embodiment of the present invention and two other imaging algorithms is shown;

[0038] Figure 6 Shown Figure 5 Schematic diagram of the contrast-to-noise ratio of the three imaging algorithms. DETAILED DESCRIPTION

[0039] Hereinafter, only certain exemplary embodiments are briefly described. As will be appreciated by those skilled in the art, the described embodiments may be modified in various ways without departing from the spirit or scope of the present invention. Therefore, the drawings and description are to be considered as illustrative in nature and not restrictive.

[0040] In the description of the present invention, it should be understood that the terms "center," "longitudinal," "transverse," "length," "width," "thickness," "up," "down," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," "outside," "clockwise," "counterclockwise," and the like, indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely for the purpose of facilitating the description of the present invention and simplifying the description. They do not indicate or imply that the devices or components referred to must have a specific orientation, be constructed, or operate in a specific orientation, and therefore should not be construed as limiting the present invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features indicated. Thus, features defined as "first" or "second" may explicitly or implicitly include one or more of the aforementioned features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0041] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood broadly. For example, they may refer to fixed, removable, or integral connections; mechanical, electrical, or intercommunication connections; direct or indirect connections through an intermediary; and internal communication between two components or interaction between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0042] In the present invention, unless otherwise expressly specified or limited, a first feature being "above" or "below" a second feature may include the first and second features being in direct contact, or may also include the first and second features not being in direct contact but being in contact via another feature between them. Furthermore, a first feature being "above," "above," and "above" a second feature may include the first feature being directly above or diagonally above the second feature, or may simply mean that the first feature is at a higher level than the second feature. A first feature being "below," "below," and "below" a second feature may include the first feature being directly above or diagonally above the second feature, or may simply mean that the first feature is at a lower level than the second feature.

[0043] The disclosure below provides many different embodiments or examples for realizing different structures of the present invention. In order to simplify the disclosure of the present invention, the components and settings of specific examples are described below. Of course, they are merely examples and are not intended to limit the present invention. In addition, the present invention may repeat reference numbers and / or reference letters in different examples. Such repetition is for the purpose of simplicity and clarity and does not in itself indicate the relationship between the various embodiments and / or settings discussed. In addition, the present invention provides examples of various specific processes and materials, but those skilled in the art will recognize the application of other processes and / or the use of other materials.

[0044] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0045] Figure 1 A flowchart of an image reconstruction method based on single-pixel imaging according to an embodiment of the present invention is shown. The image reconstruction method 10 includes:

[0046] In step S11, a random number matrix and its pseudo-inverse matrix are generated as the measurement matrix. The pseudo-inverse matrix is ​​a generalized form of the inverse matrix. Since there is no inverse matrix for a singular matrix or a non-square matrix, its pseudo-inverse matrix can be calculated using a function in MATLAB. For example, if the matrix M is an n×m random number matrix, n≠m, then the pseudo-inverse matrix M is -1 is an m×n matrix.

[0047] In step S12, based on the random number matrix M and its pseudo-inverse matrix M -1 , modulate the light beam to obtain modulated light. Figure 2The schematic diagram of an image reconstruction device using an active illumination mode according to one embodiment of the present invention is shown. The image reconstruction device 20 includes a light source 211 and a light modulator 212. A light beam emitted by the light source 211 is input to the light modulator 212. The light modulator 212 modulates a parameter of the light beam, such as the light intensity, based on a measurement matrix, and outputs modulated light. The image reconstruction device 20 will be described in further detail in the following sections.

[0048] According to a preferred embodiment of the present invention, step S12 includes: based on the random number matrix M and its pseudo-inverse matrix M -1 Generate a basis matrix, the column vectors of which are approximately orthogonal basis vectors. For example, use a computer to generate a random number matrix M, which is an n×m matrix; then calculate the pseudo-inverse matrix M of the random number matrix M -1 , is an m×n matrix. Consider the following formula for two-dimensional transformation:

[0049]

[0050] Wherein, F represents the sampling matrix of the object under test, that is, the matrix representing the digital image of the object under test, which is an n×n matrix; The spectrum matrix represents the object under test, which is an n×n matrix. Formula (1) shows the sampling matrix F and the spectrum matrix The conversion relationship of , formula (1) can be expanded as follows:

[0051]

[0052] Introducing the above transformation basis matrix sequence

[0053] {P uv |u, v=0, 1, ..., m-1} (3)

[0054] in,

[0055] P uv (x, y) = M -1 (u,x)M(y,v) (4)

[0056] If the vth column vector of the random number matrix M is denoted as M v , pseudo-inverse matrix M -1 The u-th column vector of Then the basis matrix can also be expressed as:

[0057]

[0058] This is the random number matrix M and its pseudo-inverse matrix M -1 The approximately orthogonal basis vectors constructed are the column vectors in the measurement matrix.

[0059] According to whether the light emitted by the light source is modulated before being projected onto the object to be measured, the imaging system is divided into an active illumination mode and a passive illumination mode. Figure 2 FIG. shows the schematic diagram of an image reconstruction device adopting the active illumination mode according to an embodiment of the present invention. When adopting the active illumination mode, the optical modulator 212 is located between the light source 211 and the object to be measured. The light beam emitted by the light source 211 is first modulated by the optical modulator 212 and then projected onto the object to be measured, and the object to be measured is a flat plate with a highly reflective Chinese character "light". Figure 3a FIG. shows the schematic diagram of an image reconstruction device adopting the passive illumination mode according to an embodiment of the present invention. The image reconstruction device 30 includes an optical modulator 312 and a collection module 32. When adopting the passive illumination mode, the image reconstruction device 30 itself may not have a light source. The optical modulator 312 is located between the object to be measured and the collection module 32, and mainly relies on the natural light reflected by the object to be measured or its own radiation to form an image on the imaging sensor, where the object to be measured is a flat plate with a highly reflective Chinese character "light". Figure 3b FIG. shows the schematic diagram of an image reconstruction device adopting the passive illumination mode according to another embodiment of the present invention. Different from Figure 3a that: the object to be measured is a flat plate with a hollow Chinese character "light", the light source is arranged behind the object to be measured, and the optical modulator 312 receives the light beam transmitted through the object to be measured, and after modulation, it is collected by the collection module 32. The image reconstruction device 20 and the image reconstruction device 30 will be further described in the following paragraphs.

[0060] According to a preferred embodiment of the present invention, the step S12 further includes: when adopting the active illumination mode, using each column vector in the basis matrix to modulate the light beam in sequence and project it onto the object to be measured.

[0061] According to a preferred embodiment of the present invention, the step S12 further includes: when adopting the passive illumination mode, using each column vector in the basis matrix to modulate the light beam from the object to be measured in sequence. Preferably, when the light beam from the object to be measured is too weak to affect the imaging quality, a light source 33 can be added, and the light beam emitted by the light source 33 is projected onto the object to be measured. After the optical modulator 312 receives the light beam reflected or transmitted by the object to be measured, it modulates the light beam in sequence based on each column vector in the basis matrix.

[0062] In step S13, the modulated light after passing through the object to be measured is collected to obtain the bucket detection signal. In step S12, the measurement matrix is first obtained, and then the optical modulator modulates the received light beam based on this measurement matrix, and the generated modulated light is detected by the detector and collected, which is the bucket detection signal.

[0063] According to a preferred embodiment of the present invention, step S13 includes collecting modulated light after passing through the object to be measured through a single-pixel detector. The single-pixel detector has only one photosensitive detector, and its acquisition signal-to-noise ratio is relatively high. The single-pixel detector also has a wider spectral response range and can reduce data acquisition, transmission and storage requirements based on compressed sensing theory. Therefore, it has incomparable advantages over traditional array sensors and has a wide range of applications. The single-pixel detector is, for example, a bucket detector (Bucket Detector) without spatial resolution, which is used to receive scattered light in all directions. The bucket detection signal itself is a column vector of length n×n, which becomes an n×n square matrix after deformation, that is, the spectral matrix of the object.

[0064] According to a preferred embodiment of the present invention, step S13 includes: performing base scanning data acquisition on the modulated light after being reflected or transmitted by the object to be measured, obtaining the barrel detection signal and converting it into a spectrum matrix of the object to be measured When the active illumination mode is used, the modulated light is reflected or transmitted by the object to be measured and then collected by the single pixel detector 221. The barrel detection signal is first obtained and then converted into a spectrum matrix.

[0065] According to a preferred embodiment of the present invention, step S13 includes: performing base scanning data acquisition on the modulated light obtained after modulating the light beam from the object to be measured, obtaining the bucket detection signal and converting it into a spectrum matrix of the object to be measured. When the passive illumination mode is adopted, the light beam from the object to be measured is modulated by the light modulator 312 to generate modulated light, for example, collected by the single pixel detector 314, first obtaining the bucket detection signal, and then converting it into a spectrum matrix

[0066] In step S14, based on the random number matrix M, the pseudo inverse matrix M -1 and the barrel detection signal to reconstruct the image of the object being measured.

[0067] According to a preferred embodiment of the present invention, step S14 includes: -1 and the spectrum matrix Perform inverse transformation to generate sampling matrix F and reconstruct the two-dimensional image of the object under test. The inverse transformation is performed using formula (1) as follows:

[0068]

[0069] Based on random number matrix M and its pseudo-inverse matrix M -1 , and the spectrum matrix of the object under test obtained after the barrel detection signal is deformed Then use formula (6) to reconstruct the image of the object.

[0070] In summary, the image reconstruction method 10 includes: using a random number matrix M and its pseudo-inverse matrix M -1 A measurement matrix is ​​generated. In active illumination mode, the light beam from light source 211 is modulated based on the measurement matrix. The resulting modulated light, after being reflected or transmitted by the object under test, carries the two-dimensional image information of the object under test. Alternatively, in passive illumination mode, the light beam from the object under test is modulated based on the measurement matrix. The resulting modulated light also carries the two-dimensional image information of the object under test. Finally, the modulated light is collected and an inverse transformation is performed based on the measurement matrix to reconstruct the two-dimensional image of the object under test.

[0071] According to a preferred embodiment of the present invention, the image reconstruction method 10 also includes collecting modulated light after passing through the object to be measured from different perspectives through multiple single-pixel detectors, and synthesizing a three-dimensional image of the object to be measured. Taking the active illumination mode as an example, multiple single-pixel detectors 221 are placed in different spatial positions to collect modulated light reflected or transmitted by the object to be measured from different perspectives, and the barrel detection signals of different perspectives are sequentially inversely transformed based on the measurement matrix, and finally a three-dimensional image of the object to be measured is synthesized. Optionally, a unidirectional pixel detector can also be used to collect modulated light from different perspectives in turn, and finally merge and reconstruct a three-dimensional image of the object to be measured. The hardware configuration of the passive illumination mode is also applicable to this embodiment and will not be repeated here. Among them, the algorithm involved in synthesizing three-dimensional images can refer to existing technologies.

[0072] The present invention also relates to an image reconstruction device based on single pixel imaging, referring to Figure 2 As shown, the image reconstruction device 20 adopts an active illumination mode, and includes an active illumination module 21 , an acquisition module 22 and an imaging module 23 .

[0073] The active lighting module 21 includes a light source 211 and a light modulator 212. The light modulator 212 is configured based on a random number matrix M and its pseudo-inverse matrix M. -1 The light beam emitted by the light source 211 is modulated into a modulated light, and the modulated light is projected onto the object to be measured. In other words, the light source 211 is used to emit an illumination light beam, and the light modulator 212 is used to generate a multi-encoded random number matrix M and its pseudo-inverse matrix M -1 , and then modulate the illumination beam.

[0074] The acquisition module 22 includes at least one single-pixel detector 221, which is configured to perform basic scanning data acquisition on the modulated light after it passes through the object under test, from at least one viewing angle, to obtain a bucket detection signal. In other words, the single-pixel detector 221 acquires and converts the modulated light carrying the image information of the object under test to obtain a bucket detection signal.

[0075] The imaging module 23 is coupled to the light modulator 212 and the single pixel detector 221 and is configured based on the random number matrix M, the pseudo inverse matrix M -1 The acquisition module 22 acquires the barrel detection signal and performs an inverse transformation to generate a sampling matrix F, thereby reconstructing an image of the object under test. In other words, the imaging module 23 performs an inverse transformation on the barrel detection signal acquired by the acquisition module 22 according to the reconstruction principle of coded imaging to reconstruct a two-dimensional or three-dimensional image of the object under test. When the acquisition module 22 includes a single-pixel detector, the modulated light is acquired and converted from only one perspective, and the imaging module 23 can reconstruct a two-dimensional image of the object under test. When the acquisition module 22 includes multiple single-pixel detectors, the modulated light can be acquired and converted from multiple perspectives to obtain multiple barrel detection signals. The imaging module 23 performs an inverse transformation on the multiple barrel detection signals and combines them to reconstruct a three-dimensional image of the object under test.

[0076] According to a preferred embodiment of the present invention, Figure 2 As shown, the active lighting module 21 further includes: a collimating lens group 213, which is arranged between the light source 211 and the light modulator 212, and is used to collimate and correct the light beam emitted by the light source 211, and is used to adjust the light beam to coincide with the target surface of the light modulator 212.

[0077] According to a preferred embodiment of the present invention, the active illumination module 21 further includes a first imaging lens group 214 disposed downstream of the optical path of the light modulator 212 for imaging the modulated light emitted by the light modulator 212 onto the object under test.

[0078] According to a preferred embodiment of the present invention, the acquisition module 22 also includes a second imaging lens group 222, which is arranged upstream of the optical path of the single-pixel detector 221, and is used to focus the modulated light after passing through the object to be measured, so as to increase the range of the detector receiving the light beam and image the received modulated light onto the single-pixel detector 221.

[0079] The present invention also relates to an image reconstruction device based on single pixel imaging, referring to Figure 3a and Figure 3b The image reconstruction device 30 adopts a passive illumination mode and includes an acquisition module 31 and an imaging module 32. The acquisition module 31 includes a collimating lens group 311, a light modulator 312, an imaging lens group 313 and a single pixel detector 314.

[0080] The collimating lens group 311 is configured to converge and collimate the light rays from the object to be measured; the optical modulator 312 modulates the collimated light beam based on the random number matrix and the pseudo-inverse matrix, and the imaging lens group 313 is used to image the modulated light onto the single-pixel detector 314. The single-pixel detector 314 receives the modulated light to generate a bucket detection signal and sends it to the imaging module 32. The imaging module 32 performs an inverse transformation based on the bucket detection signal to reconstruct the image of the object to be measured. Preferably, in order to increase the light rays reflected by the object to be measured, the image reconstruction device 30 may further include a light source 33, such as Figure 3a the light beam emitted by the light source 33 in Figure 3a irradiates the object to be measured, or for another example Figure 3b the light beam emitted by the light source 33 in Figure 3b is transmitted after passing through the object to be measured. The acquisition module 31 can receive the light beam reflected or transmitted by the object to be measured to improve the imaging quality.

[0081] The above methods and devices for image reconstruction based on the active illumination mode and the passive illumination mode are introduced. The following will be further described through embodiments.

[0082] Figure 4 FIG. shows a schematic diagram of image reconstruction based on single-pixel imaging according to an embodiment of the present invention. The active illumination mode is adopted. The object to be measured is, for example, a flat plate with a highly reflective Chinese character "light". First, a random number matrix and its pseudo-inverse matrix are generated, and the basis matrix is obtained through formulas (1)-(5). Each column vector in the basis matrix is used to modulate the intensity of the light beam in turn and project it onto the flat plate with the highly reflective Chinese character "light". The bucket detector is placed in front of the flat plate. Refer to Figure 2 and receive the modulated light reflected by the flat plate to generate a bucket detection signal, and then obtain the spectral matrix. Finally, the image of the flat plate is reconstructed through formula (6) and the previously generated random number matrix and its pseudo-inverse matrix. Among them, after modulating the intensity of the light beam to generate the encoded modulated light, when imaging onto the flat plate, it is equivalent to being modulated again by the highly reflective Chinese character on the flat plate. Therefore, the information of the Chinese character is carried in the reflected modulated light. After the bucket detector receives it, the image of the highly reflective Chinese character can be restored through inverse transformation.

[0083] Figure 5 FIG. shows a comparison of the image reconstruction effect diagrams of an embodiment of the present invention and the image reconstruction effect diagrams of two other imaging algorithms. The object to be measured is a flat plate with a highly reflective Chinese character "light". Image reconstruction is performed based on three single-pixel imaging algorithms respectively. Among them, the first row is the image reconstruction diagram of the single-pixel imaging algorithm of classical correlation calculation; the second row is the image reconstruction diagram of the single-pixel imaging algorithm based on the random matrix and its pseudo-inverse matrix according to an embodiment of the present invention; the third row is the image reconstruction diagram of the single-pixel imaging algorithm of Hadamard transform-based scanning; the first column to the fifth column respectively correspond to the image acquisition compression ratios of 20%, 40%, 60%, 80%, and 100%.

[0084] According to the comparison of the effects of the three imaging algorithms, it can be seen that the reconstruction effect of the classical association calculation algorithm in the first row is poor, and the pixel grayscale values ​​under different image acquisition compression ratios are relatively close, and the clarity is poor; the reconstruction effect of the single-pixel imaging algorithm of the Hadamard transform basis scan in the third row is good, and the clarity is high; the second row is the reconstruction effect based on the random matrix and its pseudo-inverse matrix of the present invention, and the image clarity is significantly improved with the increase of the compression ratio. The image clarity at a low compression ratio of 20% is slightly worse than the reconstruction effect of the classical association calculation algorithm, but the image clarity at a high compression ratio of 100% can reach the effect of the Hadamard transform imaging algorithm.

[0085] Figure 6 Shown Figure 5 Schematic diagram of the contrast-to-noise ratio of the three imaging algorithms. The horizontal axis is the image acquisition compression ratio, and the vertical axis is the contrast-to-noise ratio. The curve of the classical correlation calculation algorithm is roughly distributed in the range of 1.0-2.0 as the image acquisition compression ratio increases; the curve of the imaging algorithm based on the Hadamard transform basis scan is roughly distributed in the range of 2.5-3.0 as the image acquisition compression ratio increases; the imaging algorithm based on the random matrix and its pseudo-inverse matrix proposed in the present invention has a curve with a larger contrast-to-noise ratio range as the image acquisition compression ratio increases. When the compression ratio is lower than 50%, the contrast-to-noise ratio is lower than that of the classical correlation calculation algorithm; when the compression ratio is greater than 50%, the contrast-to-noise ratio is between the classical correlation calculation algorithm and the Hadamard transform imaging algorithm. It can be seen that the embodiments of the present invention can have a better contrast-to-noise ratio while ensuring a smaller compression ratio, achieving a balance between resource usage and image reconstruction effect.

[0086] The present invention further relates to a computer-readable storage medium, comprising computer-executable instructions stored thereon, wherein the computer-executable instructions implement the image reconstruction method 10 as described above when executed by a processor.

[0087] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. An image reconstruction method based on single-pixel imaging, comprising: S11: Generate a random number matrix and its pseudo-inverse matrix; S12: Based on the random number matrix and its pseudo-inverse matrix, modulate the light beam to obtain modulated light; S13: Collect the modulated light after passing through the object to be measured and obtain the barrel detection signal; and S14: reconstructing an image of the object under test based on the random number matrix, the pseudo-inverse matrix, and the bucket detection signal; Wherein, the step S12 comprises: generating a basis matrix based on the random number matrix and its pseudo-inverse matrix, wherein the column vectors of the basis matrix are approximately orthogonal basis vectors; When the active illumination mode is adopted, the light beam is modulated in sequence by using each column vector in the basis matrix and projected onto the object to be measured; or When the passive illumination mode is adopted, the light beam from the object to be measured is modulated in sequence using each column vector in the basis matrix.

2. The image reconstruction method according to claim 1, wherein step S13 comprises: The modulated light after being reflected or transmitted by the object to be measured is subjected to base scanning data acquisition, and the barrel detection signal is obtained and converted into a spectrum matrix of the object to be measured.

3. The image reconstruction method according to claim 1, wherein step S13 comprises: The modulated light obtained after modulating the light beam from the object to be measured is subjected to base scanning data acquisition, and the bucket detection signal is obtained and converted into a spectrum matrix of the object to be measured.

4. The image reconstruction method according to claim 2 or 3, wherein the step S14 comprises: Based on the random number matrix, the pseudo-inverse matrix and the frequency spectrum matrix, an inverse transformation is performed to generate a sampling matrix, and a two-dimensional image of the object to be measured is reconstructed. 5 . The image reconstruction method according to claim 1 , wherein step S13 comprises collecting the modulated light after passing through the object by using a single-pixel detector.

6. The image reconstruction method according to claim 5, further comprising: The modulated light passing through the object to be measured is collected from different viewing angles by multiple single-pixel detectors, and a three-dimensional image of the object to be measured is synthesized.

7. An image reconstruction device implementing the image reconstruction method based on single-pixel imaging according to any one of claims 1 to 6, comprising: an active lighting module, the active lighting module comprising a light source and a light modulator, the light modulator being configured to modulate a light beam emitted by the light source into modulated light based on a random number matrix and a pseudo-inverse matrix thereof, and project the modulated light onto the object to be measured; An acquisition module, the acquisition module comprising at least one single-pixel detector, the at least one single-pixel detector being configured to perform base scanning data acquisition on the modulated light after passing through the object to be measured at at least one viewing angle, and obtain a barrel detection signal; and An imaging module is coupled to the light modulator and the single-pixel detector and is configured to perform an inverse transformation based on a random number matrix, the pseudo-inverse matrix and the bucket detection signal to generate a sampling matrix and reconstruct a two-dimensional or three-dimensional image of the object being measured.

8. The image reconstruction device according to claim 7, wherein the active illumination module further comprises: The collimating lens group is arranged between the light source and the light modulator, and is used for collimating and correcting the light beam emitted by the light source.

9. The image reconstruction device according to claim 8, wherein the active illumination module further comprises: The first imaging lens group is arranged downstream of the optical path of the light modulator, and is used to image the modulated light onto the object to be measured.

10. The image reconstruction device according to any one of claims 7 to 9, wherein the acquisition module further comprises: The second imaging lens group is arranged upstream of the optical path of the single-pixel detector and is used to focus the modulated light after passing through the object to be measured.

11. A computer-readable storage medium comprising computer-executable instructions stored thereon, wherein the executable instructions, when executed by a processor, implement the image reconstruction method according to any one of claims 1 to 6.

Images