Single-pixel spectral imaging system and imaging method

Abstract

A single-pixel spectral imaging system and method are disclosed. The single-pixel spectral imaging system includes: an interferometer adapted to interfere with externally input white light to obtain interferometric white light; a sample holder on which a sample to be tested is placed, through which the interferometric white light passes to obtain interferometric light carrying spectral information of the sample to be tested; a spatial light modulator, which obtains at least one flip state based on at least one template of a measurement matrix, and modulates the amplitude of the interferometric light under each flip state to obtain modulated interferometric light; a single-point detector adapted to detect the intensity of the modulated interferometric light corresponding to each template; and a computer, which obtains the spectrum corresponding to each template based on the detected intensity of the modulated interferometric light corresponding to each template, and obtains an image of the sample to be tested based on the spectra of all templates and the measurement matrix, wherein the number of templates is less than the number of pixels in the image of the sample to be tested.

Description

Technical Field

[0001] This invention relates to the field of spectral imaging, and particularly to a single-pixel spectral imaging system and imaging method. Background Technology

[0002] Spectral imaging technology combines spectral and imaging techniques, recording intensity data for each pixel across multiple spectral bands. This technology can divide the imaging radiation bands into narrower bands and simultaneously image them, thus obtaining images of the same analyte across multiple spectral bands. This means simultaneously acquiring both two-dimensional spatial and one-dimensional spectral information of the object, thereby obtaining material information and chemical composition of the sample. Therefore, spectral imaging technology has not only greatly advanced the development of material detection and classification techniques but also has wide applications in fields such as air quality monitoring, water resource control, biomedicine, non-destructive testing, and biofluorescence detection.

[0003] Current spectral imagers generally employ scanning imaging. Based on the scanning method, spectral imaging techniques can be categorized into three types: oscillating broom, pushbroom, and staring. Oscillating broom spectral imaging utilizes a broom device to scan in a direction perpendicular to the spectrometer's movement, acquiring two-dimensional spatial information. However, the scanning process results in a long data acquisition time, and the broom device requires high system stability. Pushbroom spectral imaging eliminates the broom device, directly imaging the one-dimensional spatial image in the field of view onto an area array detector after dispersion, thus achieving both one-dimensional spatial and spectral imaging. Pushbroom spectral imaging is characterized by high stability, but its field of view is limited by the detector size. Staring spectral imaging differs from the previous two techniques, using filters to obtain a single-spectral-dimensional two-dimensional spatial image and acquiring the image through an area array detector, resulting in lower energy efficiency. Resolution is also limited by the filter. Furthermore, scanning spectral imaging generally requires moving parts, increasing system instability, and necessitates expensive area array detectors for data acquisition. Summary of the Invention

[0004] In view of the above problems, the present invention provides a single-pixel spectral imaging system and imaging method.

[0005] A first aspect of the present invention provides a single-pixel spectral imaging system, comprising:

[0006] An interferometer is used to cause externally input white light to interfere and produce interfering white light.

[0007] A sample holder on which a sample to be tested is placed; the interference white light passes through the sample to be tested to obtain interference light carrying the spectral information of the sample to be tested.

[0008] A spatial light modulator obtains at least one flip state based on at least one template of a measurement matrix, and modulates the amplitude of the aforementioned interference light under the action of each flip state to obtain modulated interference light;

[0009] A single-point detector is suitable for detecting the intensity of the modulated interference light corresponding to each template;

[0010] The computer obtains the spectrum corresponding to each template based on the intensity of the modulated interference light corresponding to each template, and obtains the image of the sample to be tested based on the spectra corresponding to all templates and the above measurement matrix. The number of templates is less than the number of pixels in the image of the sample to be tested.

[0011] According to an embodiment of the present invention, the above-described single-pixel spectral imaging system further includes:

[0012] The peripheral component interconnection bus acquisition card is suitable for controlling the start-up of the interferometer, controlling the playback of the templates, and sending sampling analog signals to sample the intensity of the modulated interference light corresponding to each template detected by the single-point detector with equal optical path difference, and sending the sampling results to the computer.

[0013] According to embodiments of the present invention, the interferometer described above is also suitable for interfering externally input laser light to obtain interferometric laser light;

[0014] The aforementioned single-pixel spectral imaging system also includes:

[0015] A laser detector suitable for detecting the intensity of the aforementioned interferometric lasers;

[0016] The zero-crossing comparator circuit is suitable for emitting a digital signal based on the intensity of the aforementioned interference laser. The digital signal is suitable for sampling the intensity of the modulated interference light corresponding to each template detected by the aforementioned single-point detector using equal optical path difference sampling. The digital signal is converted into the aforementioned sampled analog signal via the aforementioned peripheral component interconnect bus acquisition card.

[0017] According to an embodiment of the present invention, the above-described single-pixel spectral imaging system further includes:

[0018] The first focusing lens is suitable for focusing the interference white light emitted by the interferometer.

[0019] Optical fiber is suitable for transmitting focused interference white light to the sample to be tested.

[0020] According to an embodiment of the present invention, the above-described single-pixel spectral imaging system further includes:

[0021] The second focusing lens is suitable for focusing the modulated interference light corresponding to each template output by the above spatial light modulator to obtain focused interference light;

[0022] A collimating lens is used to collimate the focused interference light and input the collimated interference light into the single-point detector.

[0023] According to an embodiment of the present invention, the laser described above is a HeNe laser.

[0024] As a second aspect of the present invention, a single-pixel spectral imaging method is also provided, utilizing the above-described single-pixel spectral imaging system, the method comprising:

[0025] Interferometric white light is obtained by using an interferometer to cause interference with externally input white light;

[0026] The aforementioned interference white light is transmitted through the sample to be tested on the sample holder to obtain interference light carrying the aforementioned spectral information of the sample to be tested;

[0027] Using a spatial light modulator, at least one flip state is obtained based on at least one template of the measurement matrix, and the amplitude of the above interference light is modulated under the action of each flip state to obtain modulated interference light;

[0028] The intensity of the modulated interference light corresponding to each template is detected using a single-point detector;

[0029] Using a computer, the spectrum corresponding to each template is obtained based on the intensity of the modulated interference light corresponding to each template detected, and the image of the sample to be tested is obtained based on the spectra corresponding to all templates and the above measurement matrix, wherein the number of templates is less than the number of pixels in the image of the sample to be tested.

[0030] According to an embodiment of the present invention, the method further includes using a peripheral component interconnect bus acquisition card to sample the intensity of the modulated interference light corresponding to each template detected by the single-point detector with equal optical path difference, and sending the sampling results to the computer.

[0031] According to an embodiment of the present invention, obtaining the spectrum corresponding to each template based on the intensity of the modulated interference light corresponding to each template detected includes:

[0032] The sampling results of the intensity of the modulation interference light corresponding to each template are preprocessed to remove the DC component in the above sampling results, and the preprocessed result is obtained.

[0033] The above preprocessing results are subjected to atopexy to obtain the atopexy result;

[0034] The initial spectrum is obtained by zero-padding and Fourier transforming the above apodization results;

[0035] Phase correction is performed on the initial spectra to obtain the spectra corresponding to each template.

[0036] According to an embodiment of the present invention, obtaining the image of the sample to be tested based on the spectra corresponding to all templates and the above-described measurement matrix includes:

[0037] The energy values ​​of the spectra corresponding to the above measurement matrix and all templates at the target wavelength are input into the image reconstruction algorithm to obtain a constrained objective function.

[0038] The constrained objective function is transformed by using slack variables and the enhanced Lagrange multiplier method to obtain the unconstrained objective function;

[0039] The image of the sample under test at the target wavelength is obtained based on the unconstrained objective function.

[0040] According to an embodiment of the present invention, a single-point detector is used to detect modulated interference light. Compared with traditional area array detectors, no moving parts are required, which simplifies the system structure, improves system stability, and makes the system easier to use and maintain. In addition, the single-point detector is inexpensive, and the number of photons received by the single-point detector is much greater than that of traditional area array detectors, which greatly reduces image distortion caused by dark noise.

[0041] According to embodiments of the present invention, the compressed sensing theory and single-pixel spectral imaging technology are combined for imaging by using a single-point detector to detect the interference light modulated by the spatial light modulator, using a computer to obtain the spectrum corresponding to each template based on the intensity of the modulated interference light corresponding to each template, and obtaining the image of the sample under test based on the spectra of all templates and the measurement matrix (i.e., single-pixel spectral imaging technology). Since the number of templates used is less than the number of pixels in the sample image, the single-pixel spectral imaging system used in this embodiment requires less data acquisition, has a shorter sampling time, and a faster imaging speed.

[0042] According to embodiments of the present invention, conventional single-pixel spectral imaging methods transmit monochromatic light through the sample, thus requiring the light source to be changed to obtain an image of the test object in that light source band. To obtain images of the test object in multiple bands, the light source needs to be switched frequently. In the embodiments of this application, since the light transmitted through the test sample is interference white light, the spectrum corresponding to each template can be obtained based on the interference white light. Since the spectrum corresponding to each template includes the intensity corresponding to multiple wavelengths, the single-pixel spectral imaging method provided by the embodiments of the present invention can obtain images of the test sample at multiple wavelengths. Attached Figure Description

[0043] Figure 1 A block diagram of a single-pixel spectral imaging system provided according to an embodiment of the present invention is shown;

[0044] Figure 2 A flowchart of a single-pixel spectral imaging method according to an embodiment of the present invention is shown;

[0045] Figure 3 A flowchart illustrating the process of obtaining the spectrum corresponding to each template according to an embodiment of the present invention is shown;

[0046] Figure 4 A flowchart illustrating the process of obtaining an image of the sample to be tested based on the spectra corresponding to all templates and the measurement matrix according to an embodiment of the present invention is shown.

[0047] Explanation of reference numerals in the attached figures

[0048] 1 interferometer

[0049] 2 Sample racks

[0050] 3 Spatial light modulator

[0051] 4 single-point detectors

[0052] 5 computers

[0053] 6 Peripheral Component Interconnection Bus Acquisition Card

[0054] 7 laser detectors

[0055] 8-Zero-Crossing Comparator Circuit

[0056] 9 White light source

[0057] 10 laser sources

[0058] 11 First Focusing Lens

[0059] 12 fiber optic cables

[0060] 13 Second Focusing Lens

[0061] 14 collimating lenses Detailed Implementation

[0062] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the invention. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of the invention for ease of explanation. However, it will be apparent that one or more embodiments may be practiced without these specific details. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concept of the invention.

[0063] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the invention. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0064] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0065] When using expressions such as "at least one of A, B, and C", they should generally be interpreted in accordance with the meaning that is commonly understood by a person skilled in the art (e.g., "a system having at least one of A, B, and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B, and C, etc.).

[0066] Figure 1 A block diagram of a single-pixel spectral imaging system provided according to an embodiment of the present invention is shown.

[0067] like Figure 1 As shown, the single-pixel spectral imaging system includes: an interferometer 1, a sample holder 2, a spatial light modulator 3, a single-point detector 4, and a computer 5.

[0068] Interferometer 1 is used to interfere externally input white light to obtain interfering white light. Sample holder 2 holds the sample to be tested, and the interfering white light passes through the sample to obtain interfering light carrying the spectral information of the sample. Spatial light modulator 3 obtains at least one flip state based on at least one template of the measurement matrix, and modulates the amplitude of the interfering light under each flip state to obtain modulated interfering light. Single-point detector 4 is used to detect the intensity of the modulated interfering light corresponding to each template. Computer 5 obtains the spectrum corresponding to each template based on the detected intensity of the modulated interfering light corresponding to each template, and obtains an image of the sample to be tested based on the spectra of all templates and the measurement matrix, wherein the number of templates is less than the number of pixels in the image of the sample to be tested.

[0069] According to an embodiment of the present invention, the interferometer 1 can be, for example, a Michelson interferometer. The interference intensity generated by the Michelson interferometer is related to the optical path difference. The spatial light modulator 3, also known as a digital micromirror device, is a high-resolution, high-sensitivity digital micromirror array (DMD). During the imaging process, the measurement matrix to be displayed is pre-imported into the spatial light modulator 3, and the sample to be tested is placed on the sample holder. The measurement matrix is ​​generated by the computer 5.

[0070] According to an embodiment of the present invention, a single-point detector is used to detect modulated interference light. Compared with traditional area array detectors, no moving parts are required, which simplifies the system structure, improves system stability, and makes the system easier to use and maintain. In addition, the single-point detector is inexpensive, and the number of photons received by the single-point detector is much greater than that of traditional area array detectors, which greatly reduces image distortion caused by dark noise.

[0071] According to an embodiment of the present invention, the above-described pixel spectral imaging system further includes: a peripheral component interconnect bus acquisition card (PCI acquisition card) 6, which is adapted to control the start-up of the interferometer, to control the playback of the template, and to send a sampling analog signal to perform equal optical path difference sampling on the intensity of the modulated interference light corresponding to each template detected by the single-point detector, and to send the sampling result to the computer.

[0072] According to an embodiment of the present invention, the peripheral component interconnection bus acquisition card 6 is a high-speed digital-to-analog conversion data acquisition card.

[0073] According to an embodiment of the present invention, the above-described single-pixel spectral imaging system further includes: a white light source 9 and a laser source 10. The white light source 9 is suitable for emitting white light, and the laser source 10 is suitable for emitting laser light. The laser light source 9 can be, for example, a HeNe laser, and the laser source 10 can be, for example, a HeNe laser source. The HeNe laser light is interfered with by the interferometer 3 to obtain a HeNe interference laser light. The white light source 9 is an incandescent lamp source or a halogen lamp source, and the HeNe laser source light source can emit HeNe laser light with a wavelength of 632.8 nm.

[0074] According to an embodiment of the present invention, the interferometer described above is also suitable for interfering externally input laser light to obtain interferometric laser light. The single-pixel spectral imaging system further includes a laser detector 7 and a zero-crossing comparator circuit 8. The laser detector 7 is suitable for detecting the intensity of the interferometric laser light. The laser detector 7 can be a HeNe laser detector with a wavelength response range of 300nm to 1100nm, capable of detecting HeNe interferometric laser light. The zero-crossing comparator circuit 8 is suitable for emitting a pulse signal based on the intensity of the interferometric laser light. This pulse signal is used to sample the intensity of the modulated interferometric light corresponding to each template detected by the single-point detector using equal optical path difference sampling. The pulse signal is converted into the sampled analog signal via the peripheral component interconnect bus acquisition card.

[0075] The aforementioned single-point detector 4 is a silicon photodetector or a photoelectric bias detector, which has high sensitivity and signal-to-noise ratio, and is easy to extend to other bands.

[0076] According to an embodiment of the present invention, the above-described single-pixel spectral imaging system further includes: a first focusing lens 11 and an optical fiber 12. The first focusing lens 11 is adapted to focus the interference white light emitted by the interferometer 1, and the optical fiber 12 is adapted to transmit the focused interference white light to the sample to be tested. The optical fiber 12 is an SMA905 optical fiber, which is a light transmission tool.

[0077] According to an embodiment of the present invention, the above-described single-pixel spectral imaging system further includes: a second focusing lens 13 and a collimating lens 14. The second focusing lens 13 is adapted to focus the modulated interference light corresponding to each template output by the spatial light modulator to obtain focused interference light. The collimating lens 14 is adapted to collimate the focused interference light and input the collimated interference light into the single-point detector. The collimating lens 10 is a reflective collimating lens, which can efficiently couple multicolor collimated light into the optical fiber 12 (multimode fiber).

[0078] According to an embodiment of the present invention, the first focusing lens 11 and the second focusing lens 13 are convex lenses, concave mirrors, or camera lenses.

[0079] According to an embodiment of the present invention, the measurement matrix to be displayed is pre-imported into the spatial light modulator 3 during the imaging process. The sample to be tested is placed on the sample holder 2, and the PCI acquisition card controls the Michelson interferometer to start by outputting a sine wave signal, while simultaneously outputting a square wave signal to the spatial light modulator 3 to control its template playback. The white light emitted by the white light source 9 and the HeNe laser emitted by the HeNe laser source are simultaneously input into the Michelson interferometer and become interference white light and HeNe interference laser, respectively. The interference white light is irradiated onto the sample to be tested after passing through the first focusing lens 11 and the optical fiber 12 and reacts with it (i.e., it passes through the sample to be tested and is partially absorbed by the sample to be tested), resulting in interference light, which is then directed to the spatial light modulator 3. The spatial light modulator 3 dynamically modulates the interference light by playing the template of the measurement matrix and reflects the modulated interference light to the second focusing lens 13. The light is then collected by the second focusing lens 13, collimated by the collimating lens 14, and coupled into the single-point detector 4, where it awaits acquisition by the PCI acquisition card 6. On the other side, the HeNe interference laser light, after being collected by the HeNe laser detector, is transmitted to the zero-crossing comparator circuit 8. The zero-crossing comparator circuit 8 is a waveform conversion circuit that emits a rising edge pulse signal when the intensity of the HeNe interference laser light is zero and transmits it to the PCI acquisition card. The PCI acquisition card uses this pulse signal as the sampling clock signal; that is, it samples the modulated interference light collected by the single-point detector 8 once for each rising edge received, thus completing the equal optical path difference sampling. After sampling, the PCI acquisition card stores the data in the computer 5 and performs algorithm processing to obtain the spectrum and image of the sample under test. The computer 5 uses a LabVIEW graphical interface to control the input and output of the acquisition card, uses a Fourier transform algorithm to obtain the spectrum of the sample under test, and uses an image reconstruction algorithm to obtain the image of the sample under test.

[0080] Figure 2 A flowchart of a single-pixel spectral imaging method according to an embodiment of the present invention is shown.

[0081] like Figure 2 As shown, the method includes operations S1-S5.

[0082] In operation S1, the interferometer is used to interfere with the externally input white light to obtain interfering white light.

[0083] In operation S2, the aforementioned interference white light is transmitted through the sample to be tested on the sample holder to obtain interference light carrying the spectral information of the sample to be tested.

[0084] In operation S3, a spatial light modulator is used to obtain at least one flip state based on at least one template of the measurement matrix, so as to perform amplitude modulation on the above interference light under the action of each flip state to obtain modulated interference light.

[0085] In operation S4, a single-point detector is used to detect the intensity of the modulated interference light corresponding to each template.

[0086] In operation S5, a computer is used to obtain the spectrum corresponding to each template based on the intensity of the modulated interference light corresponding to each template detected, and to obtain the image of the sample to be tested based on the spectra corresponding to all templates and the above measurement matrix, wherein the number of templates is less than the number of pixels in the image of the sample to be tested.

[0087] According to an embodiment of the present invention, after the spatial light modulator has played all the measurement matrices, the computer performs algorithmic processing on the intensity data of the acquired interference light. An interference data processing algorithm is used to reconstruct the spectrum, and an image reconstruction algorithm is used to obtain an image of the object under test.

[0088] According to an embodiment of the present invention, between operation S4 and operation S5, the above-described single-pixel spectral imaging method further includes sampling the intensity of the modulated interference light corresponding to each template detected by the single-point detector using a peripheral component interconnect bus acquisition card, and sending the sampling result to the computer.

[0089] According to an embodiment of the present invention, operation S5 includes S51-S52.

[0090] Figure 3 A flowchart illustrating the process of obtaining the spectrum corresponding to each template according to an embodiment of the present invention is shown.

[0091] like Figure 3 As shown, in operation S51, the spectrum corresponding to each template is obtained based on the intensity of the modulated interference light corresponding to each template detected, including operations S511-S515.

[0092] In operation S511, the sampling results of the intensity of the modulation interference light corresponding to each template are preprocessed to remove the DC component from the above sampling results, thus obtaining the preprocessed result. The specific operation is as follows:

[0093] The preprocessing operation removes the DC component from the interference data (the intensity of the modulated interference light corresponding to each template), suppressing random noise and eliminating the influence of stray light in the measurement environment. Assuming the amplitude transmittance of the interferometer's beam splitter is t, the amplitude reflectance is r, Δ represents the optical path difference between the two interfering beams in the interferometer, v is the wavenumber of the monochromatic beam emitted by the white light source, and B0(v) represents the intensity of the monochromatic beam emitted by the white light source, then the detected interference light intensity I corresponding to each template... D The formula for calculating (v) is as follows:

[0094]

[0095]

[0096] I D (∞) represents the DC component in the interference light intensity. This DC component can be removed using equation (3) to obtain the preprocessed result, which is the final interference light intensity I0(v).

[0097] I0(v)=I D (v)-C (3)

[0098] In operation S512, the above preprocessing results are subjected to atopic cutting to obtain the atopic cutting result.

[0099] Because the moving mirror of a Michelson interferometer can only move within a finite range of [-L, +L], it can only measure a finite maximum optical path difference L. This causes the interferogram to be abruptly truncated at ±L, resulting in a sharp discontinuity and spurious sidelobes in the interference data. Apodizing the preprocessed interference light intensity can suppress sidelobes while preserving the main lobe.

[0100] Apodization is generally performed by an apodization function. This method uses the Norton-Beer window function, with a window length of N, and its function expression is:

[0101]

[0102] The result of apodization, i.e., the interference light intensity I after apodization. N (v) Derived from the following series:

[0103]

[0104] in,

[0105] Q0=sincα

[0106] Q1 = 3(sincα - cosα) / α 2

[0107] Q2 = -15[(1-3 / a)] 2 sincα+(3 / α) 2 )cosα] / α 2

[0108] Q3 = 105[(1-15 / α)] 2 cosα+3(2-5 / α) 2 )sincα] / α 4

[0109] Q4 = 945[(1-45 / α)] 2 sincα+5 / α 2 (2-21 / α2 )cosα] / α 4

[0110] α = 2πvL, where v is the monochromatic light wavenumber and L is the maximum optical path difference.

[0111] This method selects the Norton-Beer window function with the apodization degree of Strong, so the values ​​of C0 to C4 are set as follows: C0 = 0.09, C1 = 0, C2 = 0.5875, C3 = 0, C4 = 0.3225.

[0112] In operation S513, the apodized result is zero-padded and subjected to a Fourier transform to obtain the initial spectrum. To ensure that the amount of interferogram data meets the requirements of the Fast Fourier Transform, we symmetrically pad both sides of the apodized interferogram with zeros. Then, a Fourier transform is performed on the zero-padded interferogram to obtain the initial spectrum of the sample under test.

[0113]

[0114] In operation S514, a phase correction operation is performed on the initial spectrum to obtain the spectrum corresponding to each template.

[0115] Errors caused by interferometric data acquisition and by optical and electronic devices can lead to phase errors in the interferogram. Phase correction is necessary to obtain accurate spectral information. This system uses the Mertz method for phase correction. The phase correction algorithm steps are as follows:

[0116] First, the phase spectrum expression obtained by the inverse Fourier transform. for:

[0117]

[0118] In the formula, B r (v) is the real part of the spectrum B(v), B i (v) is the imaginary part of the spectrum B(v).

[0119] Then, using the phase spectrum expression The original spectrum is corrected to obtain the final corrected spectrum (i.e., the spectrum corresponding to each template):

[0120]

[0121] Figure 4 A flowchart illustrating the process of obtaining an image of the sample to be tested based on the spectra corresponding to all templates and the measurement matrix, according to an embodiment of the present invention, is shown. Figure 4 As shown, in operation S52, the image of the sample to be tested is obtained based on the spectra corresponding to all templates and the above measurement matrix, including:

[0122] Operation S521: Calculate the energy values ​​b of the above measurement matrix A and the spectra corresponding to all templates at the target wavelength V. v The input is fed into an image reconstruction algorithm to obtain a constrained objective function;

[0123] After the interferometric data processing algorithm, the entire spectrum and measurement matrix are input into the image reconstruction algorithm to obtain the image of the sample to be tested. The image reconstruction algorithm model used in this system is a total variational regularized model, and the imaging process becomes a planning problem described by equation (11) to reconstruct the image of the object to be tested.

[0124]

[0125] In the formula, A is the measurement matrix, u v b is the image of the sample under test at wavelength v. v D represents the energy value of the spectrum to be measured at wavelength v. i u v For image u v Variations of .

[0126] The total variation formula is D. i u v The calculation is as follows

[0127]

[0128] Operation S522 transforms the constrained objective function using slack variables and the enhanced Lagrange multiplier method to obtain an unconstrained objective function;

[0129] Introducing the slack variable w, the model described by equation (11) becomes:

[0130]

[0131] Since the model described by equation (14) is constrained, this algorithm uses the enhanced Lagrange multiplier method to transform the constrained model into an unconstrained objective function. The objective function is finally transformed into:

[0132]

[0133] In the formula, μ is the primary penalty factor, and β is the secondary penalty factor. λ T All are Lagrange multipliers.

[0134] Operation S523 obtains the image of the sample under test at the target wavelength based on the unconstrained objective function.

[0135] The target problem (12) is transformed into two subproblems by using the alternating direction transformation method, namely the relaxation variable w and the image of the object under wavelength v. v .

[0136] The subproblems of the slack variable w are as follows:

[0137]

[0138] Image of the object under test at wavelength v v The subproblems are as follows:

[0139]

[0140] Thus, the image reconstruction model described by equation (9) is transformed into a solution model consisting of two sub-problems, equations (13) and (14).

[0141] After operating S524 and undergoing interferometric data processing, the complete spectral image and measurement matrix are input into the image reconstruction algorithm to obtain the image of the sample under test. The image reconstruction algorithm model used in this system is a fully variational regularized model, and the imaging process becomes a programming problem to reconstruct the image of the object under test.

[0142] By using an iterative method, we first calculate w, and then calculate u. v The algorithm iterates through the Lagrange daily numbers until it converges, then outputs the image u of the sample at wavelength v. v .

[0143] According to embodiments of the present invention, the use of a single-point detector eliminates the need for moving parts, simplifying the system structure, improving system stability, and making the system easier to use and maintain.

[0144] According to embodiments of the present invention, single-point detectors are inexpensive. The number of photons received by a single-point detector is much greater than that of an area array detector, which significantly reduces image distortion caused by dark noise.

[0145] According to embodiments of the present invention, the spectral information of the analyte can be obtained using a white light source and Fourier transform spectroscopy, and image information of the analyte in each spectral band can be obtained.

[0146] According to embodiments of the present invention, the use of compressed sensing technology can reconstruct the image under test at a lower sampling rate, reducing the amount of sampling data and significantly reducing imaging time.

[0147] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A single-pixel spectral imaging system, comprising: an interferometer adapted to interfere an externally inputted white light to obtain an interference white light, the interferometer being a Michelson interferometer; a sample holder on which a sample to be measured is placed, the interference white light transmitting through the sample to be measured to obtain an interference light carrying spectral information of the sample to be measured; a spatial light modulator adapted to obtain at least one flipping state according to at least one mask of a measurement matrix, so as to modulate the interference light under the action of each flipping state to obtain a modulated interference light; a single-point detector adapted to detect intensity of the modulated interference light corresponding to each mask; a computer adapted to obtain a spectrum corresponding to each mask according to the detected intensity of the modulated interference light corresponding to each mask, and obtain an image of the sample to be measured based on an image reconstruction algorithm according to the spectra corresponding to all masks and the measurement matrix, the number of masks being less than the number of pixel points of the image of the sample to be measured; wherein the interferometer is further adapted to interfere an externally inputted laser light to obtain an interference laser light; the single-pixel spectral imaging system further comprising: a laser light detector adapted to detect intensity of the interference laser light; a zero-crossing comparison circuit adapted to emit a digital signal according to the intensity of the interference laser light, the digital signal being adapted to perform equal optical path difference sampling on the intensity of the modulated interference light corresponding to each mask detected by the single-point detector, the digital signal being converted into a sampling analog signal by a peripheral component interconnect bus acquisition card. 2.The single-pixel spectral imaging system of claim 1, further comprising: a peripheral component interconnect bus acquisition card adapted to control starting of the interferometer, adapted to control playing of the masks, and adapted to emit a sampling analog signal to perform equal optical path difference sampling on the intensity of the modulated interference light corresponding to each mask detected by the single-point detector, and send the sampling result to the computer. 3.The single-pixel spectral imaging system of claim 1, further comprising: a first focusing lens adapted to focus the interference white light emitted by the interferometer; an optical fiber adapted to transmit the focused interference white light to the sample to be measured. 4.The single-pixel spectral imaging system of claim 1, further comprising: a second focusing lens adapted to focus the modulated interference light corresponding to each mask outputted by the spatial light modulator to obtain focused interference light; a collimating mirror adapted to collimate the focused interference light and input the collimated interference light to the single-point detector.

5. The single-pixel spectral imaging system of claim 1, wherein, the laser light being a HeNe laser light; the single-point detector being a silicon photodetector or a photoelectric bias detector. 6.A single-pixel spectral imaging method using the single-pixel spectral imaging system of any one of claims 1-5, the method comprising: interfering an externally inputted white light by an interferometer to obtain an interference white light; transmitting the interference white light through a sample to be measured on a sample holder to obtain an interference light carrying spectral information of the sample to be measured; obtaining at least one flipping state according to at least one template of a measurement matrix by using a spatial light modulator, so as to modulate the interference light in amplitude under the action of each flipping state to obtain modulated interference light; detecting the intensity of the modulated interference light corresponding to each template by using a single-point detector; obtaining the spectrum corresponding to each template according to the detected intensity of the modulated interference light corresponding to each template by using a computer, and obtaining the image of the sample to be measured according to the spectra corresponding to all templates and the measurement matrix, wherein the number of templates is less than the number of pixel points of the image of the sample to be measured.

7. The single-pixel spectral imaging method according to claim 6, further comprising: equalizing the path difference of the intensity of the modulated interference light corresponding to each template detected by the single-point detector by using a peripheral component interconnect bus acquisition card, and sending the sampling result to the computer.

8. The single-pixel spectral imaging method according to claim 7, wherein, obtaining the spectrum corresponding to each template according to the detected intensity of the modulated interference light corresponding to each template comprises: preprocessing the sampling result of the intensity of the modulated interference light corresponding to each template to remove the direct current component in the sampling result to obtain a preprocessing result; apodizing the preprocessing result to obtain an apodization result; zero-padding and Fourier transforming the apodization result to obtain an initial spectrum; phase correcting the initial spectrum to obtain the spectrum corresponding to each template.

9. The single-pixel spectral imaging method of claim 8, wherein, obtaining the image of the sample to be measured according to the spectra corresponding to all templates and the measurement matrix comprises: inputting the energy values of the measurement matrix and the spectra corresponding to all templates at a target wavelength into an image reconstruction algorithm to obtain a target function with constraints; converting the target function with constraints by using a relaxation variable and an enhanced Lagrange multiplier method to obtain a target function without constraints; obtaining the image of the sample to be measured at the target wavelength according to the target function without constraints.

Images