An ultrafast compressed phase imaging system and method

Abstract

This invention belongs to the field of high-speed imaging technology and discloses an ultrafast compressed phase imaging system and method. The invention splits a time-domain broadened pulse to obtain a first pulse and a second pulse. In the signal path, the first pulse is encoded and modulated to obtain a structured light signal. This structured light signal passes through a signal acquisition unit to form a signal path light pulse containing sample information. In the reference path, the second pulse passes through an adjustment unit to form a reference path light pulse. Both the signal path light pulse and the reference path light pulse are incident on a coupler to obtain an interference signal. This interference signal is compressed and acquired by a compression acquisition unit and then incident on a signal processing unit. The signal processing unit uses a reconstruction algorithm to recover the sample information. This invention avoids the complex optical field modulation process of current compressed phase imaging, breaks through the encoding rate of spatial light modulators, significantly improves the imaging frame rate, and achieves ultrafast continuous phase imaging.

Description

Technical Field

[0001] This invention belongs to the field of high-speed imaging technology, and more specifically, relates to an ultrafast compressed phase imaging system and method. Background Technology

[0002] In the fields of biomedicine and physical sciences, optical microscopy is the most common and important analytical method and testing tool. It can directly characterize and test objects, and is an important means for people to understand and study the microscopic world. However, traditional optical microscopy can only acquire intensity information of samples and cannot directly image or measure phase information. The emergence of phase imaging technology, such as white light interferometers and quantitative phase analyzers, has enabled people to visualize or quantitatively measure the phase distribution of samples, thereby capturing highly sensitive morphological and depth information of the sample under test. Compared with other imaging techniques, this technology does not require labeling, has the advantages of minimal sample damage, and can perform high-contrast imaging of transparent samples. However, according to the classical sampling theorem, phase imaging requires a very high sampling bandwidth and a large storage space to record signals containing phase information, which limits the widespread application of phase imaging.

[0003] In recent years, the emergence of compressed sensing (CS) technology has broken the limitations of traditional sampling theorems, effectively reducing the sampling bandwidth of signals and accurately recovering the original signals using a small amount of measurement data. However, existing compressed phase imaging is limited by the imaging principle of high-speed cameras and the refresh rate of encoding devices, and cannot yet be applied to real-time continuous observation of dynamic phenomena. Furthermore, the complexity of the light field manipulation process, poor robustness, and slow computation speed of phase recovery also require further research and solutions. Summary of the Invention

[0004] This invention provides an ultrafast compressed phase imaging system and method, which solves the problems of complex optical field control, slow encoding rate, and time-consuming phase recovery calculation in existing compressed phase imaging technologies.

[0005] This invention provides an ultrafast compressed phase imaging method, comprising the following steps:

[0006] The time-domain broadened pulse is split to obtain a first pulse and a second pulse; the first pulse is sent to the signal path, and the second pulse is sent to the reference path.

[0007] In the signal path, the first pulse is encoded and modulated to obtain a structured light signal, and the structured light signal is converted into a signal path light pulse containing sample information after passing through the signal acquisition unit; in the reference path, the second pulse is converted into a reference path light pulse after passing through the adjustment unit.

[0008] Both the signal path optical pulse and the reference path optical pulse are incident on the coupler and an interference signal is obtained. The interference signal is compressed and acquired by the compression acquisition unit and then incident on the signal processing unit. The signal processing unit uses a reconstruction algorithm to recover the sample information.

[0009] Preferably, a femtosecond laser is used to generate femtosecond pulses, a time-domain dispersive device is used to broaden the femtosecond pulses in the time domain, and a beam splitter is used to perform beam splitting on the time-domain broadened pulses.

[0010] Preferably, the electro-optic modulator encodes and modulates the first pulse according to the code generated by the code generator to obtain the structured light signal.

[0011] Preferably, the first pulse is sinusoidally encoded and modulated; in the reference path, the adjustment unit includes a delay component, which is used to adjust the optical path difference between the reference path and the signal path.

[0012] Preferably, the first pulse is subjected to 01 random coding modulation; in the reference path, the adjustment unit includes a delay component and a phase interferometer, the delay component is used to adjust the optical path difference between the reference path and the signal path, and the phase interferometer is used to adjust the phase of the reference light.

[0013] Preferably, the compression acquisition unit includes a time-domain compression device and a single-pixel photodetector. The interference signal is transmitted to the time-domain compression device, which compresses the time-domain broadened pulse into a femtosecond pulse. The single-pixel photodetector captures the pulse to obtain a one-dimensional signal.

[0014] The signal processing unit includes an analog-to-digital converter and a computing device. The analog-to-digital converter converts the one-dimensional signal into a digital signal and transmits it to the computing device. The computing device uses a corresponding reconstruction algorithm to recover the sample information according to the type of coding and modulation.

[0015] Preferably, when sinusoidal coding modulation is used, the inverse Fourier transform method is used to recover the sample phase information; when 01 random coding modulation is used, the orthogonal matching pursuit algorithm is used to recover the sample information.

[0016] On the other hand, the present invention provides an ultrafast compressed phase imaging system, comprising: a time-domain stretched beam splitting unit, a pulse modulation unit, a signal acquisition unit, an adjustment unit, a coupler, a compressed acquisition unit, and a signal processing unit;

[0017] The time-domain stretching and beam splitting unit is used to perform beam splitting on the time-domain stretched pulse to obtain a first pulse and a second pulse;

[0018] The pulse modulation unit is used to encode and modulate the first pulse to obtain a structured light signal;

[0019] The signal acquisition unit is used to receive the structured light signal and generate a signal path light pulse containing sample information;

[0020] The adjustment unit is used to receive the second pulse and form a reference path light pulse;

[0021] The coupler is used to couple the signal path optical pulse and the reference path optical pulse to obtain an interference signal;

[0022] The compression acquisition unit is used to compress and acquire the interference signal;

[0023] The signal processing unit is used to recover sample information based on the compressed acquisition signal using a reconstruction algorithm;

[0024] The ultrafast compressed phase imaging system is used to implement the steps in the ultrafast compressed phase imaging method described above.

[0025] Preferably, the time-domain stretched beam splitting unit includes a femtosecond laser, a time-domain dispersive device, and a beam splitting device;

[0026] The pulse modulation unit includes an electro-optic modulator and a code generation device;

[0027] The signal acquisition unit includes a first spatial dispersive device, a focusing component, and a second spatial dispersive device arranged sequentially along the optical path;

[0028] The adjustment unit includes a delay component, or it includes a delay component and a phase interferometer;

[0029] The compressed acquisition unit includes a time-domain compression device and a single-pixel photodetector;

[0030] The signal processing unit includes an analog-to-digital converter and a computing device.

[0031] Preferably, the time-domain dispersion device uses single-mode fiber, the beam splitter uses a beam splitter or a 1×2 fiber coupler, the pattern generator uses an arbitrary waveform generator, and the first spatial dispersion device uses a first diffraction grating; the focusing assembly includes a first 4f system, a first objective lens, a second objective lens, and a second 4f system arranged sequentially along the optical path, with the sample located between the first objective lens and the second objective lens; the second spatial dispersion device uses a second diffraction grating, the delay assembly consists of several mirrors, and the time-domain compression device uses dispersion-compensating fiber.

[0032] One or more technical solutions provided in this invention have at least the following technical effects or advantages:

[0033] This invention first splits a time-domain broadened pulse to obtain a first pulse and a second pulse. The first pulse is sent to the signal path, and the second pulse is sent to the reference path. Then, in the signal path, the first pulse is encoded and modulated to obtain a structured light signal. This structured light signal passes through a signal acquisition unit to form a signal path light pulse containing sample information. In the reference path, the second pulse passes through an adjustment unit to form a reference path light pulse. Subsequently, both the signal path light pulse and the reference path light pulse are incident on a coupler to obtain an interference signal. This interference signal is compressed and acquired by a compression acquisition unit before being sent to a signal processing unit. The signal processing unit uses a reconstruction algorithm to recover the sample information. In other words, this invention utilizes off-axis interferometric imaging, designs encoded and modulated time-domain pulses for sparse acquisition of sample information, acquires compressed interference signals, and finally uses a reconstruction algorithm to recover the signal phase and intensity information. This invention effectively reduces the amount of data required for traditional phase imaging, avoids the complex light field modulation process of current compressed phase imaging, breaks through the encoding rate of spatial light modulators, significantly increases the imaging frame rate, shortens the phase recovery calculation time, and achieves ultrafast continuous phase imaging. Attached Figure Description

[0034] Figure 1 This is a schematic diagram of the framework of an ultrafast compressed phase imaging method provided in Embodiment 1 of the present invention;

[0035] Figure 2 This is a connection diagram of an ultrafast compressed phase imaging system provided in Embodiment 2 of the present invention;

[0036] Figure 3 This is a connection diagram of an ultrafast compressed phase imaging system provided in Embodiment 3 of the present invention. Detailed Implementation

[0037] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0038] Example 1:

[0039] Example 1 provides an ultrafast compressed phase imaging method, see [link to example]. Figure 1 This includes the following steps:

[0040] Step 1: Split the time-domain broadened pulse to obtain a first pulse and a second pulse; the first pulse is transferred to the signal path, and the second pulse is transferred to the reference path.

[0041] Specifically, a femtosecond laser is used to generate femtosecond pulses, a time-domain dispersion device is used to broaden the femtosecond pulses in the time domain, and a beam splitter is used to perform beam splitting on the time-domain broadened pulses.

[0042] Step 2: In the signal path, the first pulse is encoded and modulated to obtain a structured light signal. The structured light signal is then processed by the signal acquisition unit to form a signal path light pulse containing sample information. In the reference path, the second pulse is processed by the adjustment unit to form a reference path light pulse.

[0043] Specifically, the electro-optic modulator encodes and modulates the first pulse according to the code generated by the code generator to obtain the structured light signal. This step precisely modulates the code onto the time-domain waveform of the stretched pulse, maintaining synchronization between the code and the laser pulse.

[0044] Depending on whether sinusoidal coding or 0 / 1 random coding is used, different imaging schemes are employed. When using the sinusoidal coding modulation scheme for high-speed pulses, a four-step phase shift method is used to acquire samples in the signal path; when using the 0 / 1 random coding modulation scheme, a four-step phase shift is applied to the reference path.

[0045] If the first pulse is sinusoidally encoded and modulated, then in the reference path, the adjustment unit includes a delay component, which is used to adjust the optical path difference between the reference path and the signal path.

[0046] If the first pulse is subjected to 01 random coding modulation, then in the reference path, the adjustment unit includes a delay component and a phase interferometer. The delay component is used to adjust the optical path difference between the reference path and the signal path, and the phase interferometer is used to adjust the phase of the reference light.

[0047] In step 3, both the signal path optical pulse and the reference path optical pulse are incident on the coupler to obtain an interference signal. The interference signal is compressed and acquired by the compression acquisition unit and then incident on the signal processing unit. The signal processing unit uses a reconstruction algorithm to recover the sample information.

[0048] Specifically, the compression acquisition unit includes a time-domain compression device and a single-pixel photodetector. The interference signal is transmitted to the time-domain compression device, which compresses the time-domain broadened pulse into a femtosecond pulse. The single-pixel photodetector captures the pulse to obtain a one-dimensional signal. The signal processing unit includes an analog-to-digital converter and a computing device. The analog-to-digital converter converts the one-dimensional signal into a digital signal and transmits it to the computing device. The computing device uses a corresponding reconstruction algorithm based on the type of coding modulation to recover the sample information.

[0049] If sinusoidal coding modulation is used, the inverse Fourier transform method is used to recover the sample phase information; if 01 random coding modulation is used, the orthogonal matching pursuit algorithm is used to recover the sample information.

[0050] Corresponding to the method provided above, Embodiment 1 also provides an ultrafast compressed phase imaging system, including: a time-domain stretched beam splitting unit, a pulse modulation unit, a signal acquisition unit, an adjustment unit, a coupler, a compressed acquisition unit, and a signal processing unit.

[0051] The time-domain stretching and beam splitting unit is used to split the time-domain stretched pulse to obtain a first pulse and a second pulse. Specifically, the time-domain stretching and beam splitting unit includes a femtosecond laser, a time-domain dispersive device, and a beam splitting device. The time-domain dispersive device can be a single-mode fiber, and the beam splitting device can be a beam splitter or a 1×2 fiber coupler.

[0052] The pulse modulation unit is used to encode and modulate the first pulse to obtain a structured light signal. Specifically, the pulse modulation unit includes an electro-optic modulator and a pattern generator. The optical signal interface of the electro-optic modulator is connected to the first pulse, and the radio frequency signal interface of the electro-optic modulator is connected to the pattern generator. The pattern generator can be an arbitrary waveform generator.

[0053] The signal acquisition unit is used to receive the structured light signal and form a signal path light pulse containing sample information. Specifically, the signal acquisition unit includes a first spatial dispersion device, a focusing component, and a second spatial dispersion device arranged sequentially along the optical path. The first spatial dispersion device may be a first diffraction grating to disperse the pulse in space; the focusing component includes a first 4f system, a first objective lens, a second objective lens, and a second 4f system arranged sequentially along the optical path, with the sample located between the first objective lens and the second objective lens. That is, the light signal is sequentially shaped and focused onto the sample surface by the first 4f lens and the first objective lens to acquire sample information, and the signal path light pulse containing sample information is irradiated onto the second spatial dispersion device after passing through the second objective lens and the second 4f lens; the second spatial dispersion device may be a second diffraction grating.

[0054] The adjustment unit is used to receive the second pulse and form a reference path optical pulse. Specifically, the adjustment unit includes a delay component, or it includes a delay component and a phase interferometer. The delay component may consist of several mirrors.

[0055] The coupler is used to couple the signal path optical pulse and the reference path optical pulse to obtain an interference signal.

[0056] In the signal path, the electro-optic modulator modulates the optical pulses of the signal path according to the code generated by the pattern generator, producing structured light. This structured light is then shaped and focused onto the sample under test via a spatial dispersor, a 4f system, and an objective lens, achieving high-speed compressed sampling of the sample. The optical pulses are then spatially combined and incident on one end of the coupler. In the reference path, the reference optical signal is incident on the delay component. The optical path difference between the reference path and the signal path is adjusted, and the reference optical signal, after passing through the delay component, is incident on the other end of the coupler. Both the signal and reference optical signals can be collimated by a collimator before entering the coupler, and the reference optical signal can also be collimated by a collimator before entering the delay component.

[0057] The compression acquisition unit is used to compress and acquire the interference signal. Specifically, the compression acquisition unit includes a time-domain compression device and a single-pixel photodetector. The time-domain compression device can use dispersion-compensating fiber to compress the time-domain broadened pulse into a femtosecond pulse.

[0058] The signal processing unit is used to recover sample information from the compressed acquired signal using a reconstruction algorithm. Specifically, the signal processing unit includes an analog-to-digital converter and a computing device.

[0059] That is, the signal processing unit selects two different reconstruction algorithms based on the encoding to achieve signal compression and information reconstruction.

[0060] Existing compressed phase imaging requires modulation and encoding using spatial light modulators such as DMDs, with encoding refresh rates limited to kHz. The imaging frame rate is constrained by the refresh rate of the spatial light modulator, making it unsuitable for high-speed imaging. Furthermore, the control precision and imaging resolution of spatial light modulator-based imaging systems are limited by the pixel size of the modulator itself. In contrast, the time-domain modulation method used in Example 1 utilizes a high-speed electro-optic modulator to achieve GHz modulation frequencies while simultaneously encoding the time-domain waveform with nanosecond-level precision. Example 1's use of an electro-optic modulator to modulate the time-domain waveform enables high-speed encoding, overcoming the encoding rate limitation and achieving MHz scanning frame rates. Moreover, existing compressed phase imaging methods using spatial light to control cosine encoding require pre-calculation of the encoding, converting it into 0 / 1 codes before modulation, reducing encoding precision and increasing system complexity. The system provided in Example 1 is simpler and has better robustness.

[0061] Example 1 utilizes the principle of compressed sensing to recover signals without acquiring the complete number of pulse points. In the signal acquisition unit, a time-domain compression device is used to compress the stretched femtosecond pulses back into ultrashort pulses. At this time, the photodetector captures only one point for each pulse, which can greatly reduce the amount of signal data acquired.

[0062] The sampling recovery in Example 1 utilizes a traditional compressed sensing algorithm, but the system employs one-dimensional pulse scanning. The algorithm recovers the signal immediately after capturing it, which, compared to two-dimensional signal recovery, is faster and simpler, saving computational resources. It also matches the acquisition speed of the imaging system to achieve real-time imaging recovery. Furthermore, when the system uses sinusoidal encoding, the inverse Fourier transform method can be used to recover the signal, which is faster than the iterative calculations required by traditional algorithms.

[0063] The invention will now be further described in conjunction with a specific system.

[0064] Example 2:

[0065] This embodiment 2 provides an ultrafast compressed phase imaging system, see [link to documentation]. Figure 2 The optical path of the system is as follows: a femtosecond laser 101 generates femtosecond pulses, which are then stretched in the time domain by a dispersive fiber 102. A beam splitter 103 divides the stretched pulses into two paths: one as a signal path and the other as a reference path. The optical signal interface of the electro-optic modulator 105 receives the pulses from the signal path, while the radio frequency (RF) signal interface connects to an arbitrary waveform generator 104 to perform sinusoidal encoding modulation on the stretched pulses. The RF signal terminal of the femtosecond laser 101 is connected to a low-pass filter and an external clock for the arbitrary waveform generator 104. The output structured light signal is incident on a first diffraction grating 106, which disperses the pulses in space. The pulses are then shaped and focused onto the sample surface by a first 4f system 107 and a first objective lens 108 to collect sample information. The light pulses containing the sample information pass through a second objective lens 109 and a second 4f system 110 before illuminating a second diffraction grating 111. The light pulses are then combined in the spatial domain and collimated before being incident on a coupler 117. The reference optical signal is incident on a delay assembly consisting of a first reflector 112, a second reflector 113, a third reflector 114, a fourth reflector 115, and a fifth reflector 116. This adjusts the optical path difference between the reference path and the signal path. The reference optical signal, after passing through the delay assembly, is collimated and incident on a coupler 117, interfering with the signal from the signal path. The interference signal is connected to a dispersion-compensating fiber 118, which compresses the time-domain stretched signal. The signal is then captured by a photodetector 119 and converted into a digital signal by an analog-to-digital converter, which transmits it to a computing device. A reconstruction algorithm is then used to recover the sample phase information.

[0066] Example 3:

[0067] This embodiment 3 provides an ultrafast compressed phase imaging system, see [link to documentation]. Figure 3The optical path of the system is as follows: a femtosecond laser 201 generates laser light; the optical signal is stretched in the time domain through a dispersive fiber 202, and the stretched pulse is split into two paths using a beam splitter 203, one as the signal path and the other as the reference path. The optical signal interface of the electro-optic modulator 205 is connected to the pulse of the signal path, and the radio frequency signal interface is connected to an arbitrary waveform generator 204 to perform 01 encoding modulation on the stretched pulse. The radio frequency signal terminal of the femtosecond laser 201 is connected to a low-pass filter and connected to the external clock of the arbitrary waveform generator 204. The output structured light signal is incident on the first diffraction grating 206, which disperses the pulse in space, and then passes through the first 4f system 207 and the first objective lens 208 to be shaped and focused onto the sample surface to collect sample information. The light pulse containing the sample information passes through the second objective lens 209 and the second 4f system 210 and then illuminates the second diffraction grating 211. The light pulse is then combined in the spatial domain and collimated before being incident on the coupler 218. The reference path optical signal is incident on an adjustment unit consisting of a first reflector 212, a phase interferometer 213, a second reflector 214, a third reflector 215, a fourth reflector 216, and a fifth reflector 217. This unit adjusts the optical path difference between the reference path and the signal path, and also adjusts the phase of the reference light. The adjusted reference path optical signal is then collimated and incident on a coupler 218, where it interferes with the signal from the signal path. The interference signal is connected to a dispersion-compensating fiber 219, which compresses the time-domain stretched signal. The signal is then captured by a photodetector 220 and converted into a digital signal by an analog-to-digital converter, which transmits it to a computing device. A reconstruction algorithm is then used to recover the sample phase information.

[0068] In Embodiment 2 or Embodiment 3, the femtosecond laser 101 is selected as a pulsed laser with a center wavelength of 1550nm, a spectral width of 15nm, a pulse width of 100fs, and a repetition frequency of 101.7MHz; the dispersive fiber 102 is a single-mode fiber with a dispersion coefficient of 17ps / km / nm; the electro-optic modulator 105 is a 40Gbps Mach-Zehnder modulator in the 1550nm band; the arbitrary waveform generator (AWG) 104 is selected as Keysight Technologies M8195A; the first objective lens 108 and the second objective lens 109 are selected as Mitutoyo M Plan Apo NIR HR with a numerical aperture of 0.65 and a magnification of 50x; the collimator is selected as Thorlabs F260FC-1550; the 4f lens is selected as having a focal length of f=100mm and a focal length of f=50mm; and the photodetector 119 is a 10G bandwidth PD.

[0069] In summary, this invention is simple to control, can effectively improve the imaging frame rate, reduce sampling data, and accurately and quickly recover sample information.

[0070] Finally, it should be noted that the above specific embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to examples, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. An ultrafast compressed phase imaging method, characterized in that, Includes the following steps: The time-domain broadened pulse is split to obtain a first pulse and a second pulse; the first pulse is sent to the signal path, and the second pulse is sent to the reference path. In the signal path, the electro-optic modulator encodes and modulates the first pulse according to the code generated by the code generator to obtain a structured light signal. The structured light signal is then processed by the signal acquisition unit to form a signal path light pulse containing sample information. In the reference path, the second pulse forms a reference path optical pulse after passing through the adjustment unit; Both the signal path optical pulse and the reference path optical pulse are incident on the coupler and an interference signal is obtained. The interference signal is compressed and acquired by the compression acquisition unit and then incident on the signal processing unit. The signal processing unit uses a reconstruction algorithm to recover the sample information. The compression acquisition unit includes a time-domain compression device and a single-pixel photodetector. The interference signal is transmitted to the time-domain compression device, which compresses the time-domain broadened pulse into a femtosecond pulse. The single-pixel photodetector captures the pulse to obtain a one-dimensional signal. The signal processing unit includes an analog-to-digital converter and a computing device. The analog-to-digital converter converts the one-dimensional signal into a digital signal and transmits it to the computing device. The computing device uses a corresponding reconstruction algorithm to recover the sample information according to the type of coding and modulation.

2. The ultrafast compressed phase imaging method according to claim 1, characterized in that, Femtosecond pulses are generated using a femtosecond laser, the femtosecond pulses are broadened in the time domain using a time-domain dispersive device, and the time-domain broadened pulses are split using a beam splitter.

3. The ultrafast compressed phase imaging method according to claim 1, characterized in that, The first pulse is sinusoidally encoded and modulated; in the reference path, the adjustment unit includes a delay component, which is used to adjust the optical path difference between the reference path and the signal path.

4. The ultrafast compressed phase imaging method according to claim 1, characterized in that, The first pulse is subjected to 01 random coding modulation; in the reference path, the adjustment unit includes a delay component and a phase interferometer, the delay component is used to adjust the optical path difference between the reference path and the signal path, and the phase interferometer is used to adjust the phase of the reference light.

5. The ultrafast compressed phase imaging method according to claim 1, characterized in that, When sinusoidal coding modulation is used, the inverse Fourier transform method is used to recover the sample phase information; when 01 random coding modulation is used, the orthogonal matching pursuit algorithm is used to recover the sample information.

6. An ultrafast compressed phase imaging system, characterized in that, include: The system includes a time-domain broadened beam splitting unit, a pulse modulation unit, a signal acquisition unit, an adjustment unit, a coupler, a compressed acquisition unit, and a signal processing unit. The time-domain stretching and beam splitting unit is used to perform beam splitting on the time-domain stretched pulse to obtain a first pulse and a second pulse; The pulse modulation unit includes an electro-optic modulator and a code generator. The pulse modulation unit is used to encode and modulate the first pulse to obtain a structured light signal. The signal acquisition unit is used to receive the structured light signal and generate a signal path light pulse containing sample information; The adjustment unit is used to receive the second pulse and form a reference path light pulse; The coupler is used to couple the signal path optical pulse and the reference path optical pulse to obtain an interference signal; The compression acquisition unit includes a time-domain compression device and a single-pixel photodetector, and the compression acquisition unit is used to compress and acquire the interference signal; The signal processing unit includes an analog-to-digital converter and a computing device. The signal processing unit is used to recover sample information based on the compressed acquired signal using a reconstruction algorithm. The ultrafast compressed phase imaging system is used to implement the steps in the ultrafast compressed phase imaging method as described in any one of claims 1-5.

7. The ultrafast compressed phase imaging system according to claim 6, characterized in that, The time-domain stretching beam splitting unit includes a femtosecond laser, a time-domain dispersive device, and a beam splitting device; The signal acquisition unit includes a first spatial dispersive device, a focusing component, and a second spatial dispersive device arranged sequentially along the optical path; The adjustment unit includes a delay component, or it includes a delay component and a phase interferometer.

8. The ultrafast compressed phase imaging system according to claim 7, characterized in that, The time-domain dispersion device uses single-mode fiber, the beam splitter uses a beam splitter or a 1×2 fiber coupler, the pattern generator uses an arbitrary waveform generator, and the first spatial dispersion device uses a first diffraction grating; the focusing assembly includes a first 4f system, a first objective lens, a second objective lens, and a second 4f system arranged sequentially along the optical path, with the sample located between the first objective lens and the second objective lens; the second spatial dispersion device uses a second diffraction grating, the delay assembly consists of several mirrors, and the time-domain compression device uses dispersion-compensating fiber.

Images