Single-pixel imaging system based on DMD and scanning mirror and high-resolution field-of-view expansion method thereof

Abstract

The invention discloses a single-pixel imaging system based on a DMD and a scanning mirror and a high-resolution view field expansion method thereof, and belongs to the technical field of laser imaging, and the system comprises a light path unit, the DMD, the scanning mirror, a single-photon detector, a time correlation single photon counting (TCSPC) system and a control unit. The light path unit controls the scanning mirror to scan and expand an image reflected by a target object to the DMD, and echo light signals are collected and coupled to enter the single-photon detector; and the DMD is connected with the TCSPC system to trigger a synchronizing signal to carry out data marking. In the forming process of the optimized image of the scanning mirror, the control unit gradually changes the position angle information of the scanning mirror until the position with the highest signal-to-noise ratio for calibration, expands an imaging view field, scans and expands an image reflected on a target object to the DMD, realizes efficient sampling through compressed sensing, and adopts an imaging algorithm for image recovery, reconstruction and splicing. According to the invention, large-view-field full-automatic high-speed scanning imaging of a long-distance target object can be realized, and real-time capturing and monitoring of a moving object can be realized.

Description

Technical Field

[0001] The invention belongs to the technical field of laser imaging, and relates to a single-pixel imaging system based on a DMD and a scanning mirror and a high-resolution field of view expansion method thereof. Background Art

[0002] Single-pixel imaging is a new type of computational imaging technology that uses a single pixel for imaging and reconstructs the intensity information of the object to be imaged through an algorithm, breaking through the limitations of traditional matrix imaging technology. This technology can use the intensity information of light to image objects without the need for a spatial sampling matrix, and ultimately achieves imaging of the object by continuously changing the lighting structure and accumulating related information. Due to its unique advantages in signal acquisition and processing, it can combine compressed sensing in the data acquisition step to reduce data storage and data transmission requirements, and can still obtain effective images under low light conditions, making it very suitable for detection in complex environments such as strong noise, high scattering, and long distances.

[0003] However, in the detection of ultra-distant targets, the telescope is used as a receiving device for reflected light, and its effect of receiving light is closely related to the magnification of the telescope. The resolution of high-magnification receiving devices is limited by the design of the optical system and the target distance, while the imaging field of view is limited by the size of the lens field of view. There is inevitably a problem that it is difficult to balance resolution and imaging field of view. Huang et al. integrated traditional point-by-point scanning imaging and single-pixel technology and proposed a scanning single-pixel imaging lidar solution. The scanning mirror controls the light source to scan distant objects, expands to DMD for compressed sensing imaging, and then merges and reconstructs, thereby achieving an imaging resolution of 896×128 pixels for a transmitting tower 7km away [J.Huang et al., "Scanning single-pixel imaging lidar," Optics Express, vol.30, no.21, 2022.]. Although this method can save most of the time spent on imaging the background area, the efficiency is still not high enough because it is first scanned and then expanded to DMD to image each sub-area, which makes the scanning time relatively long and has certain limitations. In the present invention, the laser is first emitted to the target object, and then expanded to the DMD through the lens group from the scanning mirror for imaging, which can greatly save the imaging range of the background area, improve the imaging efficiency, and effectively expand the field of view. Summary of the invention

[0004] In response to the defects of the prior art or the need for improvement, the present invention proposes a high-resolution field of view expansion method for a single-pixel imaging system based on DMD and a scanning mirror, thereby solving the problem of field of view expansion limitations caused by long-distance detection in traditional single-pixel imaging systems and low imaging resolution when identifying small target objects in the field of view.

[0005] To solve the above problems, the technical solution adopted by the present invention is:

[0006] A single pixel imaging system based on DMD and scanning mirror, the single pixel imaging system comprises an optical path unit, a control unit, a single photon detector 3 and a TCSPC (time correlated single photon counting) system 2; specifically:

[0007] The optical path unit is used to illuminate the entire target object 14 after the laser 13 emits the laser, the distance between the target object and the scanning mirror is about 37.5 cm, and control the high-precision scanning mirror 12 to scan and expand the image reflected by the target object to the DMD 10 through the imaging lens 11, and the distance between the DMD 10 and the scanning mirror 12 is about 12.5 cm. The optical path unit collects the target echo light signal, filters out stray light and noise, and then collects and couples the filtered echo light signal into the single-photon detector 3;

[0008] The control unit obtains the scanning step length and number of steps through the distance, and is used to gradually change the position angle information of the scanning mirror 12 until the position with the highest signal-to-noise ratio is reached to achieve fine scanning imaging;

[0009] The single photon detector 3 is used to detect the target echo signal in the fiber coupling mirror 9 and process the received echo signal into a photon counting signal;

[0010] The TCSPC system 2 is used to receive the photon counting signal and measure the number of DMD mask switching under the action of the synchronization signal output by the DMD 10, and then perform data marking and segmentation to achieve image reconstruction.

[0011] Preferably, the optical path unit includes a laser 13, an imaging lens 11, a scanning mirror 12, a DMD 10, a converging lens 8, and a fiber coupling mirror 9; specifically: the laser 13 is used to emit laser light to illuminate a target object 14 about 37.5 cm away; the imaging lens 11 collects the echo signal reflected by the laser 13 to the target object 14 to the scanning mirror 12, and scans the image of the target object to the DMD 10. When the scanning mirror 12 is working, the angle of the incident light is inconsistent at each position, so the output light path of the DMD 10 will change with the change of the position of the scanning mirror. The DMD 10 is used to spatially modulate the incident light. The deflection state of each micromirror is controlled to achieve precise control of light and thus realize high-resolution imaging. The DMD target surface size is only 0.7 inches and has a 45° angle with the horizontal direction to ensure that the incident and outgoing light paths are within the same degree of freedom. Each micromirror of DMD10 is attached to its own support axis, and the storage unit of each micromirror can be electronically addressed independently, allowing each mirror to be electrostatically tilted to both sides by +12° and -12°, and the total number of photons collected in the +12° direction of DMD10; the imaging lens is used to couple the outgoing light on DMD10 into the coupling mirror; the fiber coupling mirror 9 is used to couple the collected echo signal into the single-photon detector 3.

[0012] Preferably, the control unit includes a DAC control module 1, a scanning mirror controller 6, a DMD controller 7, an MCU single chip computer 5 and a NUC processor 4; specifically: the DAC control module 1 is located on the MCU single chip computer 5, and is used to output ±10V analog voltage to control the ±10mrad deflection of the scanning mirror; the scanning mirror controller 6 is connected to the scanning mirror 12, and is used to drive the scanning mirror to move; the DMD controller 7 is connected to the DMD10 through a flat cable, and is connected to the NUC controller 4 through RJ45, and the Hadamard mask is input into the DMD10, which is used to drive the micromirror deflection rate of the DMD10, so as to achieve precise control of light, and then generate high The MCU 5 obtains the movement step length and number of steps of the scanning mirror 12 according to the position of the target object 14, and controls the large-field fine scanning of the image of the target object to the DMD 10; the NUC controller transmits the photon data collected by the TCSPC system 2 through the USB3.0 interface, and after completing the scanning of the imaging area of ​​the target object, processes it according to the scanning logic combined with the optimized imaging algorithm, realizes the splicing imaging of the target object and displays it, and realizes fully automatic and fast large-field scanning imaging.

[0013] Furthermore, the MCU microcontroller 5 and NUC processor 4 in the control unit package control the single-pixel imaging system, with the NUC processor 4 as the main control and the MCU microcontroller 5 as the auxiliary control to control the single-pixel imaging system and achieve super-resolution imaging of the target object in a large field of view.

[0014] Furthermore, the optical path unit extends the image scanning reflected from the target object to the DMD 10, which can greatly save the imaging time consumption of the background area and improve the efficiency compared with the traditional object scanning.

[0015] A method for expanding the high-resolution field of view of a single-pixel imaging system based on a DMD and a scanning mirror, using the single-pixel imaging system of the DMD 10 and the scanning mirror 12, comprises the following steps:

[0016] Step S1, the laser 13 emits laser light to illuminate the target object 14, and the echo signal returns along the original path and converges onto the scanning mirror 12 through the imaging lens 11;

[0017] Step S2, MCU 5 controls scanning mirror 12 to scan echo signal to DMD10, DMD and TCSPC are initialized and relevant parameters are set, after successful initialization, echo signal enters fiber coupling mirror 9 through DMD10, single photon detector 3 detects echo signal, TCSPC system 2 collects photon counting signal, NUC processor 4 analyzes collected photon data, combines optimized imaging algorithm to perform image denoising, splicing and reconstruction, and obtains final large field of view imaging result image. It includes the following sub-steps:

[0018] Step 2.1, the MCU single chip computer 5 first controls the scanning mirror 12 according to the distance between the DMD10 and the scanning mirror 12, calculates the step length and number of steps required at this time, and then performs a "Z"-shaped scan; in order to obtain the best position of the PSNR of the image on the DMD, the voltage is increased by 0.1V each time, that is, the angle of the scanning mirror is changed by 0.1mrad, the scanning mirror 12 moves the image returned from the target object on the DMD10 by one position, and the DMD10 reconstructs the image, and calculates the PSNR (peak signal-to-noise ratio) of the image at this position at this time, and so on, calculates the PSNR of the reconstructed image on the DMD10 through the scanning mirror movement in turn, takes the position with the highest PSNR as the mark position, and records the voltage value output by the MCU single chip computer 5 at this time;

[0019] Step 2.2, the target image is divided into n×n blocks, and the scanning mirror 12 stops at n 2The optimal PSNR position (the voltage value output by the MCU 5) of each position should be kept for a period longer than the time for a group of DMD masks to be played. Usually, the time for a group of masks to be played is set to 4 seconds. Each time the mask is switched, a trigger signal will be output to the TCSPC system 2 for data marking, which is convenient for subsequent image processing.

[0020] Step 2.3, the data collected by the TCSPC system and the Hadamard mask are used as inputs of the reconstruction algorithm to adapt it to the spatial imaging of the DMD and the scanning mirror. The algorithm uses the first "1" flag in the collected data as the starting signal and the last "1" as the ending signal. The imaging matrix is ​​obtained according to the logic of compressed sensing, and then normalized. It can also denoise, splice, and reconstruct each block image extended from the scanning mirror 12 to the DMD10.

[0021] In particular, the scanning mirror 12 and the DMD 10 are used to set the adaptation range of the DMD according to the following formula:

[0022] Adaptation range

[0023] Among them, d represents the DMD adaptation range, L represents the distance of the target object, θ represents the maximum deflection angle of the scanning mirror, V represents the maximum output voltage of the DAC module, and U represents the given input voltage. It can be seen that with the change of distance and voltage, a super-resolution image of the target object can be obtained;

[0024] Preferably, the target image is reconstructed after the range adaptation, which can be expressed as follows:

[0025] Image information to be tested X = A -1 Y=A -1 AO=O

[0026] Among them, O is the target vector (size is N×1), which is formed by the target object through N Hadamard mask vectors arranged in columns, A is the mask observation matrix and is a reversible matrix with size N×N. The mask vectors are linearly independent, that is, the observation matrix A must have a reversible matrix A -1 , Y is the measurement value vector (size is N×1);

[0027] Step 2.4, through step 2.1, step 2.2 and step 2.3, the scanning mirror under the control of the MCU microcontroller and the DMD ensure the super-resolution imaging effect of the system, obtain a clearer field of view, and realize the fully automatic fine high-speed scanning imaging of the system with a large field of view.

[0028] The beneficial effects of the present invention are:

[0029] (1) The present invention controls a high-precision scanning mirror as a scanning unit and combines it with a DMD to perform fast and high-resolution imaging, thereby achieving large-field, fully automatic, high-speed scanning imaging of distant target objects.

[0030] (2) The present invention can simultaneously generate target images at various locations by optimizing the imaging algorithm, and then perform denoising, splicing and reconstruction.

[0031] (3) The present invention realizes high signal-to-noise ratio imaging of smaller targets in a large field of view through a scanning mirror and a DMD, which is conducive to the real-time capture and monitoring of moving objects. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 A schematic diagram of the working principle of a single-pixel imaging system based on a DMD and a scanning mirror provided in a specific embodiment of the present invention;

[0033] Figure 2 A single pixel imaging spatial structure diagram based on DMD and scanning mirror provided in a specific embodiment of the present invention

[0034] Figure 3 An electronic control system solution for a single pixel imaging system based on a DMD and a scanning mirror is provided in a specific embodiment of the present invention;

[0035] Figure 4 A flowchart of a single-pixel imaging method based on a DMD and a scanning mirror is provided for a specific embodiment of the present invention.

[0036] Figure 5 (a) is the adaptation diagram of the scanning mirror on the DMD. Figure 5 (b) is the structural diagram of DMD and scanning mirror Figure 6 A single-pixel imaging experiment diagram based on DMD and scanning mirror is provided for a specific embodiment of the present invention, wherein Figure (a) is the original image, Figure (b) is the imaging experiment diagram using DMD alone, and Figure (c) is the imaging experiment diagram of the linkage control of DMD and scanning mirror.

[0037] In all the drawings, the same figure numbers are used to represent the same elements or structures, including: 1 DMD controller, 2 TCSPC system, 3 single-photon detector, 4 NUC processor, 5 MCU microcontroller, 6 scanning mirror controller, 7 DAC control module, 8 converging lens, 9 fiber-coupled mirror, 10 DMD, 11 imaging lens, 12 scanning mirror, 13 laser, 14 target object. DETAILED DESCRIPTION

[0038] In order to facilitate the understanding of the present invention, the present invention is described in more detail below in conjunction with the accompanying drawings and specific embodiments. The preferred embodiments of the present invention are shown in the accompanying drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described in this specification. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] The present invention designs a single-pixel imaging system based on the linkage control of DMD and scanning mirror, and combines the algorithm to achieve fast and high-resolution image reconstruction of small targets in a large field of view.

[0040] Figure 1 It is a schematic diagram of the working principle of a super-resolution, large-field-of-view single-pixel imaging system based on DMD and a scanning mirror in an embodiment of the present invention. It can be seen from the figure that the main components include: DMD controller 1, TCSPC system 2, single-photon detector 3, NUC processor 4, MCU single-chip computer 5, scanning mirror controller 6, DAC control module 7, converging lens 8, fiber-coupled mirror 9, DMD 10, imaging lens 11, scanning mirror 12, laser 13, and target object 14.

[0041] In the embodiment of the present invention, the connection relationship and function of each component are as follows: the laser 13 emits laser light to irradiate the target object 14, and the reflected echo signal passes through the imaging lens 11 to the scanning mirror 12. The MCU single chip 5 controls the DAC control module 7 to output analog voltage to the scanning mirror controller 6, and calculates the appropriate scanning step length and scanning step number according to the distance between the DMD 10 and the scanning mirror 12. The scanning mirror 12 scans the image on the target object in a "Z" shape to the DMD 10 for image reconstruction in turn. The light signal reflected from the DMD 10 is coupled into the fiber coupling mirror 9 through the converging lens 8, so that the optical signal The signal is connected to the single-photon detector 3 through an optical fiber. The single-photon detector 3 processes the optical signal into a photon counting signal. The TCSPC system 2 collects the photon counting signal. Every time a mask on the DMD10 is played, the trigger signal on the DMD controller 1 will generate a marking signal to the TCSPC system 2. Each synchronization signal displays a flag bit of "1" in the collected data; the collected photon data is output to the NUC processor 4 via USB3.0. The optimized TVAL3 algorithm simultaneously reconstructs multiple sets of data in a large field of view and performs denoising and splicing to achieve high-resolution, large-field-of-view imaging of the target object.

[0042] In the embodiment of the present invention, the scanning mirror and the DMD in the optical path system can realize real-time alignment between the image reflected by the scanning mirror and the DMD by calculating the known distance, without distortion or overlap of the boundaries. At the same time, the optimized TVAL3 algorithm reconstructs multiple sets of scanned data at the same time, thereby realizing large field of view, fast speed, high resolution and fully automatic imaging at different distances.

[0043] In the embodiment of the present invention, a laser 13 with a wavelength of 488 nm is used.

[0044] In the embodiment of the present invention, the wavelength detection range of the single photon detector 3 is 400-1060nm, and the dark count is 50cps.

[0045] In the embodiment of the present invention, the time resolution of the TCSPC system 2 is 16 ps, and the maximum count value of the unit is 65535.

[0046] In the embodiment of the present invention, the flip frequency of DMD10 is 27995 Hz, and the scale of the micromirror array is 1024×768.

[0047] In the embodiment of the present invention, the minimum rotation angle of a single dimension of the scanning mirror 12 is 1 μrad, and the scanning range is ±10 mrad.

[0048] In the embodiment of the present invention, the accuracy of the output voltage of the DAC control module 7 is 1 mV, and it can output an analog voltage signal of ±10V.

[0049] Figure 3 An electronic control system solution is provided for the specific embodiment system of the present invention. It can be seen from the figure that the electronic control system is mainly composed of two parts. First, the MCU single-chip computer uses SPI to control the DAC module to give the scanning mirror controller voltage signal control to realize scanning the image from the scanning mirror to the DMD. Each time the mask on the DMD is played, a synchronization pulse signal is output to the TCSPC. The TCSPC system marks the scanning process of the target object and communicates with the NUC processor through USART. The NUC processor mainly sends synchronization signals to the MCU single-chip computer through USART, and receives the photon counting information recorded by the TCSPC system through USB3.0, and processes it according to the compressed sensing logic combined with the optimized TVAL3 algorithm to realize the imaging of the target object and display it.

[0050] Among them, the control process in completing a scanning imaging process is as follows Figure 4As shown. First, initialize the DMD and set the flip frequency of the DMD to 128Hz. For a set of 512 Hadamard masks, it takes 4 seconds to play the complete video. The playback mode is internal loop playback, and the trigger mode is the rising edge. Then initialize the TCSPC and set the impedance to 50Ω to improve the transmission efficiency of the electrical signal and reduce signal interference. The threshold level is 1000mV, and the trigger mode is the rising edge, which is consistent with the trigger mode of the DMD to facilitate data collection. Then set the angle of the scanning mirror to match the DMD, the DMD starts to reset and play, and then the scanning mirror starts scanning. The target size of the DMD is imaged on the DMD in turn according to the calculated scanning step length and scanning step number. The scanning mirror outputs a ±10V voltage through the DAC module on the MCU to control its deflection angle ±10mrad to control the area of ​​the target object irradiated on the DMD, as shown in Figure 5 (a) As shown. The laser reflected by the DMD is collected by the receiving device and processed to reconstruct the image. Figure 5 The parameters of the simple optical system DMD and scanning mirror shown in (b), such as the lens parameter f, the deflection angle θ, the deflection voltage U of the scanning mirror and the adaptation range d of the DMD, can be calculated to determine the mathematical logical relationship between them.

[0051]

[0052] Then, the mathematical relationship between the adaptation range d of the DMD and the deflection voltage U can be calculated.

[0053] d=Ltan0.573U (3)

[0054] When we obtain the mathematical relationship between various parameters, we can easily control the scanning mirror accurately so that it is projected onto the DMD, thus avoiding the influence of errors and external interference.

[0055] Evaluate the image quality and get the best scanning position as the calibration position. The optical signal is collected by the single-photon detector and output to the TTL level to TCSPC. The dark count rate is 50c / s, the time jitter is about 500ps, and the dead time is 50ns. Open the file to read the data and determine whether the number of flags is an integer multiple of 512. If not, end the process because there are 512 masks in total and a set must be played completely to form an image. If so, use flag 1 as the start signal and the last 1 as the end signal. According to the compressed sensing algorithm, the imaging matrix is ​​obtained and normalized to obtain the reconstructed image. After the image is denoised, the images are spliced ​​and combined into a large image in the scanning order to achieve imaging of the target object and display it, ending the scanning imaging.

[0056] Figure 6This is the experimental result. For the picture with TYUT printed on it, the DMD and scanning mirror are used to jointly image. The original image is 256×256 pixels. Figure 6 As shown in (a), the final reconstruction is obtained after splicing. Figure 6 The 512×512 pixel image shown in (c) was sampled 512 times in total, with a sampling rate of 0.8%. Statistics of all the collected data show that the number of photons collected to restore the image is approximately 11637416, which translates to approximately 15 photons used to restore the image per pixel in each measurement. Figure 6 (b) is the image without the scanning mirror, that is, the DMD is used alone for imaging without using the scanning mirror for expansion, and Figure 6 (a) is similar, and the sampling is also 512 times, with a sampling rate of 0.8%. The data is also statistically analyzed to find that the number of photons collected to restore the entire image is about 4376057, forming 4 groups of data, and the number of photons used per pixel is about 17. For imaging targets with the same field of view, the imaging effect of the linkage of DMD and scanning mirror is greatly improved compared with the imaging effect of DMD alone. Under the same conditions, the pixel size is also increased by 4 times. The PSNR of the linkage imaging of DMD and scanning mirror is 20.90dB, and the imaging effect of DMD alone is 10.18dB. The PSNR is increased by about 2 times, realizing linkage control imaging in a large field of view.

[0057] The present invention aims to solve the problem of low imaging resolution in traditional single-pixel imaging systems due to field of view expansion limitations and identification of small target objects in the field of view. The present invention proposes a field of view expansion imaging method based on DMD and scanning mirror, which scans and expands the image reflected on the target object to the DMD, realizes efficient sampling through compressed sensing, and in the process of forming the scanning mirror optimized image, controls the scanning mirror to adapt to the DMD with high precision (the error is controlled within ±3mV), obtains the position of the optimal PSNR, expands the imaging field of view, and uses the optimized TVAL3 algorithm to restore and reconstruct the image, and then splices it. For 2nd-order imaging, the imaging resolution of the target object is improved by 4 times (256×256-512×512) and the signal-to-noise ratio is about 2 times (10.18-20.90dB), which can be extended to nth-order imaging, and the imaging resolution of the target object can be improved to n. 2 This new imaging method will gradually adapt to the existing beam scanning imaging system or wide field of view camera with additional scanning system, and will be conducive to the widespread application of large field of view imaging.

[0058] It should be pointed out that, according to the needs of implementation, the various steps / components described in this application can be split into more steps / components, and two or more steps / components or partial operations of steps / components can be combined into new steps / components to achieve the purpose of the present invention.

[0059] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A single pixel imaging system based on DMD and scanning mirror, characterized in that: The single-pixel imaging system comprises an optical path unit, a control unit, a single-photon detector (3) and a time-correlated single-photon counting system (2), wherein the time-correlated single-photon counting system (2) is defined as TCSPC; specifically: The optical path unit is used to illuminate the entire target object (14) after the laser (13) emits laser light, control the high-precision scanning mirror (12) to scan and expand the image reflected by the target object to the DMD (10) through the imaging lens (11), and collect the target echo light signal, filter it, and then collect and couple the filtered echo light signal into the single-photon detector (3); The control unit obtains the scanning step length and the number of steps according to the distance, and is used to gradually change the position angle information of the scanning mirror (12) until the position with the highest signal-to-noise ratio is reached to achieve fine scanning imaging; The single photon detector (3) is used to detect the target echo signal in the optical fiber coupling mirror (9) and process the received echo signal into a photon counting signal; The TCSPC is used to receive the photon counting signal and measure the number of DMD mask switching under the action of the synchronization signal output by the DMD (10), and then perform data marking and segmentation to achieve image reconstruction.

2. A single pixel imaging system based on DMD and scanning mirror according to claim 1, characterized in that: The optical path unit comprises a laser (13), an imaging lens (11), a scanning mirror (12), a DMD (10), a converging lens (8), and a fiber coupling mirror (9); specifically: The laser (13) emits laser light to illuminate the target object (14); the echo signal reflected by the target object (14) irradiated by the laser (13) is collected through the imaging lens (11) to the scanning mirror (12), and the image of the target object is scanned to the DMD (10); the incident light is spatially modulated by the DMD (10), and the deflection state of each micromirror of the DMD (10) is controlled to achieve precise control of the light, thereby achieving high-resolution imaging; the converging lens (8) is used to couple the output light on the DMD (10) into the coupling mirror; the optical fiber coupling mirror (9) is used to couple the collected echo signal into the single-photon detector (3).

3. A single pixel imaging system based on DMD and scanning mirror according to claim 2, characterized in that: The target surface of the DMD (10) has an angle of 45° with the horizontal direction, and the incident and outgoing light paths are within the same degree of freedom. Each micromirror of the DMD (10) is attached to a support shaft, and the storage unit of each micromirror can be electronically addressed independently.

4. The single pixel imaging system based on DMD and scanning mirror according to claim 1, characterized in that: The control unit comprises a DAC control module (1), a scanning mirror controller (6), a DMD controller (7), an MCU single chip computer (5) and a NUC processor (4); specifically: The DAC control module (1) is located on the MCU single chip (5) and is used to output ±10V analog voltage to control the ±10mrad deflection of the scanning mirror; the scanning mirror controller (6) is connected to the scanning mirror (12) and is used to drive the scanning mirror (12) to move; the DMD controller (7) is respectively connected to the DMD (10) and the NUC processor (4), inputs the mask into the DMD (10), and is used to drive the micromirror deflection rate of the DMD (10), thereby achieving precise control of light, generating high-precision graphics, and outputting a synchronization signal to the TCSP The photon counting signal collected by the TCSPC is marked, and each synchronization signal is displayed as a flag bit "1" in the collected data; the MCU single chip computer (5) obtains the movement step length and number of steps of the scanning mirror (12) from the position of the target object (14), and controls to realize large-field fine scanning of the image of the target object to the DMD (10); the NUC controller receives the photon data collected by the TCSPC, and after completing the scanning of the imaging area of ​​the target object, processes it according to the imaging algorithm, realizes the splicing imaging of the target object and displays it, and realizes fully automatic and fast large-field scanning imaging.

5. The single pixel imaging system based on DMD and scanning mirror according to claim 4, characterized in that: The MCU single-chip computer (5) and the NUC processor (4) in the control unit are used to control the single-pixel imaging system, with the NUC processor (4) serving as the main control and the MCU single-chip computer (5) serving as the auxiliary control, to control the single-pixel imaging system and achieve super-resolution imaging of a target object in a large field of view.

6. A high-resolution field of view expansion method based on the single-pixel imaging system according to any one of claims 1 to 5, characterized in that: The following steps are involved: Step S1, a laser (13) emits laser light to illuminate a target object (14), and an echo signal returns along the original path and converges onto a scanning mirror (12) through an imaging lens (11); In step S2, the MCU single chip computer (5) controls the scanning mirror (12) to scan the echo signal onto the DMD (10), the DMD and the TCSPC are initialized and relevant parameters are set, after the initialization is successful, the echo signal then enters the fiber coupling mirror (9) through the DMD (10), the echo signal is detected by the single photon detector (3), the TCSPC collects the photon counting signal, the collected photon data is analyzed by the NUC processor (4), and the image is denoised, spliced ​​and reconstructed in combination with the optimized imaging algorithm to obtain the final large field of view imaging result image.

7. A high-resolution field of view expansion method according to claim 6, characterized in that: The step S2 comprises the following sub-steps: Step 2.1, the MCU single chip computer (5) first controls the scanning mirror (12) according to the distance between the DMD (10) and the scanning mirror (12), calculates the step length and number of steps required at this time, and then performs a "Z"-shaped scan; in order to obtain the best position of the PSNR of the image formed on the DMD, the voltage is increased by 0.1V each time, the scanning mirror (12) moves the image returned from the target object on the DMD (10) by one position, the DMD (10) performs image reconstruction, and calculates the peak signal-to-noise ratio PSNR of the image formed at this position at this time, and so on, calculates the PSNR of the reconstructed image on the DMD (10) through the movement of the scanning mirror in sequence, takes the position with the highest PSNR as the mark position, and records the voltage value output by the MCU single chip computer (5) at this time; Step 2.2, the target image is divided into n×n blocks, and the scanning mirror (12) stops at n 2 The optimal PSNR of each position is reached, and the dwell time exceeds the time for a set of masks to be played by the DMD. Each time the mask is switched, a trigger signal is output to the TCSPC to mark the data. Step 2.3, using the data collected by TCSPC and the Hadamard mask as inputs of the reconstruction algorithm to adapt them to the spatial imaging of the DMD (10) and the scanning mirror; using the first "1" flag in the collected data as the start signal and the last "1" as the end signal to obtain the imaging matrix, and then performing normalization processing to achieve denoising, splicing and reconstruction of each block image extended from the scanning mirror (12) to the DMD (10); Step 2.4, through step 2.1, step 2.2 and step 2.3, the scanning mirror (12) under the control of the MCU single chip (5) and the DMD (10) ensures the super-resolution imaging effect of the system, obtains a clear field of view, and realizes fully automatic fine high-speed scanning imaging of the system with a large field of view.

8. A high-resolution field of view expansion method according to claim 7, characterized in that: In the step 2.2, the playing time of a set of masks is set to 4 seconds.

9. The high-resolution field of view expansion method according to claim 7, characterized in that: In step 2.3, the scanning mirror (12) and the DMD (10) are configured to have an adaptation range of the DMD (10) according to the following formula: Adaptation range Wherein, d represents the DMD (10) adaptation range, L represents the target object distance, θ represents the maximum deflection angle of the scanning mirror, V represents the maximum output voltage of the DAC module, and U represents the given input voltage.

10. A high-resolution field of view expansion method according to claim 9, characterized in that: In step 2.3, after the range is adapted, the target image is reconstructed, which is expressed as follows: Image information to be tested X = A -1 Y=A -1 AO=O Among them, O is the target vector, size N×1, which is composed of the target object through N mask vectors arranged in columns; A is the mask observation matrix and is a reversible matrix, size N×N, and the observation matrix A has a reversible matrix A -1 ; Y is the measurement value vector, size is N×1.

Images