CN109361833B - Transmission method of single photon compression video transmission device - Google Patents

Abstract

The invention relates to the field of weak light communication, in particular to a transmission device of a single photon compression video, which comprises a sending end and a receiving end, wherein the sending end and the receiving end are communicated through optical fibers, the sending end comprises an imaging lens, a compression modulator, a DMD (digital micromirror device) controller and a first focusing lens, and light irradiates on an object to be transmitted and is imaged on the DMD in real time through the imaging lens and the imaging lens in sequence; the receiving end comprises a second focusing lens, a PMT single photon detector, a compression demodulator and a computer, and the optical signal transmitted by the optical fiber is collected into the PMT single photon detector through the second focusing lens. The invention is based on single photon detection technology, the detector works in a photon technology mode, and the information is demodulated by the number of single photon pulses output by the detector, so that the invention has high detection sensitivity, and can realize remote communication with lower power consumption.

Description

A transmission method of a single-photon compressed video transmission device

technical field

The invention relates to the field of weak optical video communication, in particular to a transmission method of a single-photon compressed video transmission device in the field of weak optical communication.

Background technique

Single-photon imaging is a method to achieve extremely weak light imaging by detecting and counting single photons. It has been widely used in biomedical imaging, night vision, lidar, and astronomical spectroscopy. In order to achieve high-sensitivity detection imaging, several kinds of detectors with spatial resolution have been developed, such as intensified CCD (ICCD), electron multiplying CCD (EM-CCD), and APD array. ICCDs and EM-CCDs operating in photon counting mode require very high frame rates and very low circuit noise, so deep cooling and very high cost are required. The resolution of current APD arrays is still low due to fabrication difficulties and unstable performance. Another high-resolution imaging method is to use point detectors, such as avalanche photodiodes (APDs) operating in Geiger mode, single-photon sensitive photomultiplier tubes (PMTs), or small APD arrays for scanning imaging, but due to photons The collection efficiency is low and the imaging time is very long.

Single-pixel imaging based on compressed sensing (CS) theory provides a new solution to the above problems. In the single-pixel imaging scheme, the target is imaged on a digital micromirror device (DMD), which is modulated by the DMD and focused on a single-point detector. A two-dimensional image can be reconstructed using a series of light intensity values detected by a single-point detector and a measurement matrix loaded into the DMD. Due to the CS sampling theory, the number of measurements and imaging time can be significantly reduced. In 2012, Yu Wenkai et al. proposed a single-photon compression imaging scheme using a single-photon detector on the basis of single-pixel imaging. And it is proved theoretically and experimentally that this scheme has higher sensitivity compared with the traditional multi-pixel imaging scheme.

Compared with a single image, the sequence of video time-continuous still images can describe objective things more vividly, vividly and in real time, so it has been widely used. The general method of video transmission is to use software to compress multiple images after collecting multiple images at the sending end, generate video file data, and then use classic communication means, such as mobile communication network, bluetooth, wifi or video cable, It is transmitted to the receiving end by means of USB cable, network cable, etc.

Based on the single-photon compression imaging technology, this project proposes a device and method for single-photon compressed video transmission, which directly compresses and modulates the image imaged on the DMD, and transmits the modulated weak light to the video through the optical fiber. At the receiving end, single-photon detection and counting are performed at the receiving end, and then the frame-by-frame signal reconstruction is performed using the compressed sensing theory, so as to realize the transmission of imaging video under extremely weak light conditions. In order to realize the video transmission, the DMD modulation matrix and the demodulation of the single-photon pulse signal are specially designed.

SUMMARY OF THE INVENTION

The purpose of the present invention is to design a transmission method of a single-photon compressed video transmission device in order to realize the transmission of compressed imaging video under extremely weak light conditions.

In order to realize the purpose of the present invention, the technical means adopted in the present invention are:

A transmission device for single-photon compressed video, including a sending end and a receiving end, the sending end and the receiving end communicate through an optical fiber, and the sending end includes an imaging lens, an imaging lens, a modulator, a DMD controller, a DMD, a first a focusing lens, wherein an imaging lens, an imaging lens and a DMD are sequentially arranged on the imaging optical path of the object at the sending end, and the first focusing lens is arranged on the reflected optical path of the DMD;

The modulator is connected with the DMD controller, and the generated random measurement matrix is output to the DMD controller, and the DMD controller is connected with the DMD;

The modulator comprises a DMD inversion control signal generator, an SDRAM, a matrix loading module, and an SDRAM connected in sequence, the matrix loading module is connected with the matrix generating module, and the matrix loading module outputs a random measurement matrix;

The receiving end includes a second focusing lens, a single photon detector, a demodulator and a PC arranged in sequence;

The demodulator includes a pulse stretching module, a frame header identification module, a gated signal generation module, a gated photon counting module, a matrix generation module, a first USB interface communication module and a second USB interface communication module;

The input signal of the pulse stretching module is the single-photon pulse signal output by the single-photon detector; the pulse stretching module is respectively connected with the frame header identification module and the gated photon counting module, and the frame header identification module and the gated signal are generated. The modules are connected, the gated signal generation module is connected with the gated photon counting module, the gated photon counting module is connected with the first USB interface communication module, the first USB interface communication module is connected with the PC, and the matrix generation module is connected with the The second USB interface communication module is connected, and the second USB interface communication module is connected with the PC.

The frame header identification module includes a pulse envelope extraction module, a square wave width measurement module, a first threshold register, a second threshold register, and a square wave width comparator; the pulse envelope extraction module is connected with the square wave width measurement module; The square wave width measurement module, the first threshold value register, and the second threshold value register are respectively connected with the square wave width comparator.

A method for transmitting single-photon compressed video, comprising the following steps:

1) Set the matrix generation parameters and sampling parameters to generate the measurement matrix:

Set the matrix generation parameters of the matrix generation module in the sending end and the receiving end. The matrix generation parameters include but are not limited to the matrix size P*Q, the measurement matrix type, the sampling times m, and the sampling frequency f; the measurement matrix generated by the sending end will store into the SDRAM; the measurement matrix generated by the receiving end will be sent to the PC end through the second USB interface communication module;

2) The imaging object is imaged on the DMD mirror surface through the imaging lens and the imaging lens;

3) Modulate the image on the DMD mirror surface;

3.1) Detect the rising edge of the Start signal, and proceed to the next step if the detection is successful, otherwise wait;

3.2) The DMD inversion control signal generator of the modulator outputs a group of DMD inversion control signals, and the described group of DMD inversion control signals are equally spaced pulse signals, the frequency of the modulating signal is equal to the preset sampling frequency, and the number of pulses is the preset sampling frequency. Set the sampling times m+3;

3.3) Transmit the frame header of a frame of image

When the matrix loading module of the modulator detects the rising edge of the first pulse of a group of DMD inversion control signals, it loads the matrix with all "1" to the DMD controller, and detects the second pulse of a group of DMD inversion control signals. On the rising edge, load the matrix with all "0" to the DMD controller;

3.4) Perform m compression sampling on the image on the DMD mirror

The matrix loading module of the modulator detects m pulses after the second pulse of a group of DMD inversion control signals, and every time a rising edge of a pulse is detected, the matrix loading module reads a measurement matrix from the SDRAM to the DMD controller;

3.5) Transmit the end of a frame of image

When the matrix loading module of the modulator detects the rising edge of the last pulse of a group of DMD inversion control signals, it loads the matrix with all "0" to the DMD controller;

3.6) After receiving the matrix, the DMD controller controls the micromirror on the DMD to flip, and the first focusing lens 106 collects the reflected light in the +12 degree direction of the DMD micromirror and enters the optical fiber, and reaches the receiving end after long-distance transmission;

3.7) Repeat 3.2)-3.6) above, and send images continuously until the rising edge of the stop signal is detected;

4) The second converging lens at the receiving end converges the light output from the optical fiber to the single-photon detector, and the single-photon detector converts the optical signal into a single-photon pulse signal;

5) Demodulate the single-photon pulse signal

5.1) The single-photon pulse signal is input to the pulse stretching module for stretching;

5.2) Input the stretched single-photon pulse signal into the frame header identification module and the gated photon counting module at the same time;

5.3) After the frame header identification module successfully identifies the frame header of a frame of image, it outputs a start pulse signal to the gate control signal generation module;

5.4) After detecting the rising edge of the start pulse signal, the gated signal generation module outputs a group of gated signals to the gated photon counting module; the gated signal is an equally spaced pulse signal, and the frequency of the signal is equal to the preset sampling Frequency f, the number of pulses is the preset sampling times m+1;

5.5) When the gated photon counting module detects the rising edge of the gated signal; it judges whether it is the first synchronous control pulse of a group of gated signals, if so, clear the counter and input the gated photon counting module The single-photon pulse stretched signal counts from 0 again; if not, the pulse and the stretched single-photon pulse signal count in the previous pulse interval are output to the PC through the second USB interface communication module, and the counter is cleared at the same time. The stretched single-photon pulse counts from 0 again;

6) The PC stores a series of single-photon pulse count values received in step 5.5) and the measurement matrix received in step 1), and reconstructs the image using compressed sensing theory;

7) Repeat steps 5) and 6) to continuously reconstruct multiple frames of images, and use the multiple frames of images to reconstruct a video.

The step of identifying the frame header includes:

1) The pulse envelope extraction module extracts the envelope of the stretched single-photon pulse signal. The extraction method is to combine the pulses in the stretched single-photon pulse signal with a relatively small pulse interval into a square wave pulse, and the stretched single-photon pulse signal becomes the envelope square. wave pulse signal;

2) The square wave width measurement module measures the width of the square wave pulse in the envelope square wave pulse signal, and outputs the measured value to the square wave width comparator. The measurement method is to measure the high frequency clock when the rising source of the square wave pulse is detected. Counting starts from zero, and stops counting when the falling source of the square wave pulse is detected. At this time, the count value of the high-frequency clock represents the width of the square wave pulse;

3) The square wave width comparator compares the measured value input by the square wave width measurement module with the values in the first threshold register and the second threshold register. If the measured value is greater than the first threshold register and less than the second threshold register, it means that the frame The head recognition is successful, and the start pulse signal is output to the gate control signal generation module, the value in the first threshold register is greater than the width value of a single-photon pulse after being stretched, and the value in the second threshold register is greater than 1/2 times The DMD toggle control signal period value is less than 2 times the DMD toggle control signal period.

Beneficial effects of the present invention:

1. The present invention adopts a single-photon compression imaging method based on single-photon detection technology and single-pixel camera technology, which has extremely high detection sensitivity and can realize video transmission under extremely weak light imaging conditions.

2. The transmitting end of the video transmission device of the present invention does not perform image sampling and reconstruction, so it does not require image compression, but directly transmits the modulated light to the receiving end, and then uses the compressed sensing theory to reconstruct the image after photon detection and counting at the receiving end. The complex encoding and decoding process of the traditional video transmission system is eliminated.

3. The receiving end of the present invention uses a single-photon detector to detect the optical signal transmitted by the optical fiber, and has very high sensitivity, so it can realize ultra-long-distance video transmission.

Description of drawings

1 is a schematic structural diagram of a single-photon compressed video transmission device of the present invention;

Fig. 2 is the modulator structure block diagram of the single-photon compressed video transmission device of the present invention;

3 is a block diagram of the demodulator structure of the single-photon compressed video transmission device of the present invention;

Fig. 4 is the frame header identification module structural block diagram of the present invention;

Fig. 5 is the image modulation sequence diagram of the transmitting end of the present invention;

Fig. 6 is the demodulation sequence diagram of the receiving end of the present invention;

FIG. 7 is a timing diagram of frame header signal extraction according to the present invention.

Detailed ways

Example: See Figures 1-7.

The present invention discloses a single-photon compressed video transmission device, as shown in FIG. 1 , including a sending end and a receiving end, the sending end 1 and the receiving end 2 communicate through an optical fiber 3, and the sending end 1 includes an imaging lens 101. Imaging lens 102, modulator 103, DMD controller 104, DMD 105, first focusing lens 106, the imaging lens 101, imaging lens 102 and DMD 105 are sequentially arranged on the imaging optical path of the object at the sending end 1, the first focusing lens The lens 106 is arranged on the reflected light path of the DMD105;

The modulator 103 is connected to the DMD controller 104, and outputs the generated random measurement matrix to the DMD controller, and the DMD controller is connected to the DMD;

The modulator 103 includes a DMD inversion control signal generator 1031, an SDRAM 1034, a matrix loading module 1032, and an SDRAM 1034 connected in sequence. The matrix loading module 1032 is connected to the matrix generating module 1033, and the matrix loading module 1032 outputs a random measurement matrix.

As shown in FIG. 3 , the receiving end includes a second focusing lens 201 , a single-photon detector 202 , a demodulator 203 and a PC 204 arranged in sequence; the demodulator 203 includes a pulse stretching module 2031 and a frame header identifying module 2032 , gated signal generation module 2033, gated photon counting module 2034, matrix generation module 2035, first USB interface communication module 2036 and second USB interface communication module 2037; the input signal of the pulse stretching module 2031 is a single photon detector The single-photon pulse signal output by 202; the pulse stretching module is respectively connected with the frame header identification module 2032 and the gated photon counting module 2034, and the frame header identification module 2032 is connected with the gated signal generation module 2033, and the gated signal The generation module 2033 is connected with the gated photon counting module 2034, the gated photon counting module 2034 is connected with the first USB interface communication module 2036, the first USB interface communication module 2036 is connected with the PC 204, and the matrix generation module 2035 is connected with the second USB interface communication module 2036. The USB interface communication module 2037 is connected, and the second USB interface communication module 2037 is connected to the PC 204 . The frame header identification module 2032 includes a pulse envelope extraction module 20321, a square wave width measurement module 20322, a first threshold register 20323, a second threshold register 20324, and a square wave width comparator 20325; the pulse envelope extraction module 20321 and The square wave width measurement module 20322 is connected; the square wave width measurement module 20322 , the first threshold value register 20323 and the second threshold value register 20324 are respectively connected to the square wave width comparator 20325 .

When the device is working, at the sending end, each frame of image of the scene to be tested is imaged on a digital micromirror device (Digital Micromirror Device, DMD) in real time through an imaging lens and an imaging lens in turn. After the parameter setting is completed, the compressed video transmission starts. For each frame of the video, the single-photon compressed video transmission control device loads m+3 random measurement matrices to the DMD, of which 2 frame measurement matrices (all "1" matrix and all "0" Matrix) as the frame header, and a full "0" matrix for 1 frame as the frame tail. Each micromirror of the DMD can independently achieve ±12° deflection under the control of the loaded random binary matrix. In the experiment, a first focusing lens is set in the +12° reflection direction of the DMD micromirror, and the micromirror reflects light through the lens. Collect and couple into the optical fiber for long-distance transmission; at the receiving end, the optical signal transmitted by the optical fiber is collected into the single-photon detector through the focusing lens, and after the frame header signal is extracted by the demodulator receiving end, the discrete output of the detector is modulated for each time. The single-photon pulse counts photons, and sends each photon count value as a measurement value to the PC; after the frame end signal is successfully extracted, it means that the transmission of one frame of compressed image is completed, and then the next frame of compressed image is extracted. The PC performs frame-by-frame restoration of the compressed sensing image according to the received measurement value sequence and measurement matrix, and finally forms a video stream.

In this case, SDRAM: Synchronous Dynamic Random Access Memory, synchronous dynamic random access memory, synchronization means that the memory operation requires a synchronous clock, and the sending of internal commands and data transmission are based on it; dynamic means that the storage array needs to be constantly refreshed to ensure The data is not lost; random means that the data is not stored in a linear order, but freely specifies the address for data read and write. DMD: Digital Micromirror Device, a digital micromirror device, is a kind of optical switch, which uses a rotating mirror to realize the opening and closing of the optical switch. The DMD is used to perform random spatial modulation on each frame of image in the video to be transmitted according to the 0-1 random mask loaded by the measurement matrix loading module 132 .

3. a transmission method of single photon compressed video, is characterized in that, comprises the following steps:

1) Set the matrix generation parameters and sampling parameters to generate the measurement matrix:

Set the matrix generation parameters of the matrix generation module in the sending end 1 and the receiving end 2, the matrix generation parameters include but are not limited to the matrix size P*Q, the measurement matrix type, the sampling times m, and the sampling frequency f; the measurement generated by the sending end 1 The matrix will be stored in the SDRAM; the measurement matrix generated by the receiving end will be sent to the PC204 end through the second USB interface communication module 2037;

2) The imaging object is imaged on the mirror surface of the DMD105 through the imaging lens 101 and the imaging lens 102;

3) The image on the mirror surface of DMD105 is modulated, and its working sequence is shown in Figure 5, which specifically includes the following steps;

3.1) Detect the rising edge of the Start signal, and proceed to the next step if the detection is successful, otherwise wait;

3.2) The DMD inversion control signal 1031 generator of the modulator 103 outputs a group of DMD inversion control signals, the group of DMD inversion control signals are equally spaced pulse signals, the frequency of the modulation signal is equal to the preset sampling frequency, the number of pulses is the preset sampling times m+3;

3.3) Transmit the frame header of a frame of image

When the matrix loading module of the modulator 103 detects the rising edge of the first pulse of a group of DMD inversion control signals, it loads a matrix of all "1"s to the DMD controller, and detects the second pulse of a group of DMD inversion control signals. At the rising edge of the pulse, load the matrix with all "0" to the DMD controller;

3.4) Perform m compression sampling on the image on the DMD mirror

The matrix loading module 1032 of the modulator 103 detects m pulses after the second pulse of a set of DMD inversion control signals, and each time a rising edge of a pulse is detected, the matrix loading module 1032 reads a measurement matrix from the SDRAM to the DMD controller;

3.5) Transmit the end of a frame of image

When the matrix loading module 1032 of the modulator 103 detects the rising edge of the last pulse of a group of DMD inversion control signals, it loads the matrix with all "0"s to the DMD controller;

3.6) After receiving the matrix, the DMD controller 104 controls the micromirror on the DMD 105 to flip. The first focusing lens 106 collects the reflected light in the +12 degree direction of the DMD micromirror into the optical fiber, and reaches the receiving end 2 after long-distance transmission;

3.7) Repeat the above (3.2-3.6), and send images continuously until the rising edge of the stop signal is detected;

4) the second converging lens 201 of the receiving end 2 converges the light output from the optical fiber to the single-photon detector 202, and the single-photon detector 202 converts the optical signal into a single-photon pulse signal;

5) Demodulate the single-photon pulse signal

5.1) The single-photon pulse signal is input to the pulse stretching module 2031 for stretching;

5.2) Input the stretched single-photon pulse signal into the frame header identification module 2032 and the gated photon counting module 2034 simultaneously;

5.3) After the frame header identification module 2032 successfully identifies the frame header of a frame of image, it outputs a start pulse signal to the gate control signal generation module 2033; the working sequence of the frame header identification is shown in Figure 7, and the specific steps are as follows:

5.3.1) The pulse envelope extraction module extracts the envelope of the stretched single-photon pulse signal. The extraction method is to combine the pulses in the stretched single-photon pulse signal with a relatively small pulse interval into a square wave pulse, and the stretched single-photon pulse signal becomes a packet. network square wave pulse signal;

5.3.2) The square wave width measurement module measures the width of the square wave pulse in the envelope square wave pulse signal, and outputs the measured value to the square wave width comparator. The high-frequency clock starts counting from zero, and stops counting when the falling source of the square wave pulse is detected. At this time, the count value of the high-frequency clock represents the width of the square wave pulse.

5.3.3) The square wave width comparator compares the measured value input by the square wave width measurement module with the values in the first threshold register and the second threshold register. If the measured value is greater than the first threshold register and less than the second threshold register, then Indicates that the frame header is successfully recognized, and outputs a start pulse signal to the gate control signal generation module. 2 times the DMD toggle control signal period value and less than 2 times the DMD toggle control signal period.

5.4) After the gated signal generation module 2033 detects the rising edge of the start pulse signal, it outputs a group of gated signals to the gated photon counting module 2034; the gated signal is an equally spaced pulse signal, and the frequency of the signal is equal to the preset The sampling frequency f, the number of pulses is the preset sampling times m+1;

5.5) When the gated photon counting module detects the rising edge of the gated signal; it judges whether it is the first synchronous control pulse of a group of gated signals, if so, clear the counter and input the gated photon counting module The single-photon pulse stretched signal counts from 0 again; if not, the pulse and the stretched single-photon pulse signal count in the previous pulse interval are output to the PC through the second USB interface communication module 6, and the counter is cleared at the same time. The input stretched single-photon pulse starts counting from 0 again;

6) The PC stores a series of single-photon pulse count values received in step 5.5 and the measurement matrix received in step 1, and reconstructs the image using compressed sensing theory;

The reconstruction method of each frame of image is to take the measurement matrix as Φ and the photon count value as the measurement value y, and solve the convex optimization problem shown in Equation (1) to restore the original image x with high probability and high precision:

7) Repeat steps 5 and 6, reconstruct multiple frames of images continuously, and reconstruct a video by using the multiple frames of images.

In this case, the sending-end matrix generating module and the receiving-end matrix generating module are set to the same parameters, so the generated matrices are also the same. The generated matrix consists of three parts: frame header matrix, m-time measurement matrix and frame tail matrix. The frame header matrix includes the first all-1 matrix and the second all-zero matrix of each group of matrices. The frame tail The matrix is the last all "0" matrix. The frame header matrix and the frame tail matrix are only used for image modulation at the sending end so that the receiving end can extract and demodulate, and only the m-times measurement matrix is used for frame-by-frame reconstruction of the image at the receiving end.

The above descriptions are only the embodiments of the present invention, and are not intended to limit the scope of the patent of the present invention. All equivalent transformations made by using the contents of the description and drawings of the present invention or directly or indirectly applied in the relevant technical fields are similarly included in the present invention. The invention is within the scope of patent protection.

Claims (2)

1. A transmission method of a single-photon compressed video transmission device, the device comprises a sending end and a receiving end, the sending end (1) and the receiving end (2) communicate through an optical fiber (3), The sending end (1) includes an imaging lens (101), an imaging lens (102), a modulator (103), a DMD controller (104), a DMD (105), and a first focusing lens (106). At the end (1), an imaging lens (101), an imaging lens (102) and a DMD (105) are arranged in sequence on the imaging optical path of the object, and the first focusing lens (106) is arranged on the reflected optical path of the DMD (105); The modulator (103) is connected to the DMD controller (104), and outputs the generated random measurement matrix to the DMD controller, and the DMD controller is connected to the DMD; The modulator (103) includes a DMD inversion control signal generator (1031), an SDRAM (1034), a matrix loading module (1032), and an SDRAM (1034) connected in sequence, the matrix loading module (1032) and the matrix generating module (1033) connect, the matrix loading module (1032) outputs a random measurement matrix; The receiving end includes a second focusing lens (201), a single-photon detector (202), a demodulator (203) and a PC (204) arranged in sequence; The demodulator (203) includes a pulse stretching module (2031), a frame header identification module (2032), a gated signal generation module (2033), a gated photon counting module (2034), a matrix generation module (2035), a A USB interface communication module (2036) and a second USB interface communication module (2037); The input signal of the pulse stretching module (2031) is the single-photon pulse signal output by the single-photon detector (202); the pulse stretching module is respectively connected with the frame header identification module (2032) and the gated photon counting module (2034) , the frame header identification module (2032) is connected with the gated signal generation module (2033), the gated signal generation module (2033) is connected with the gated photon counting module (2034), the gated photon counting module ( 2034) is connected to the first USB interface communication module (2036), the first USB interface communication module (2036) is connected to the PC (204), and the matrix generation module (2035) is connected to the second USB interface communication module (2037), The second USB interface communication module (2037) is connected to the PC (204); the frame header identification module (2032) includes a pulse envelope extraction module (20321), a square wave width measurement module (20322), and a first threshold register (20323), a second threshold register (20324), a square wave width comparator (20325); the pulse envelope extraction module (20321) is connected to the square wave width measurement module (20322); the square wave width measurement module (20322) , the first threshold register (20323), and the second threshold register (20324) are respectively connected with the square wave width comparator (20325), characterized in that the method includes the following steps: 1) Set the matrix generation parameters and sampling parameters to generate the measurement matrix: Set the matrix generation parameters of the matrix generation module in the transmitting end (1) and the receiving end (2). The matrix generating parameters include but are not limited to the matrix size P*Q, the measurement matrix type, the sampling times m, and the sampling frequency f; the transmitting end (1) The generated measurement matrix will be stored in the SDRAM; the measurement matrix generated by the receiving end will be sent to the PC (204) end through the second USB interface communication module (2037); 2) The imaging object is imaged on the mirror surface of the DMD (105) through the imaging lens (101) and the imaging lens (102); 3) Modulate the image on the mirror surface of DMD (105); 3.1) Detect the rising edge of the Start signal, if the detection is successful, proceed to the next step, otherwise wait; 3.2) The DMD inversion control signal (1031) generator of the modulator (103) outputs a group of DMD inversion control signals, the group of DMD inversion control signals are pulse signals at equal intervals, and the frequency of the modulation signal is equal to the preset sampling frequency , the number of pulses is the preset sampling times m+3; 3.3) Transmit the frame header of a frame of image When detecting the rising edge of the first pulse of a group of DMD inversion control signals, the matrix loading module of the modulator (103) loads a matrix of all "1"s to the DMD controller, and detects the first pulse of a group of DMD inversion control signals. At the rising edge of the two pulses, load the matrix with all "0" to the DMD controller; 3.4) Perform m compression sampling on the image on the DMD mirror The matrix loading module (1032) of the modulator (103) detects m pulses after the second pulse of a group of DMD inversion control signals, and every time a rising edge of a pulse is detected, the matrix loading module (1032) reads from the SDRAM Take a measurement matrix to the DMD controller; 3.5) Transmit the end of a frame of image When the matrix loading module (1032) of the modulator (103) detects the rising edge of the last pulse of a group of DMD inversion control signals, it loads the matrix with all "0"s to the DMD controller; 3.6) After receiving the matrix, the DMD controller (104) controls the micromirror on the DMD (105) to flip, and the first focusing lens 106 collects the reflected light in the +12 degree direction of the DMD micromirror and enters the optical fiber, and reaches the receiver after long-distance transmission. end(2); 3.7) Repeat the above 3.2)-3.6), and continuously send images until the rising edge of the stop signal is detected; 4) The second converging lens (201) at the receiving end (2) converges the light output from the optical fiber to the single-photon detector (202), and the single-photon detector (202) converts the optical signal into a single-photon pulse signal; 5) Demodulate the single-photon pulse signal 5.1) The single-photon pulse signal is input to the pulse stretching module (2031) for stretching; 5.2) Input the stretched single-photon pulse signal into the frame header identification module (2032) and the gated photon counting module (2034) at the same time; 5.3) After the frame header identification module (2032) successfully identifies the frame header of a frame of image, it outputs a start pulse signal to the gate control signal generation module (2033); 5.4) The gated signal generation module (2033) outputs a set of gated signals to the gated photon counting module (2034) after detecting the rising edge of the start pulse signal; The frequency is equal to the preset sampling frequency f, and the number of pulses is the preset sampling times m+1; 5.5) When the gated photon counting module detects the rising edge of the gated signal, it will judge whether it is the first synchronous control pulse of a group of gated signals. If so, clear the counter and input the gated photon counting module. The single-photon pulse stretched signal counts from 0 again; if not, the pulse and the stretched single-photon pulse signal count in the previous pulse interval are output to the PC through the second USB interface communication module (6), and the counter is cleared at the same time , start counting from 0 again for the input stretched single-photon pulse; 6) The PC stores a series of single-photon pulse count values received in step 5.5) and the measurement matrix received in step 1), and reconstructs the image using compressed sensing theory; 7) Repeat steps 5) and 6) to continuously reconstruct multiple frames of images, and use the multiple frames of images to reconstruct a video. 2. the transmission method of a kind of single-photon compressed video transmission device according to claim 1, is characterized in that, the step of described frame header identification comprises: 1) The pulse envelope extraction module extracts the envelope of the stretched single-photon pulse signal. The extraction method is to combine the pulses in the stretched single-photon pulse signal with a relatively small pulse interval into a square wave pulse, and the stretched single-photon pulse signal becomes the envelope square. wave pulse signal; 2) The square wave width measurement module measures the width of the square wave pulse in the envelope square wave pulse signal, and outputs the measured value to the square wave width comparator. The measurement method is to measure the high frequency clock when the rising source of the square wave pulse is detected. Counting starts from zero, and stops counting when the falling source of the square wave pulse is detected. At this time, the count value of the high-frequency clock represents the width of the square wave pulse; 3) The square wave width comparator compares the measured value input by the square wave width measurement module with the values in the first threshold register and the second threshold register. If the measured value is greater than the first threshold register and less than the second threshold register, it means that the frame If the head is recognized successfully, a start pulse signal is output to the gate control signal generation module, the value in the first threshold register is greater than the width value of a single photon pulse after being stretched, and the value in the second threshold register is greater than 1/2 times The DMD toggle control signal period value is less than 2 times the DMD toggle control signal period.

Images