Underwater single photon laser radar imaging method and system based on gating SPAD array

Abstract

The invention discloses an underwater single photon laser radar imaging method and system based on a gated SPAD array, and belongs to the field of photoelectric detection and image processing. Through the underwater single photon TG-SPAD array range gating imaging system, real-time intensity imaging and high-precision depth imaging of an underwater scene target are realized, and the three-dimensional target reconstruction precision can reach a millimeter level. The system comprises an onshore synchronous control and image processing module and an underwater laser radar module installed in a sealed waterproof shell. The underwater laser radar module comprises a pulse blue-green laser emission module, a single photon TG-SPAD array detection imaging module, and a multipath power supply and signal transmission module. The single-photon TG-SPAD array detection module has two modes of intensity imaging and gating imaging. The pulse blue-green laser transmitting module comprises a pulse blue-green laser and a beam expander. The multi-path power supply and signal transmission module comprises a multi-path power supply, a photoelectric composite watertight cable and a bidirectional optical transceiver. The system has the advantages of simple structure, high stability, low power consumption and high integration level.

Description

Technical Field

[0001] The present invention relates to an underwater single-photon laser radar imaging method and system, and in particular to an underwater single-photon laser radar imaging method and system based on a gated SPAD array, and belongs to the technical field of photoelectric detection and image processing. Background Art

[0002] my country is a major maritime country, and underwater security is related to the country's core security and interests. In coastal areas, near bays and islands and reefs, due to poor water quality, daytime sunlight is severely scattered underwater in shallow waters, and underwater visibility is limited; in deep seas and at night, lighting conditions are severely limited, and underwater unmanned platforms, frogman units and other underwater divers and equipment responsible for underwater reconnaissance, raids and special operations are faced with difficulties such as water scattering affecting visibility and restrictions on day and night environmental conditions.

[0003] At present, the core technology in the field of marine target detection is sonar imaging technology. Although light waves are strongly scattered and absorbed in water, and have greater attenuation than sound waves, and the observable distance is relatively short, light waves have the advantages of good directivity, small divergence, and fast propagation speed. Compared with sonar imaging technology, underwater photoelectric imaging technology has the advantages of high imaging resolution, high frame rate, consistent image and visual habits, and simple system structure. It is an effective way to obtain underwater target scene situation in offshore waters.

[0004] Although the use of high-performance underwater CCD / CMOS imaging, polarization imaging and a variety of corresponding image processing methods have been developed to improve the underwater imaging distance to a certain extent, the improvement is limited (to 1 to 2 times the water body attenuation length AL). At the same time, technologies such as ICCD / ICMOS gated imaging have been developed. Through pulsed laser illumination + time-gated imaging technology, water backscattering is reduced, and the underwater imaging distance (to 4 to 6 times AL) and detail resolution are significantly improved. Through synchronization and delay time control, underwater three-dimensional scene images are calculated. However, the system often requires a high-power lighting system, which has high requirements for underwater platforms, and the gated time width (tens of ns) and repetition rate are limited (hundreds of hertz). The distance resolution of the three-dimensional scene is low, which is difficult to adapt to the application requirements of underwater floating platforms.

[0005] Single-photon imaging is one of the key development directions of photoelectric detection and imaging technology at home and abroad. Single-photon avalanche diode (SPAD) array devices can read the time information and spatial position of a single photon at high speed and accurately. They have the advantages of extremely high sensitivity, picosecond time resolution and low cost, and have shown broad development and application prospects in the field of land lidar detection. At present, there are two main modes of technology that can achieve picosecond resolution in the field of SPAD array imaging: one is time-correlated single photon counting (TCSPC) technology, and the photon statistical timing accuracy determined by the time-to-digital converter (TDC) can reach the picosecond level. The second is time-gated (TG) gating technology, which can obtain images corresponding to different detection times through gated delay offset or gate width modulation. Due to the huge amount of data, the current SPAD array size that TCSPC technology is suitable for is limited (current products are about 64×64 or 320×1), and the imaging resolution is low; and because the number of bits of TDC is fixed, the resolution of TDC is inversely proportional to the dynamic range, making it difficult to achieve long-distance detection while having high temporal resolution. Especially when the backscattering of water bodies is severe, the pixel-level photon response histogram obtained by TCSPC technology must be eliminated through subsequent processing technology to remove path scattering in order to extract and reconstruct the target distance image. The processing algorithm is complex and the detection distance is limited. The working distance of TG gating technology is only limited by the minimum laser clock input frequency of the detector, and is independent of the gate width, TDC accuracy, and gate delay offset accuracy. The interference of the front-end medium backscattering can be filtered out through gating, making it easy to implement large-area array applications.

[0006] Existing underwater single-photon gated imaging methods and systems can often only achieve intensity imaging, and cannot obtain three-dimensional information of the target. In addition, they lack the necessary waterproof housing and underwater signal transmission devices, and their underwater application environments such as the ocean are limited. There is an urgent need for an underwater single-photon lidar imaging method and system that combines a gated SPAD array with high-frequency pulsed laser emission, gated selection, underwater signal transmission and image processing modules to achieve long-distance high-precision imaging underwater or in other strong scattering medium environments. In addition, this technology is also applicable to autonomous driving, remote sensing, non-line-of-sight imaging, biology and other fields. Summary of the invention

[0007] The present invention provides an underwater single-photon lidar imaging method and system based on a gated SPAD array. The technical problem to be solved is: through an underwater single-photon TG-SPAD array range-gated imaging system, real-time intensity imaging and high-precision depth imaging of underwater scene targets are achieved, and the three-dimensional target reconstruction accuracy reaches the millimeter level. The present invention has the advantages of simple structure and high integration.

[0008] The purpose of the present invention is achieved through the following technical solutions.

[0009] The present invention discloses an underwater single-photon laser radar imaging method based on a gated SPAD array, comprising the following steps:

[0010] Step 1: Measure the blind pixel information of the single-photon TG-SPAD array detector, the initial intensity distribution of the laser spot, and the synchronous trigger delay time between the laser and the detector to obtain the blind pixel mask M(i, j), the initial intensity distribution of the laser spot A(i, j), and the synchronous trigger delay time t0;

[0011] Step 2: The laser pulse irradiates the detection target, the detector is triggered by the laser pulse, the host computer sends the gate trigger and timing control signal to the detector, and the detector collects the cumulative echo photon number under different gating conditions to obtain the continuous intensity image matrix I (i, j, s);

[0012] Step 3: Preprocess the continuous intensity image matrix I(i,j,s) by combining the blind pixel mask M(i,j) and the initial intensity distribution A(i,j) of the laser spot to obtain the preprocessed intensity image matrix I A (i, j, s);

[0013] Step 4: Intensity image matrix I A (i, j, s) to perform underwater single-photon image denoising and obtain the denoised intensity image matrix I B (i, j, s);

[0014] Step 5: Combine the synchronous trigger delay time t0 to denoise the intensity image matrix I B (i, j, s) performs depth information reconstruction based on the gated SPAD array to obtain the depth information matrix D of the target, and then obtains the underwater single-photon three-dimensional reconstructed image R through pseudo-color processing.

[0015] The specific implementation method of step one is:

[0016] For the onshore darkroom environment, the blind pixel information of the single-photon TG-SPAD array detector is measured without external light source to obtain the blind pixel mask M(i, j);

[0017] Place a white board with uniform reflectivity, and collect the initial intensity distribution A(i,j) of the laser spot and the photon return time t at a distance d between the detector, the laser and the white board. d ;

[0018] Calculate the photon return time corresponding to the distance d c is the speed of light, then the synchronous trigger delay time t0 = t d -t'.

[0019] The specific implementation of step 2 is as follows:

[0020] Method 1: The laser pulse irradiates the detection target, the detector is triggered by the laser pulse, and the host computer sends the gate trigger and timing control signal to the detector. The gate trigger signal is composed of the parameters of the gate step number S per frame, the initial gate width t w , gate delay t1 and gate delay offset t2, the timing control signal is determined by the parameter laser repetition rate L R , the integration time t corresponding to the generation of a single gate image by the TG-SPAD array detector i , frame number F r The detector collects the cumulative echo photon number after the gates with different offset step sizes are opened multiple times within the integration time in a pulse cycle, and obtains a continuous intensity image matrix I(i, j, s) with a time difference of picoseconds between two adjacent step gate images;

[0021] Method 2: The laser pulse irradiates the detection target, the detector is triggered by the laser pulse, and the host computer sends the gate trigger and timing control signal to the detector. The gate trigger signal is composed of the parameters of the gate step number S per frame, the initial gate width t w , Gate width modulation parameter Δt w , gate delay t1 and gate delay offset t2 (t2 = 0), the timing control signal is determined by the parameter laser repetition rate L R , the integration time t corresponding to the generation of a single gate image by the TG-SPAD array detector i , frame number F r The detector collects the cumulative echo photons after the gate is opened multiple times within the integration time in one pulse cycle, and the gate width increases gradually under different pulse cycles to obtain a continuous intensity image matrix I (i, j, s) with a time difference of nanoseconds between two frames of gated images corresponding to two adjacent pulse cycles.

[0022] The specific implementation method of step 2 method 1 is:

[0023] The initial emission energy of the pulse laser is E0, and the laser repetition rate is L R , the divergence angle after beam expansion is θ, the distance between the target and the laser is l, and the target reflectivity is RT , the water body attenuation length is l0, assuming the target is a Lambertian radiator, the radiance L0 at the target is:

[0024]

[0025] For the detection imaging module, the focal length of the lens is f, which satisfies l>>f, the lens aperture number is F, and the lens transmittance is ρ. Then the irradiance E1 of the target reflected light reaching the detector image plane after passing through the lens is:

[0026]

[0027] For the single-photon TG-SPAD array detector, the pixel size of a single pixel is m×n, the single-photon detector filling factor is η, the quantum efficiency is q, and the integration time corresponding to the detector generating a single gate image is t i , then the energy received at each pixel is E2:

[0028] E2=q·E1·η·m·n·t i ·L R

[0029] Combined with the pulse laser wavelength λ, the number of photons received per pixel per pulse by the single-photon TG-SPAD array detector is N photon :

[0030]

[0031] Where h is Planck's constant and c is the speed of light;

[0032] The gate signal G(t) can be expressed as:

[0033]

[0034] Among them, δ(t) is the unit pulse function;

[0035] The refractive index of water is n, and the laser pulse width is τ (τ ≥ w), then the S i The intensity image matrix I obtained by the detector corresponding to the gate Si (i,j,S i ):

[0036]

[0037] Continuously collect the cumulative echo photon number of each pixel of the detector under S-step picosecond gate delay offset to obtain the continuous intensity image matrix I in each frame k (i,j,s), a total of F r Frame, for each step of gated image, the continuous intensity image matrix I(i,j,s) is obtained by averaging multiple frames:

[0038]

[0039] The specific implementation method of step 2 method 2 is:

[0040] The initial emission energy of the pulse laser is E0, and the laser repetition rate is L R , the divergence angle after beam expansion is θ, the distance between the target and the laser is l, and the target reflectivity is R T , the water body attenuation length is l0, assuming the target is a Lambertian radiator, the radiance L0 at the target is:

[0041]

[0042] For the detection imaging module, the focal length of the lens is f, which satisfies l>>f, the lens aperture number is F, and the lens transmittance is ρ. Then the irradiance E1 of the target reflected light reaching the detector image plane after passing through the lens is:

[0043]

[0044] For the single-photon TG-SPAD array detector, the pixel size of a single pixel is m×n, the single-photon detector filling factor is η, the quantum efficiency is q, and the integration time corresponding to the detector generating a single gate image is t i , then the energy received at each pixel is E2:

[0045] E2=q·E1·η·m·n·t i ·L R

[0046] Combined with the pulse laser wavelength λ, the number of photons received per pixel per pulse by the single-photon TG-SPAD array detector is N photon :

[0047]

[0048] Where h is Planck's constant and c is the speed of light;

[0049] Gate signal G(t):

[0050]

[0051] The refractive index of water is n, and the laser pulse width is τ (τ ≥ t w ), then the intensity image matrix I obtained by the detector corresponding to the kth gate in each frame is k (i,j,k):

[0052]

[0053] The average of S gated images with the same gate width in each frame is calculated to obtain the average intensity image matrix I in each frame. Fri (i,j,F ri ):

[0054]

[0055] Continuous acquisition F r The cumulative echo photon number of each pixel of the detector under nanosecond gate width modulation in the frame is calculated, and the average intensity image matrix in each frame is calculated to obtain the continuous intensity image matrix I(i,j,s).

[0056] The specific implementation method of step three is:

[0057] For each pixel (i, j), the image matrix I after blind pixel correction is M (i,j,s):

[0058]

[0059] in, is the set of four neighbors of pixel (i, j):

[0060]

[0061] The pixel array size of the detector is M×N. For the image matrix I after blind pixel correction M (i, j, s), the target detection distance corresponding to the s-th gated image is d'=2c[t1+(s-1)t2], and the intensity distribution of the laser spot can be expressed as follows under ideal conditions:

[0062]

[0063] Then the preprocessed intensity image matrix I A (i,j,s):

[0064]

[0065] The specific implementation method of step 4 is:

[0066] First, remove the through-scattering effect of water on the target reflected light;

[0067] Intensity image matrix I obtained by underwater single-photon TG-SPAD array A (i,j,s) represents the target reflection component I rs (i,j,s), forward scattering component I fs (i,j,s) and the backscattered component I bs The sum of (i,j,s):

[0068] I A(i,j,s)=I rs (i,j,s)+I fs (i,j,s)+I bs (i,j,s)

[0069] Considering that the backscatter of underwater targets is much greater than the forward scatter (I bs (i,j,s)>>I fs (i,j,s)), intensity image matrix I A (i,j,s) is expressed as:

[0070] I A (i,j,s)=I rs (i,j,s)+I bs (i,j,s)

[0071] Combined with the water body attenuation length l0, the distance l between the target and the laser is an estimated adjustable parameter, and the target reflection component I rs (i,j,s) and the backscattered component I bs (i,j,s) is expressed as:

[0072]

[0073] Among them, I r (i, j, s) is the target reflection component after removing the influence of water body through-scattering; is the intensity image matrix I A The average value of the sum of the top 0.2% of pixels in grayscale value in (i, j, s);

[0074] Get the target reflection component I after removing the influence of water body through-scattering r (i,j,s):

[0075]

[0076] Secondly, the target reflection component I is removed to remove the influence of water body through-scattering r (i,j,s) performs background filtering;

[0077] The background threshold b is set based on the characteristic parameters of the single-photon TG-SPAD array detector and the illumination conditions at the target.

[0078] For method 1, for each pixel (i, j) in the image, select the image matrix I in S frames r (i,j,s), calculate the standard deviation of (i,j):

[0079]

[0080] For method 2, for each pixel (i, j) in the image, in Fr Amplitude gated image matrix I r (i,j,s), calculate the standard deviation of (i,j):

[0081]

[0082] Further obtain the image matrix I after background filtering b (i,j,s):

[0083]

[0084] Finally, for the image matrix I b (i, j, s), filter using the BM3D method to obtain the denoised intensity image matrix I B (i,j,s).

[0085] For step 2, method 1, the implementation method of step 5 is:

[0086] For each pixel (i,j), the pixel centroid

[0087]

[0088] Due to the limitation of laser repetition rate, the detection distance of the target in a single laser cycle is limited. For a longer detection distance, the time corresponding to multiple laser cycles must be accumulated. The synchronous trigger delay time is t0. Combined with the refractive index n of the water body, the calculation formula of the pixel distance matrix D is:

[0089]

[0090] The underwater single-photon three-dimensional reconstructed image R is then obtained through pseudo-color processing.

[0091] For step 2, method 2, the implementation method of step 5 is:

[0092] For each pair of adjacent images I B (i,j,s) and I B (i,j,s+1), there is an intensity difference ΔI B (i, j, s) and intensity change threshold Δk;

[0093] Due to the limitation of laser repetition rate, the detection distance of the target in a single laser cycle is limited. For a longer detection distance, the time corresponding to multiple laser cycles must be accumulated. The synchronous trigger delay time is t0. Combined with the refractive index n of the water body, when the first intensity difference ΔI B When (i,j,s)≥Δk, the target distance D(i,j) of the pixel is:

[0094]

[0095] The underwater single-photon three-dimensional reconstructed image R is then obtained through pseudo-color processing.

[0096] The present invention also discloses a gated SPAD array underwater single-photon laser radar imaging system based on the above method, including an onshore synchronous control and image processing module and an underwater laser radar module installed inside a sealed waterproof housing. The underwater laser radar module includes a pulsed blue-green laser emission module, a single-photon TG-SPAD array detection imaging module, and a multi-channel power supply and signal transmission module. Among them, the synchronous control and image processing module controls the multi-channel power supply and signal transmission module to realize the signal control of the underwater system, and performs blind pixel correction and underwater single-photon image denoising on the data collected by the TG-SPAD detection imaging module to obtain a two-dimensional spatial intensity map of the target, and continuously obtains the intensity image matrix of multiple target scenes through the gated delay offset method of mode one or the gate width modulation method of mode two, so as to realize underwater target detection intensity imaging and high-precision three-dimensional image reconstruction based on the single-photon SPAD array and perform image display. The pulsed blue-green laser emission module emits laser for pulsed illumination of the target scene. The single-photon TG-SPAD array detection module has two modes: intensity imaging and gated imaging. The intensity imaging mode controls the amount of light entering each image by setting the integration time of the single-photon TG-SPAD array detector, and collects single or multiple intensity images by setting the number of frames; the gated imaging mode uses time-gated TG technology. The SPAD array detector only opens and records the photon information reaching the SPAD array detector within the gate window triggered by the pulsed laser, gated trigger signal and timing control signal, thereby effectively eliminating the serious backscattering and other interference noises of the water body in front of the imaging target scene. The main parameters of the TG-SPAD imaging mode are integration time (Integration time, ms level), number of frames (Frames), number of gate steps (Gatesteps) per frame, gate width (Gate width, ns level), gate delay offset (Gate offset, ps level), etc. By adjusting the parameter settings and applying algorithm processing, underwater single-photon two-dimensional and three-dimensional imaging at different distances can be achieved. The multi-channel power supply and signal transmission module is used to provide isolated power supply for each module of the system and realize signal communication and control between the underwater system and the onshore host computer.

[0097] The synchronous control and image processing module is implemented by the host computer. By measuring the synchronous trigger delay time between the laser and the detector, adjusting the time parameters in the host computer, setting the gate trigger signal and the timing control signal, accurately controlling the detector's gate opening time and gate width, and realizing effective gating of the target distance. The synchronous control and image processing module sends instructions such as integration time, number of frames, number of gate steps, gate width, gate delay offset, gate width modulation parameters, etc., and quickly changes the measurement parameters and state of the underwater system through the multi-channel power supply and signal transmission module, which can adapt to the image acquisition requirements of different environments. After the host computer obtains the image signal collected by the detector, it obtains and displays the underwater single-photon two-dimensional intensity image or the underwater single-photon three-dimensional reconstructed image through the image processing method.

[0098] The pulsed blue-green laser emission module includes a pulsed blue-green laser and a beam expander. The laser generates a high-intensity, monochromatic and highly coherent laser beam. In order to match the time resolution of the single-photon detector, a high-repetition-rate pulsed laser is required. The laser output beam has a small divergence angle. After the beam is shaped by the beam expander and the light intensity distribution is optimized, it can output a laser beam suitable for long-distance transmission and covering the lens detection field of view to illuminate the target area.

[0099] The single-photon TG-SPAD array detection imaging module includes a TG-SPAD array detector and a lens, which is used to realize underwater single-photon intensity imaging and gated imaging functions. In the gated imaging mode, the synchronization trigger signal of the SPAD array is provided by the laser output signal, and the two are connected by a synchronization connection line. The focal length of the lens is selected according to the field of view of the detection target scene, the working distance requirements, and the size of the photosensitive surface of the SPAD array detector.

[0100] The multi-channel power supply and signal transmission module includes a multi-channel power supply, an optoelectronic composite watertight cable and a bidirectional optical terminal. The multi-channel power supply realizes the conversion of the required isolated power supply provided by the external 220V power supply to each functional module and device in the system; the optoelectronic composite watertight cable is a professional composite cable of electrical and optical cables, which meets the remote power supply and communication requirements under predetermined water pressure conditions; the bidirectional optical terminal receives the instructions of the host computer to the underwater module, and converts the electrical signals output by each module into optical signals, uses optical fiber for long-distance data transmission, reduces electromagnetic interference, improves the speed and stability of signal transmission, and realizes the conversion of optical signals to electrical signals through USB devices on the PC side.

[0101] Preferably, according to actual use requirements, a picosecond delay device can be added to the single-photon TG-SPAD imaging module, and connected to the laser and the SPAD array through a synchronization connection line, so as to adjust the amplitude of the pulse signal to adapt it to the SPAD array and ensure the stability of the synchronization signal jitter between the laser and the SPAD. The laser output pulse signal is used as the input of the picosecond delay device, and the electrical pulse signal output by the picosecond delay device is used to trigger the single-photon SPAD array detector.

[0102] Preferably, the pulsed blue-green laser is connected to the beam expander via an optical fiber, allowing the laser and the beam expander to be separated in space, thereby reducing the risk of alignment errors during system installation leading to reduced beam quality.

[0103] Preferably, a narrow-band filter corresponding to the laser band is installed at the supporting lens of the single-photon TG-SPAD array detector and the optical window of the system sealed waterproof housing to filter out stray light of other wavelengths in the environment and improve imaging quality.

[0104] Beneficial effects:

[0105] 1. The present invention discloses an underwater single-photon lidar imaging method and system based on a gated SPAD array, with a single-photon TG-SPAD array detector as the core component. The host computer quickly changes the underwater system's integration time, the number of gate steps per frame, the gate width, the gate delay offset and other measurement parameter settings through a multi-channel power supply and a signal transmission module. The detector collects the cumulative number of echo photons under different gating conditions, and the synchronization control and image processing module performs subsequent algorithm processing. It can realize single-photon array detection and imaging in a variety of environments such as deep sea or turbid water, and has large-array high-resolution real-time intensity imaging and high-precision three-dimensional imaging functions. The three-dimensional target reconstruction accuracy can reach millimeter level at most, and it is expected to achieve ultra-long-distance imaging.

[0106] 2. The present invention discloses an underwater single-photon lidar imaging method and system based on a gated SPAD array, which adopts an optoelectronic composite watertight cable and a bidirectional optical terminal to realize the mutual conversion and transmission of electrical signals and optical signals, and the underwater components are placed in a sealed waterproof casing, which can cope with extreme conditions such as deep-sea pressure and temperature and realize remote control. The signal transmission speed between the onshore synchronous control and image processing module and the underwater lidar module is fast and the stability is high.

[0107] 3. The present invention discloses an underwater single-photon lidar imaging method and system based on a gated SPAD array. The underwater part integrates pulsed laser emission, single-photon TG-SPAD array imaging, multi-channel power supply and signal transmission functions. The overall system has the advantages of simple structure, strong stability, low power consumption and high integration. BRIEF DESCRIPTION OF THE DRAWINGS

[0108] Figure 1 It is a schematic diagram of the flow of an underwater single-photon lidar imaging method based on a gated SPAD array of the present invention.

[0109] Figure 2 This is a structural block diagram of an underwater single-photon lidar imaging system based on a gated SPAD array of the present invention.

[0110] Figure 3 This is a schematic diagram of the gate control principle of the present invention.

[0111] Figure 4 Schematic diagram of gated trigger and timing control signal in the present invention, (a) is the gated delay offset reconstruction method of mode 1, and (b) is the gate width modulation method of mode 2.

[0112] Figure 5 It is a structural schematic diagram of a specific embodiment of an underwater single-photon lidar imaging method and system based on a gated SPAD array of the present invention.

[0113] In the figure: 1-power supply, 2-single photon detector, 3-lens, 4-glass window, 5-optical terminal, 6-partition, 7-beam expander, 8-laser, 9-host computer, 10-optical composite cable, 11-waterproof housing.

[0114] Figure 6 It is a schematic diagram of the SOLIDWORKS layout mode of the internal structure fixed frame of a specific embodiment of an underwater single-photon lidar imaging method and system based on a gated SPAD array of the present invention. DETAILED DESCRIPTION

[0115] In order to more clearly describe the technical solution of the present invention, the invention content will be further explained below in conjunction with the accompanying drawings and specific embodiments.

[0116] Embodiment 1:

[0117] like Figure 1 As shown, this embodiment discloses an underwater single-photon lidar imaging method based on a gated SPAD array, and the specific implementation steps are as follows:

[0118] Step 1: Measure the blind pixel information of the single-photon TG-SPAD array detector, the initial intensity distribution of the laser spot, and the synchronous trigger delay time between the laser and the detector to obtain the blind pixel mask M(i, j), the initial intensity distribution of the laser spot A(i, j), and the synchronous trigger delay time t0.

[0119] For the onshore darkroom environment, the blind pixel information of the single-photon TG-SPAD array detector is measured without external light source to obtain the blind pixel mask M(i, j);

[0120] Place a white board with uniform reflectivity, and collect the initial intensity distribution A(i,j) of the laser spot and the photon return time t at a distance d between the detector, the laser and the white board. d ;

[0121] Calculate the photon return time corresponding to the distance d c is the speed of light, then the synchronous trigger delay time t0 = t d -t'.

[0122] Step 2: The laser pulse irradiates the detection target, and the detector is triggered by the laser pulse. The host computer sends the gate trigger and timing control parameters to the detector. The gate trigger parameters specifically include the number of gate steps S per frame, the initial gate width t w , gate delay t1 and gate delay offset t2, the timing control parameters specifically include laser repetition rate L R , the integration time t corresponding to the generation of a single gate image by the TG-SPAD array detector i , frame number F r With the synchronous trigger delay time t0, the detector collects the cumulative echo photon number after the gates with different offset step sizes are opened multiple times within the integration time in a pulse cycle, and obtains a continuous intensity image matrix I(i, j, s) with a time difference of picoseconds between two adjacent step gate images.

[0123] Specifically, the initial emission energy of the pulsed laser is E0, and the laser repetition rate is L R , the divergence angle after beam expansion is θ, the distance between the target and the laser is l, and the target reflectivity is R T , the water body attenuation length is l0, assuming the target is a Lambertian radiator, the radiance L0 at the target is:

[0124]

[0125] For the detection imaging module, the focal length of the lens is f, which satisfies l>>f, the lens aperture number is F, and the lens transmittance is ρ. Then the irradiance of the target reflected light reaching the detector image plane after passing through the lens is:

[0126]

[0127] For the single-photon TG-SPAD array detector, the pixel size of a single pixel is m×n, the single-photon detector filling factor is η, the quantum efficiency is q, and the integration time corresponding to the detector generating a single gate image is t i , then the energy received at each pixel is E2:

[0128] E2=q·E1·η·m·n·t i ·L R

[0129] Combined with the pulse laser wavelength λ, the number of photons received per pixel per pulse by the single-photon TG-SPAD array detector is N photon :

[0130]

[0131] Among them, h is Planck's constant and c is the speed of light.

[0132] Gating trigger and timing control signals are as follows Figure 4 As shown in (a), the gate signal G(t) can be expressed as:

[0133]

[0134] Among them, δ(t) is the unit pulse function.

[0135] The refractive index of water is n, and the laser pulse width is τ (τ ≥ w), then the S i The intensity image matrix I obtained by the detector corresponding to the gate Si (i,j,S i ):

[0136]

[0137] Continuously collect the cumulative echo photon number of each pixel of the detector under S-step picosecond gate delay offset to obtain the continuous intensity image matrix I in each frame k (i,j,s), a total of F r Frame, for each step of gated image, the continuous intensity image matrix I(i,j,s) is obtained by averaging multiple frames:

[0138]

[0139] Step 3: Preprocess the continuous intensity image matrix I(i,j,s) by combining the blind pixel mask M(i,j) and the initial intensity distribution A(i,j) of the laser spot to obtain the preprocessed intensity image matrix I A (i,j,s).

[0140] Specifically, for each pixel (i, j), if the pixel is a blind pixel, that is, M(i, j) = 1, it is replaced with the average value of the non-blind pixels in its four neighborhoods. If all the neighborhoods are blind pixels, the pixel value is set to 0; if the pixel is a non-blind pixel, that is, M(i, j) = 0, the original value is retained. Thus, the image matrix I after blind pixel correction is obtained M (i,j,s):

[0141]

[0142] in, is the set of four neighbors of pixel (i, j):

[0143]

[0144] The pixel array size of the detector is M×N. For the image matrix I after blind pixel correction M (i, j, s), the target detection distance corresponding to the s-th gated image is d'=2c[t1+(s-1)t2], and the intensity distribution of the laser spot can be expressed as follows under ideal conditions:

[0145]

[0146] Then the preprocessed intensity image matrix I A (i,j,s):

[0147]

[0148] Step 4: Intensity image matrix I A (i, j, s) to perform underwater single-photon image denoising and obtain the denoised intensity image matrix I B (i,j,s).

[0149] Specifically, the intensity image matrix I obtained by the underwater single-photon TG-SPAD array is A (i,j,s) represents the target reflection component I rs (i,j,s), forward scattering component I fs (i,j,s) and the backscattered component I bs The sum of (i,j,s):

[0150] I A (i,j,s)=I rs (i,j,s)+I fs (i,j,s)+I bs (i,j,s)

[0151] Considering that the backscatter of underwater targets is much greater than the forward scatter (I bs (i,j,s)>>I fs (i,j,s)), intensity image matrix I A (i,j,s) is expressed as:

[0152] I A (i,j,s)=I rs (i,j,s)+I bs (i,j,s)

[0153] Combined with the water body attenuation length l0, the distance l between the target and the laser is an estimated adjustable parameter, and the target reflection component I rs(i,j,s) and the backscattered component I bs (i,j,s) is expressed as:

[0154]

[0155] Among them, I r (i, j, t) is the target reflection component after removing the influence of water body through-scattering; is the intensity image matrix I A The average value of the sum of the top 0.2% of the pixels in the grayscale value of (i, j, t). In the specific embodiment, according to different underwater detection distances and lighting conditions, the sum of pixels with appropriate grayscale values ​​is selected and averaged to obtain the parameter

[0156] Above, except for the target reflection component I affected by water body penetration scattering r (i,j,s):

[0157]

[0158] Secondly, the target reflection component I is removed to remove the influence of water body through-scattering r (i, j, s) for background filtering. Combining the single-photon TG-SPAD array detector characteristic parameters (dark count rate, peak detectability, etc.) and the target illumination conditions, the background threshold b is set. In this embodiment, according to different underwater detection distances (1-16m), the value range of the background threshold is b = [4, 25].

[0159] For each pixel (i, j) in the image, the image matrix I is gated in S frames r (i,j,s), calculate the standard deviation of (i,j):

[0160]

[0161] Further obtain the image matrix I after background filtering b (i,j,s):

[0162]

[0163] Finally, for the image matrix I b (i, j, s), the BM3D method is used for filtering. In this embodiment, the parameter value α = [10, 25] is selected to obtain the denoised intensity image matrix I B (i,j,s).

[0164] Step 5: Combine the synchronous trigger delay time t0 to denoise the intensity image matrix I B(i, j, s) performs depth information reconstruction based on the gated SPAD array to obtain the depth information matrix D of the target, and then obtains the underwater single-photon three-dimensional reconstructed image R through pseudo-color processing.

[0165] For each pixel (i,j), the pixel centroid

[0166]

[0167] Or combine the laser response function h of the detector and obtain the pixel centroid according to the pixel cross-correlation method

[0168] Due to the limitation of laser repetition rate, the detection distance of the target in a single laser cycle is limited. For a longer detection distance, the time corresponding to multiple laser cycles must be accumulated. The synchronous trigger delay time is t0. Combined with the refractive index n of the water body, the calculation formula of the pixel distance matrix D is:

[0169]

[0170] The underwater single-photon three-dimensional reconstructed image R is then obtained through pseudo-color processing.

[0171] like Figure 5As shown. The gated SPAD array underwater single-photon laser radar imaging system based on the above method disclosed in this embodiment includes an onshore synchronous control and image processing module and an underwater laser radar module installed inside a sealed waterproof housing. The underwater laser radar module includes a pulsed blue-green laser emission module, a single-photon TG-SPAD array detection imaging module, and a multi-channel power supply and signal transmission module. The single-photon SPAD array imaging module includes a single-photon detector 2 and a lens 3. Among them, the single-photon TG-SPAD array detector 2 is a SPAD512S single-photon array detector of PI Imaging, which has both intensity imaging and gated gating imaging modes. The detector array size is 512×512, the pixel pitch is 16.38μm, the fill factor is higher than 50%, the peak photon detection rate at 520nm is higher than 50%, the minimum gate width can reach 5ns, the minimum gate delay offset is 17.9ps, the receivable laser synchronization frequency is 10 / 20MHz, and the external laser synchronization trigger signal voltage requirement is 0-1.8V. The lens 3 is a 135mm F1.8 DG HSM|Art lens from SIGMA, with a focal length of 135mm and a maximum aperture of F1.8. The underwater field of view of the single-photon detector 2 is about 2.7°, and an object about 2.3m high at 50m underwater can be fully imaged. The laser emission module includes a laser 8 and a beam expander 7. Among them, the laser 8 uses the VisUV-532 picosecond laser from PicoQuant of Germany, with a central wavelength of 532nm, an output light pulse width of less than 85ps, an internally triggered laser repetition rate of 1-20MHz, a single pulse energy of 5nJ at a repetition rate of 10MHz, an average power of 100mW, a peak power of 60W, and a synchronous output signal amplitude of 3.3V. In this embodiment, the laser operating frequency is 10MHz, and the laser 8 is connected to the single-photon detector 2 through a synchronous connection line. At this time, the underwater photon non-mixing working distance of the single-photon TG-SPAD array detector 2 is 11.3m. When there is no obvious obstruction in front of the target and the algorithm is used to filter out the underwater scattered light, the working distance can be further increased. The beam expander 7 uses the product of Beijing Gaopule Optoelectronics Technology Co., Ltd., so that the laser divergence angle is adjustable. The beam expander magnification can be adjusted according to the actual target distance and size, and the maximum divergence angle is 45°. In this embodiment, the laser and the beam expander are coaxial by adding a translation stage. The power supply and transmission module includes an optical terminal 5, a power supply 1 and an optoelectronic composite cable 10. The optical terminal 5 uses the FYZ-PW69 optical terminal of the Fengying sub-brand. The product has 4 USB ports and can simultaneously transmit 4 signals through LC fiber optic connectors. The single-fiber transmission distance is up to 250m. The optoelectronic composite cable 14 is a customized product of the W series underwater connector of the Swiss LEMO company. The cable length inside the cabin is 1m and about 100m outside the cabin. The plug-in self-locking connector uses an 8-hole 308 series product, which can simultaneously transmit 6 electrical signals and 2 optical fiber signals.The power supply 1 uses the 4NIC-Q73G-2L switching power supply of Aerospace Changfeng Chaoyang Power Supply Co., Ltd., with an input parameter of AC220V±10%50Hz single-phase electricity, and output terminals of three groups of DC12V / 48W and one group of DC5V / 25W, of which two groups of 12V terminals are used to power the laser 8 and the optical terminal 5, respectively, and one group of 5V terminals is used to power the single-photon TG-SPAD array detector 2. The master and slave detector output signals of the single-photon TG-SPAD array detector 2 are connected to the optical terminal 5 through a USB connection line, and the optical fiber transmission signal of the optical terminal 5 and the power line of the power supply 1 are connected to the photoelectric composite cable 10 through a connector, and are connected to the ground equipment, i.e., the host computer 9 after being transmitted for 100m through the photoelectric composite cable 10. The synchronous control and image processing module includes the host computer 9. In this embodiment, the system model HP ZHAN 66Pro G1 is used, and the processor is intel(R)Core(TM)i5-82500 CPU@1.60GHz(8CPUs). Other structures of the power supply include a glass window 4, a partition 6 and a waterproof shell 11. Among them, the glass window 4 is made of two pieces of quartz glass with a diameter of 120mm and a thickness of 20mm. The material is JGS1, and the double-sided 550nm anti-reflection film is coated as an optical window for laser emission and echo reception. A partition 6 is installed inside the system between the two windows. It is made of 6061 aluminum alloy material and the outer surface is coated with matte black paint to prevent the laser beam from reflecting or scattering in the cabin to generate noise and affect the imaging quality of the single-photon TG-SPAD array detector 2. The waterproof shell 11 is made of 7075 aluminum alloy material and can withstand a water depth of 100m. The overall layout of the internal structure of the system is as follows. Figure 6 As shown, the outer surfaces of the cabin structure are all sprayed with matte black paint. Under ideal conditions, the depth resolution of the method described in this embodiment can reach 2.7 mm.

[0172] In this embodiment, the picosecond gated delay offset is realized by the gating function of the single-photon TG-SPAD array detector. Among them, the picosecond gated delay offset can also be realized by connecting a picosecond delay device to the laser 8, and then connecting the picosecond delay device to the single-photon TG-SPAD array detector. Among them, the picosecond delay device can use the PSD-065-A-MOD picosecond delay device of Micro Photon Devices, with a delay range of 0-50ns, an accuracy of 10ps, a pulse width adjustment range of 1-250ns, and an accuracy of 3.3ns. It can receive an input signal amplitude of -2V~+3V. After being connected to the laser output signal, the picosecond delay device can output a standard TTL signal that meets the detector synchronization trigger signal requirements. The picosecond delay device is connected to the single-photon TG-SPAD array detector 2 and the laser 8 through a synchronization connection line. When this method is used under ideal conditions, the depth resolution can reach 1.5mm.

[0173] In addition, the above system can be used for single-photon imaging on the road and can be used normally without the optoelectronic composite cable 10 and the waterproof housing 11.

[0174] Embodiment 2:

[0175] like Figure 1 As shown, this embodiment discloses an underwater single-photon lidar imaging method based on a gated SPAD array, and the specific implementation steps are as follows:

[0176] Step 1: Measure the blind pixel information of the single-photon TG-SPAD array detector, the initial intensity distribution of the laser spot, and the synchronous trigger delay time between the laser and the detector to obtain the blind pixel mask M(i, j), the initial intensity distribution of the laser spot A(i, j), and the synchronous trigger delay time t0.

[0177] For the onshore darkroom environment, the blind pixel information of the single-photon TG-SPAD array detector is measured without external light source to obtain the blind pixel mask M(i, j);

[0178] Place a white board with uniform reflectivity, and collect the initial intensity distribution A(i,j) of the laser spot and the photon return time t at a distance d between the detector, the laser and the white board. d ;

[0179] Calculate the photon return time corresponding to the distance d c is the speed of light, then the synchronous trigger delay time t0 = t d -t'.

[0180] Step 2: The laser pulse irradiates the detection target, and the detector is triggered by the laser pulse. The host computer sends the gate trigger and timing control parameters to the detector. The gate trigger parameters specifically include the number of gate steps S per frame, the initial gate width t w , Gate width modulation parameter Δt w , gate delay t1 and gate delay offset t2 (t2=0), the timing control parameters specifically include laser repetition rate L R , the integration time t corresponding to the generation of a single gate image by the TG-SPAD array detector i , frame number F r And the synchronous trigger delay time t0, the detector collects the cumulative echo photon number after the gate is opened multiple times within the integration time in one pulse cycle, the gate width increases gradually under different pulse cycles, and the continuous intensity image matrix I(i, j, s) with a time difference of nanoseconds between two frames of gated images corresponding to two adjacent pulse cycles is obtained.

[0181] The initial emission energy of the pulse laser is E0, and the laser repetition rate is L R , the divergence angle after beam expansion is θ, the distance between the target and the laser is l, and the target reflectivity is RT , the water body attenuation length is l0, assuming the target is a Lambertian radiator, the radiance L0 at the target is:

[0182]

[0183] For the detection imaging module, the focal length of the lens is f, which satisfies l>>f, the lens aperture number is F, and the lens transmittance is ρ. Then the irradiance E1 of the target reflected light reaching the detector image plane after passing through the lens is:

[0184]

[0185] For the single-photon TG-SPAD array detector, the pixel size of a single pixel is m×n, the single-photon detector filling factor is η, the quantum efficiency is q, and the integration time corresponding to the detector generating a single gate image is t i , then the energy received at each pixel is E2:

[0186] E2=q·E1·η·m·n·t i ·L R

[0187] Combined with the pulse laser wavelength λ, the number of photons received per pixel per pulse by the single-photon TG-SPAD array detector is N photon :

[0188]

[0189] Among them, h is Planck's constant and c is the speed of light.

[0190] For gate trigger and timing control signals, see the attached Figure 4 As shown in (b), the gate signal G(t) is:

[0191]

[0192] Among them, δ(t) is the unit pulse function.

[0193] The refractive index of water is n, and the laser pulse width is τ (τ ≥ t w ), then the intensity image matrix I obtained by the detector corresponding to the kth gate in each frame is k (i,j,k):

[0194]

[0195] The average of S gated images with the same gate width in each frame is calculated to obtain the average intensity image matrix I in each frame. Fri (i,j,F ri ):

[0196]

[0197] Continuous acquisition F r The cumulative echo photon number of each pixel of the detector under nanosecond gate width modulation in the frame is calculated, and the average intensity image matrix in each frame is calculated to obtain the continuous intensity image matrix I(i,j,s).

[0198] Step 3: Preprocess the continuous intensity image matrix I(i,j,s) by combining the blind pixel mask M(i,j) and the initial intensity distribution A(i,j) of the laser spot to obtain the preprocessed intensity image matrix I A (i,j,s).

[0199] Specifically, for each pixel (i, j), if the pixel is a blind pixel, that is, M(i, j) = 1, it is replaced with the average value of the non-blind pixels in its four neighborhoods. If all the neighborhoods are blind pixels, the pixel value is set to 0; if the pixel is a non-blind pixel, that is, M(i, j) = 0, the original value is retained. Thus, the image matrix I after blind pixel correction is obtained M (i,j,s):

[0200]

[0201] in, is the set of four neighbors of pixel (i, j):

[0202]

[0203] The pixel array size of the detector is M×N. For the image matrix I after blind pixel correction M (i, j, s), the target detection distance corresponding to the s-th gated image is d'=2c[t1+(s-1)t2]. Since t2=0, it can be simplified to d'=2ct1. The intensity distribution of the laser spot can be expressed as follows under ideal conditions:

[0204]

[0205] Then the preprocessed intensity image matrix I A (i,j,s):

[0206]

[0207] Step 4: Intensity image matrix I A (i, j, s) to perform underwater single-photon image denoising and obtain the denoised intensity image matrix I B (i,j,s).

[0208] Specifically, the through-scattering effect of water on the target reflected light is first removed. The intensity image matrix I obtained by the underwater single-photon TG-SPAD arrayA (i,j,s) represents the target reflection component I rs (i,j,s), forward scattering component I fs (i,j,s) and the backscattered component I bs The sum of (i,j,s):

[0209] I A (i,j,s)=I rs (i,j,s)+I fs (i,j,s)+I bs (i,j,s)

[0210] Considering that the backscatter of underwater targets is much greater than the forward scatter (I bs (i,j,s)>>I fs (i,j,s)), intensity image matrix I A (i,j,s) is expressed as:

[0211] I A (i,j,s)=I rs (i,j,s)+I bs (i,j,s)

[0212] Combined with the water body attenuation length l0, the distance l between the target and the laser is an estimated adjustable parameter, and the target reflection component I rs (i,j,s) and the backscattered component I bs (i,j,s) is expressed as:

[0213]

[0214] Among them, I r (i, j, s) is the target reflection component after removing the influence of water body through-scattering; is the intensity image matrix I A The average value of the sum of the top 0.2% of the pixels in the grayscale value of (i, j, s). In the specific embodiment, according to different underwater detection distances and lighting conditions, the sum of pixels with appropriate grayscale values ​​is selected and averaged to obtain the parameter

[0215] The above removes the target reflection component I affected by water body penetration scattering r (i,j,s):

[0216]

[0217] Secondly, the target reflection component I is removed to remove the influence of water body through-scattering r(i, j, s) for background filtering. Combined with the single-photon TG-SPAD array detector characteristic parameters (dark count rate, peak detectability, etc.) and the target illumination conditions, the background threshold b is set. In this embodiment, according to different underwater detection distances (1-16m), the background threshold value range is b = [4, 25]. For each pixel (i, j) in the image, in F r Amplitude gated image matrix I r (i,j,s), calculate the standard deviation of (i,j):

[0218]

[0219] Further obtain the image matrix I after background filtering b (i,j,s):

[0220]

[0221] Finally, for the image matrix I b (i, j, s), filter using the BM3D method to obtain the denoised intensity image matrix I B (i,j,s).

[0222] Step 5: Combine the synchronous trigger delay time t0 to denoise the intensity image matrix I B (i, j, s) performs depth information reconstruction based on the gated SPAD array to obtain the depth information matrix D of the target, and then obtains the underwater single-photon three-dimensional reconstructed image R through pseudo-color processing.

[0223] For each pair of adjacent images I B (i,j,s) and I B (i,j,s+1), there is an intensity difference ΔI B (i, j, s) and intensity change threshold Δk;

[0224] An intensity change threshold Δk is set. In this embodiment, the threshold is selected as Δk=[4,25] according to different target distances.

[0225] Due to the limitation of laser repetition rate, the detection distance of the target in a single laser cycle is limited. For a longer detection distance, the time corresponding to multiple laser cycles must be accumulated. The synchronous trigger delay time is t0. Combined with the refractive index n of the water body, when the first intensity difference ΔI B When (i,j,s)≥Δk, the target distance D(i,j) of the pixel is:

[0226]

[0227] The underwater single-photon three-dimensional reconstructed image R is then obtained through pseudo-color processing.

[0228] The experimental system used in this embodiment is the same as that in embodiment 1. In this embodiment, nanosecond gate width modulation is realized by the gating function of the single-photon TG-SPAD array detector, and the gate width modulation parameter Δt w The depth resolution can reach 0.15m under ideal conditions. w , the system depth resolution can be further improved.

[0229] The above description is only a specific embodiment of the present invention and is not intended to limit the protection scope of the present invention. The present invention may also have multiple embodiments. Within the spirit and principle of the present invention, those skilled in the art can make various corresponding changes and deformations according to the present invention, and any modifications, equivalent substitutions, improvements, etc. made should be included in the protection scope of the present invention.

Claims

1. An underwater single-photon lidar imaging method based on a gated SPAD array, characterized in that: The following steps are included: Step 1: Measure the blind pixel information of the single-photon TG-SPAD array detector to obtain the blind pixel mask M(i, j); measure the initial intensity distribution A(i, j) of the laser spot and the synchronous trigger delay time t0 between the laser and the detector; Step 2: The laser pulse irradiates the detection target, the detector is triggered by the laser pulse, the host computer sends the gate trigger and timing control signal to the detector, and the detector collects the cumulative echo photon number under different gating conditions to obtain the continuous intensity image matrix I (i, j, s); Step 3: Preprocess the continuous intensity image matrix I(i,j,s) by combining the blind pixel mask M(i,j) and the initial intensity distribution A(i,j) of the laser spot to obtain the preprocessed intensity image matrix I A (i, j, s); Step 4: Intensity image matrix I A (i, j, s) to perform underwater single-photon image denoising and obtain the denoised intensity image matrix I B (i, j, s); Step 5: Combine the synchronous trigger delay time t0 to denoise the intensity image matrix I B (i, j, s) performs depth information reconstruction based on the gated SPAD array to obtain the depth information matrix D of the target, and then obtains the underwater single-photon three-dimensional reconstructed image R through pseudo-color processing.

2. The underwater single-photon laser radar imaging method based on a gated SPAD array as claimed in claim 1, characterized in that: The specific implementation methods of step 2 include the following two methods: Method 1: The laser pulse irradiates the detection target, the detector is triggered by the laser pulse, and the host computer sends the gate trigger and timing control signal to the detector. The gate trigger signal is composed of the parameters of the gate step number S per frame, the initial gate width t w , gate delay t1 and gate delay offset t2, the timing control signal is determined by the parameter laser repetition rate L R , the integration time t corresponding to the generation of a single gate image by the TG-SPAD array detector i , frame number F r The detector collects the cumulative echo photon number after the gates with different offset step sizes are opened multiple times within the integration time in a pulse cycle, and obtains a continuous intensity image matrix I(i, j, s) with a time difference of picoseconds between two adjacent step gate images; Method 2: The laser pulse irradiates the detection target, the detector is triggered by the laser pulse, and the host computer sends the gate trigger and timing control signal to the detector. The gate trigger signal is composed of the parameters of the gate step number S per frame, the initial gate width t w , Gate width modulation parameter Δt w , gate delay t1 and gate delay offset t2 (t2 = 0), the timing control signal is determined by the parameter laser repetition rate L R , the integration time t corresponding to the generation of a single gate image by the TG-SPAD array detector i , frame number F r The detector collects the cumulative echo photons after the gate is opened multiple times within the integration time in one pulse cycle, and the gate width increases gradually under different pulse cycles to obtain a continuous intensity image matrix I (i, j, s) with a time difference of nanoseconds between two frames of gated images corresponding to two adjacent pulse cycles.

3. The underwater single-photon laser radar imaging method based on a gated SPAD array as claimed in claim 1, characterized in that: The specific implementation method of step one is: For the onshore darkroom environment, the blind pixel information of the single-photon TG-SPAD array detector is measured without external light source to obtain the blind pixel mask M(i, j); Place a white board with uniform reflectivity, and collect the initial intensity distribution A(i,j) of the laser spot and the photon return time t at a distance d between the detector, the laser and the white board. d ; Calculate the photon return time corresponding to the distance d c is the speed of light, then the synchronous trigger delay time t0 = t d -t'.

4. The underwater single-photon laser radar imaging method based on a gated SPAD array as claimed in claim 2, characterized in that: The specific implementation method of step 2 method 1 is: The initial emission energy of the pulse laser is E0, and the laser repetition rate is L R , the divergence angle after beam expansion is θ, the distance between the target and the laser is l, and the target reflectivity is R T , the water body attenuation length is l0, assuming the target is a Lambertian radiator, the radiance L0 at the target is: For the detection imaging module, the focal length of the lens is f, which satisfies l>>f, the lens aperture number is F, and the lens transmittance is ρ. Then the irradiance E1 of the target reflected light reaching the detector image plane after passing through the lens is: For the single-photon TG-SPAD array detector, the pixel size of a single pixel is m×n, the single-photon detector filling factor is η, the quantum efficiency is q, and the integration time corresponding to the detector generating a single gate image is t i , then the energy received at each pixel is E2: E2=q·E1·η·m·n·t i ·L R Combined with the pulse laser wavelength λ, the number of photons received per pixel per pulse by the single-photon TG-SPAD array detector is N photon : Where h is Planck's constant and c is the speed of light; The gate signal G(t) can be expressed as: Among them, δ(t) is the unit pulse function; The refractive index of water is n, and the laser pulse width is τ (τ ≥ w), then the S i The intensity image matrix I obtained by the detector corresponding to the gate Si (i,j,S i ): Continuously collect the cumulative echo photon number of each pixel of the detector under S-step picosecond gate delay offset to obtain the continuous intensity image matrix I in each frame k (i,j,s), a total of F r Frame, for each step of gated image, the continuous intensity image matrix I(i,j,s) is obtained by averaging multiple frames:

5. The underwater single-photon laser radar imaging method based on a gated SPAD array as claimed in claim 2, characterized in that: The specific implementation method of step 2 method 2 is: The initial emission energy of the pulse laser is E0, and the laser repetition rate is L R , the divergence angle after beam expansion is θ, the distance between the target and the laser is l, and the target reflectivity is R T , the water body attenuation length is l0, assuming the target is a Lambertian radiator, the radiance L0 at the target is: For the detection imaging module, the focal length of the lens is f, which satisfies l>>f, the lens aperture number is F, and the lens transmittance is ρ. Then the irradiance E1 of the target reflected light reaching the detector image plane after passing through the lens is: For the single-photon TG-SPAD array detector, the pixel size of a single pixel is m×n, the single-photon detector filling factor is η, the quantum efficiency is q, and the detector integration time is t i , then the energy received at each pixel is E2: E2=q·E1·η·m·n·t i ·L R Combined with the pulse laser wavelength λ, the number of photons received per pixel per pulse by the single-photon TG-SPAD array detector is N photon : Where h is Planck's constant and c is the speed of light; Gate signal G(t): Among them, δ(t) is the unit pulse function; The refractive index of water is n, and the laser pulse width is τ (τ ≥ t w ), then the intensity image matrix I obtained by the detector corresponding to the kth gate in each frame is k (i,j,k): Average the S gated images with the same gate width in each frame to obtain the average intensity image matrix in each frame. Continuous acquisition F r The cumulative echo photon number of each pixel of the detector under nanosecond gate width modulation in the frame is calculated, and the average intensity image matrix in each frame is calculated to obtain the continuous intensity image matrix I(i,j,s).

6. The underwater single-photon laser radar imaging method based on a gated SPAD array as claimed in claim 3, characterized in that: The specific implementation method of step three is: For each pixel (i, j), the image matrix I after blind pixel correction is M (i,j,s): in, is the set of four neighbors of pixel (i, j): The pixel array size of the detector is M×N. For the image matrix I after blind pixel correction M (i, j, s), the target detection distance corresponding to the s-th gated image is d'=2c[t1+(s-1)t2], and the intensity distribution of the laser spot can be expressed as follows under ideal conditions: Then the preprocessed intensity image matrix I A (i,j,s): The specific implementation method of step 4 is: First, remove the through-scattering effect of water on the target reflected light; Intensity image matrix I obtained by underwater single-photon TG-SPAD array A (i,j,s) represents the target reflection component I rs (i,j,s), forward scattering component I fs (i,j,s) and the backscattered component I bs The sum of (i,j,s): I A (i,j,s)=I rs (i,j,s)+I fs (i,j,s)+I bs (i,j,s) Considering that the backscatter of underwater targets is much greater than the forward scatter (I bs (i,j,s)>>I fs (i,j,s)), intensity image matrix I A (i,j,s) is expressed as: I A (i,j,s)=I rs (i,j,s)+I bs (i,j,s) Combined with the water body attenuation length l0, the distance l between the target and the laser is an estimated adjustable parameter, and the target reflection component I rs (i,j,s) and the backscattered component I bs (i,j,s) is expressed as: Among them, I r (i, j, s) is the target reflection component after removing the influence of water body through-scattering; I is the intensity image matrix I A The average value of the sum of the top 0.2% of pixels in grayscale value in (i, j, s); Get the target reflection component I after removing the influence of water body through-scattering r (i,j,s): Secondly, the target reflection component I is removed to remove the influence of water body through-scattering r (i,j,s) performs background filtering; The background threshold b is set based on the characteristic parameters of the single-photon TG-SPAD array detector and the illumination conditions at the target. For method 1, for each pixel (i, j) in the image, select the image matrix I in S frames r (i,j,s), calculate the standard deviation of (i,j): For method 2, for each pixel (i, j) in the image, the image matrix I is gated in Fr r (i,j,s), calculate the standard deviation of (i,j): Further obtain the image matrix I after background filtering b (i,j,s): Finally, for the image matrix I b (i, j, s), filter using the BM3D method to obtain the denoised intensity image matrix I B (i,j,s).

7. An underwater single-photon laser radar imaging method based on a gated SPAD array as claimed in claim 2 or 4, characterized in that: For step 2, method 1, the implementation method of step 5 is: For each pixel (i,j), the pixel centroid Due to the limitation of laser repetition rate, the detection distance of the target in a single laser cycle is limited. For a longer detection distance, the time corresponding to multiple laser cycles must be accumulated; The synchronous trigger delay time is t0. Combined with the water refractive index n, the calculation formula of the pixel distance matrix D is: The underwater single-photon three-dimensional reconstructed image R is then obtained through pseudo-color processing.

8. An underwater single-photon laser radar imaging method based on a gated SPAD array as claimed in claim 2 or 5, characterized in that: For step 2 method 2, the implementation method of step 5 is: For each pair of adjacent images I B (i,j,s) and I B (i,j,s+1), there is an intensity difference ΔI B (i, j, s) and intensity change threshold Δk; Due to the limitation of laser repetition rate, the detection distance of the target in a single laser cycle is limited. For a longer detection distance, the time corresponding to multiple laser cycles must be accumulated; The synchronous trigger delay time is t0, combined with the water refractive index n, when the first intensity difference ΔI B When (i,j,s)≥Δk, the target distance D(i,j) of the pixel is: The underwater single-photon three-dimensional reconstructed image R is then obtained through pseudo-color processing.

9. A system for implementing the method according to any one of claims 1 to 6, characterized in that: It includes an onshore synchronous control and image processing module and an underwater laser radar module installed inside a sealed waterproof housing; The underwater laser radar module includes a pulsed blue-green laser emission module, a single-photon TG-SPAD array detection imaging module, and a multi-channel power supply and signal transmission module; Among them, the synchronous control and image processing module controls the multi-channel power supply and signal transmission module to realize the signal control of the underwater system, and performs blind pixel correction and underwater single-photon image denoising on the data collected by the TG-SPAD detection imaging module to obtain the target two-dimensional spatial intensity map, and continuously obtains the intensity image matrix of multiple target scenes through gated offset or gate width modulation, realizing underwater target detection intensity imaging and high-precision three-dimensional image reconstruction based on the single-photon SPAD array and image display; The pulsed blue-green laser emission module emits laser for target scene illumination; The single-photon TG-SPAD array detection module has two modes: intensity imaging and gated imaging. The intensity imaging mode controls the amount of light entering each image by setting the integration time of the single-photon TG-SPAD array detector, and collects single or multiple intensity images by setting the number of frames; the gated imaging mode uses time-gated TG technology. The SPAD array detector only turns on and records the photon information reaching the SPAD array detector within the preset delay time gate window synchronously controlled with the pulsed laser, thereby effectively eliminating the serious backscattering and other interference noise of the water body in front of the imaging target scene; the main parameters of the TG-SPAD imaging mode are integration time, number of frames, number of gate steps per frame, gate width, and gate delay offset. By adjusting the parameter settings and applying algorithm processing, underwater single-photon two-dimensional and three-dimensional imaging at different distances can be achieved; The multi-channel power supply and signal transmission module is used to provide isolated power supply for each module of the system and realize signal communication and control between the underwater system and the onshore host computer.

Images