Simulation Method and System for Single-Photon LiDAR Imaging of Space Debris

By using a 3D mesh model and a Poisson model, the radial distance and reflectivity are accurately calculated, solving the problem of insufficient accuracy in imaging simulation in existing technologies and achieving higher quality space debris imaging simulation.

CN123066792APending Publication Date: 2026-11-10HUAZHONG UNIV OF SCI & TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611582483.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-13
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

Existing single-photon lidar imaging simulation methods cannot accurately reflect the impact of the actual size, surface undulations, and attitude changes of distant space debris targets on the number of echo photons, resulting in low accuracy of imaging simulation.

Method used

A 3D mesh model of space fragments is used and rendered into a 2D image with the same resolution as the SPAD array. A ray array is generated for intersection detection, and the radial distance, equivalent reflectivity and effective reflective area are calculated. The depth map and intensity map are reconstructed by combining the Poisson model and TCSPC histogram processing, and denoising and accurate modeling of background photon count rate are performed.

Benefits of technology

It improves the simulation accuracy of long-distance space debris imaging, reduces the calculation deviation of echo photon count, enhances the quality of depth and intensity maps, and reduces the impact of noise.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123066792A_ABST
    Figure CN123066792A_ABST
Patent Text Reader

Abstract

This invention discloses a single-photon lidar imaging simulation method and system for space debris, belonging to the field of photoelectric detection imaging technology. The invention employs a three-dimensional mesh model of the space debris target, reflecting the influence of the target's actual size and surface undulations on the spatial distribution of echo photons. The model is rendered into a two-dimensional image with the same resolution as a SPAD array. A ray array is generated based on the array's resolution and projected onto the surface of the three-dimensional mesh model for intersection detection, accurately calculating the true physical values ​​such as radial distance, equivalent reflectivity, and effective reflective area. Based on this, an echo photon expectation calculation formula is designed to calculate the echo photon expectation based on radial distance, equivalent reflectivity, and effective reflective area. This formula also considers the squared distance attenuation caused by two-way propagation, causing the echo photon expectation of distant space targets to attenuate on the order of the fourth power of the distance, thus improving the accuracy of the imaging simulation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of photoelectric detection and imaging technology, and more specifically, relates to a single-photon lidar imaging simulation method and system for space debris. Background Technology

[0002] Space debris is a typical non-cooperative target in the space environment. With the increasing number of spacecraft in orbit and space activities, the impact of space debris on spacecraft operational safety and space resource utilization is becoming increasingly prominent. Long-range detection and imaging of space debris can provide fundamental data for space debris monitoring, target characteristic analysis, orbit identification, and the design of related detection systems. Therefore, establishing imaging simulation methods that can reflect the geometric characteristics, propagation conditions, and detector response characteristics of space debris is of great significance.

[0003] Visible light and infrared passive imaging rely on the target's own radiation or external illumination conditions. In space environments such as long distances, low illumination, and conditions with or against the light, target contrast and imaging quality are easily limited. Active laser imaging, on the other hand, can illuminate the target with a controlled light source, acquiring target distance information while reducing the impact of ambient lighting changes on the detection process. Therefore, it is suitable for long-range space debris imaging research. Lidar, by actively emitting lasers and receiving target echoes, can obtain target distance information based on the echo's flight time, making it one of the effective means of active imaging and detection of space targets.

[0004] Space debris typically exhibits characteristics such as small size, long distance, uncertain attitude, and complex surface scattering properties. For non-cooperative targets at distances of several kilometers to hundreds of kilometers, the emitted beam undergoes spatial expansion during propagation, resulting in a limited number of photons intercepted by the target. Furthermore, photons reflected from the target must undergo further distance attenuation before entering the receiving aperture, often resulting in effective echoes at the single-photon level. Simultaneously, background light from direct sunlight, Earth reflection, and lunar reflection, along with detector dark counts, time jitter, dead time, and afterpulses, all contribute to the time-correlated single-photon count histogram. These factors are interconnected, and relying solely on geometric rendering or simple Gaussian noise layering is insufficient to reflect the actual active single-photon imaging conditions of distant space debris. Therefore, single-photon lidar imaging simulation methods are frequently employed for imaging.

[0005] However, existing single-photon lidar imaging simulation methods typically simplify space debris targets into single-point targets or planar targets with uniform reflectivity, and calculate the number of echo photons using the target's nominal distance, overall cross-sectional area, and preset reflectivity. This makes it difficult to reflect the influence of the target's actual size, surface undulations, and attitude changes on the spatial distribution of echo photons. Especially under long-range imaging conditions, the projected size of space debris targets on the detection surface is small, and their outlines may only occupy a portion of the pixels. Furthermore, the distance, orientation, and effective reflective area of ​​the target surface differ for different pixels. Using uniform target parameters can easily lead to inaccurate calculations of the number of echo photons for each pixel.

[0006] Furthermore, in echo photon link modeling, existing methods typically treat targets as point targets or extended targets. For small, distant targets with an effective reflective area that does not change with detection distance and whose projected range is smaller than the laser illumination area, the number of echo photons typically decreases by the fourth power of the distance. For extended targets that fully cover the laser illumination area or the corresponding probe pixel's projected range, the number of echo photons typically decreases by the square of the distance because the effective area involved in reflection increases with the square of the distance. This process usually requires pre-determining or manually setting the target type or the target's effective reflective area. However, the projected size of distant space debris may be on a similar order of magnitude to the projected range of the probe pixel on the target plane. Different pixels of the same target may simultaneously exist in states of complete coverage, partial coverage, and no coverage, and these states change with the detection distance and target attitude. If a single distance attenuation relationship or a fixed effective reflective area is used for the entire target, it is difficult to accurately describe the number of echo photons received by each pixel under different coverage states, which may lead to calculation deviations in the expected number of echo photons over different time periods, affecting the simulation accuracy of depth and intensity maps. Summary of the Invention

[0007] In view of the above-mentioned defects or improvement needs of the existing technology, the present invention provides a single-photon lidar imaging simulation method and system for space debris, so as to solve the technical problem of low accuracy of existing imaging simulation.

[0008] To achieve the above objectives, in a first aspect, the present invention provides a single-photon lidar imaging simulation method for space debris, applied to a single-photon lidar detection system, comprising: Obtain a 3D mesh model of the space debris target; with the goal of rendering the 3D mesh model into a 2D image with the same resolution as the SPAD array in the system, generate a ray array based on the array resolution and project it onto the surface of the 3D mesh model for intersection detection; for each pixel position in the 2D image, calculate the radial distance from the optical center of the system along the ray direction to the corresponding object point, as well as the equivalent reflectivity and effective reflective area of ​​the object point, based on the ray hit results within it; After the transmitting unit in the system emits a laser pulse toward the array, it targets each SPAD detector s in the array at the [missing information - likely a specific time or position]. k unit time period Perform the following calculations: The expected echo photon received by s under a single laser pulse ;based on dark count rate of s and the corresponding background photon count rate ,calculate Triggering expectation of the next s Substituting this into the Poisson model yields... The probability of a probe being triggered in the next s is then calculated. The probability that the next s will be triggered at least once. And s records the probability of detecting the event. ;in, The number of photons contained in a single emission pulse; for Normalized time response; , , These represent the pixel positions corresponding to the system's optical center along the ray direction to point s. The radial distance to the corresponding object point, the equivalent reflectivity of the object point, and the effective reflective area; and These are the optical transmittance and emission solid angle of the emitting unit, respectively. Let s be the effective receiving area; Atmospheric transmittance; The optical transmittance of the receiving unit in the system; ; This refers to the number of unit time periods during the entire detection process; probability , … and Nonnegation and normalization constitute the first discrete probability distribution corresponding to s; For each SPAD detector, each laser pulse emitted by the transmitting unit is independently sampled according to the corresponding first discrete probability distribution to obtain the TCSPC histogram of each SPAD detector, and then the depth map and intensity map of the space debris target are reconstructed.

[0009] More preferably, the above Normalized time response ;in, unit time period The starting time; Speed ​​of light; composite standard deviation ; The pulse width of the laser pulse; This refers to the time jitter of the SPAD; The above trigger expectation for: ;in, Let s be the photon detection probability; The duration of a unit of time interval.

[0010] More preferably, The probability that the next s will be detected at least once and The probability of detecting the event is recorded in the next s. They are respectively: and ;in, For the first j unit time period The probability that the inner s is triggered by noise at least once.

[0011] More preferably, the above-mentioned single-photon lidar imaging simulation method further includes: after obtaining the intensity map of the space debris target, performing a noise reduction operation on the obtained intensity map of the space debris target: For each SPAD detector s, calculate the mean noise value of its inner group interval from the TCSPC histogram. The values ​​are then substituted into the Poisson model for calculation. The resulting probabilities are used as the probability that s is triggered by noise once within each unit time period. This allows for the calculation of the probability that s is triggered by noise at least once within each unit time period and the probability that it is mistakenly recorded as a detection event after being triggered by noise. Let s be the photon detection probability; The duration of a unit of time; probability , … and Nonnegation and normalization constitute the second discrete probability distribution corresponding to s; and The first k unit time period The probability that s is triggered by noise at least once and the probability that s is mistakenly recorded as a detection event after being triggered by noise; For each SPAD detector, each laser pulse emitted by the transmitting unit is independently sampled according to the corresponding second discrete probability distribution to obtain the TCSPC histogram of each SPAD detector due to noise, and then the estimated intensity map of the noise is reconstructed. The intensity value at each pixel location in the intensity map of the space debris target is subtracted from the intensity value at the corresponding pixel location in the estimated intensity map. When the difference is less than 0, the intensity value at that pixel location in the intensity map of the space debris target is set to 0, thereby achieving noise reduction of the intensity map of the space debris target.

[0012] More preferably, The probability that the next s is triggered by noise at least once. and The probability that the event 's' is mistakenly recorded as a probe event after being triggered by noise. They are respectively: and ;in, For the first j unit time period The probability that the inner s is triggered by noise at least once.

[0013] More preferably, after obtaining the TCSPC histogram of the SPAD detector or the TCSPC histogram of the SPAD detector due to noise, the TCSPC histogram is further subjected to post-pulse correction and non-extended dead time statistical correction.

[0014] More preferably, the aforementioned background photon count rate for: Among them, the spectral radiance formed by the sun directly shining at time s, the sunlight reflected by the Earth at time s, and the sunlight reflected by the moon at time s is in the band. Average value within ; Let be the bandwidth of the filter in s; For the wavelength band of the filter in s; , , All are coefficients; The spectral radiance formed when the sun shines directly on you at wavelength s The value at; The spectral radiance formed by the Earth reflecting sunlight up to s at wavelength The value at; The spectral radiance formed by sunlight reflected from the moon at wavelength s The value at; Solar spectral irradiance at Earth's orbit in wavelength The value at; For Earth at wavelength Spectral albedo at that location; For the moon at wavelength Spectral albedo at that location; The receiving solid angle of the receiving unit; It is Planck's constant; The speed of light; The wavelength at which the laser is emitted by the emitting unit.

[0015] More preferably, after obtaining the three-dimensional mesh model, center correction, scale reconstruction and attitude transformation operations are also performed on it.

[0016] More preferably, the above aims to render the three-dimensional mesh model into a two-dimensional image with the same resolution as the SPAD array in the system. A ray array is generated based on the array resolution and projected onto the surface of the three-dimensional mesh model for intersection detection. For each pixel in the two-dimensional image, based on the ray hit results within it, the radial distance from the system's optical center along the ray direction to the corresponding object point, as well as the equivalent reflectivity and effective reflective area of ​​that object point, are calculated, including: With the goal of rendering the 3D mesh model into a 2D image with the same resolution as the SPAD array in the system, an oversampling strategy is used to generate a ray array based on the array resolution, and the array is projected onto the surface of the 3D mesh model for intersection detection. For each pixel in the 2D image, the hit results of all sub-pixel rays within it are aggregated, and the radial distance from the optical center of the system along the ray direction to the corresponding object point, as well as the equivalent reflectivity and effective reflective area of ​​the object point, are calculated.

[0017] More preferably, the system optical center is located at the pixel position corresponding to s along the ray direction. radial distance of the corresponding object point The equivalent reflectance of the object point and effective reflective area They are respectively: , and ;in, The pixel position corresponding to s The set of subpixel rays that hit the surface of the 3D mesh model at the location; For the system optical center along the sub-pixel ray l To pixel position The distance to the corresponding object point; The pixel position corresponding to s The number of sub-pixel ray hits at a given location; when the pixel position corresponding to s... When the target is hit, the corresponding base reflectivity For space debris targets at pixel locations The material reflectivity at the location; when the pixel position corresponding to s When the target is not hit, the corresponding base reflectivity ; Directional scattering solid angle ; pixel position The normal vector at the corresponding object point; Let be the direction vector of the optical axis of s in the world coordinate system; Preset to positive numbers; The hit rate; This is the oversampling factor; The number of all sub-pixel rays in the pixel corresponding to s; and These are the focal length parameters in the horizontal and vertical directions of the camera intrinsic parameter matrix, respectively.

[0018] In a second aspect, the present invention provides a single-photon lidar detection system, comprising: a SPAD array, a transmitting unit, a receiving unit, and a processing unit; The processing unit is used to execute the single-photon lidar imaging simulation method for space debris provided in the first aspect of the present invention.

[0019] In summary, the above-described technical solutions conceived in this invention can achieve the following beneficial effects: 1. This invention provides a single-photon lidar imaging simulation method for space debris. It employs a three-dimensional mesh model of the space debris target, reflecting the influence of the target's actual size and surface undulations on the spatial distribution of echo photons. The method renders the three-dimensional mesh model as a two-dimensional image with the same resolution as the SPAD array in the system. A ray array is generated based on the array's resolution and projected onto the surface of the three-dimensional mesh model for intersection detection. This allows for accurate calculation of physical values ​​such as radial distance, equivalent reflectivity, and effective reflective area. Based on this, a formula for calculating the expected echo photon distribution is designed. This method calculates the echo photon expectation based on radial distance, equivalent reflectivity, and effective reflective area. Radial distance determines the echo arrival time and distance attenuation during the outward and return journeys. Effective reflective area reflects the actual coverage of the target within the corresponding pixel, and equivalent reflectivity reflects the influence of the target surface orientation on the number of echo photons. This allows changes in target size, attitude, and pixel coverage to directly affect the echo photon expectation calculation for each detected pixel, reducing biases caused by using the overall target distance, fixed reflective area, and uniform reflection parameters. The formula simultaneously considers the squared distance attenuation generated during outward beam spread and return reception. For small targets at long distances where the effective reflective area does not change with distance, the echo photon expectation attenuates on the order of the fourth power of the distance. When the target only covers a portion of the pixels, the echo photon expectation is adjusted using the effective reflective area of ​​the corresponding pixel, thus more accurately describing the echo characteristics of spatial debris under different distances, attitudes, and pixel coverage states, improving the accuracy of depth and intensity map imaging simulations.

[0020] 2. Furthermore, the single-photon lidar imaging simulation method provided by the present invention performs a noise reduction operation on the intensity map of the space debris target after obtaining the intensity map of the space debris target, thereby further improving the quality of the obtained intensity map.

[0021] 3. Furthermore, the single-photon lidar imaging simulation method provided by the present invention, after obtaining the TCSPC histogram of the SPAD detector or the TCSPC histogram of the SPAD detector due to noise, also performs post-pulse correction and non-extended dead-time statistical correction on the TCSPC histogram; through the above correction, the deviation of the ideal independent detection model from the actual total count can be reduced, and the obtained TCSPC histogram is closer to the statistical response of the actual detector in terms of additional count and count saturation.

[0022] 4. Furthermore, the single-photon lidar imaging simulation method provided by this invention calculates the background photon count rate. The study simultaneously considered the spectral radiance generated when sunlight directly hits the SPAD, when sunlight reflected from the Earth reaches the SPAD, and when sunlight reflected from the Moon reaches the SPAD. Spectral radiation generated by direct sunlight, Earth reflection, and lunar reflection were modeled separately. The spatial background photon count rate received by a single SPAD was calculated from the comprehensive spectral radiance within the observed band. Taking into account the background radiation from the Sun, Earth, and Moon, the resulting background photon count rate... The accuracy is high, which further improves the accuracy of imaging simulation.

[0023] 5. Furthermore, the single-photon lidar imaging simulation method provided by the present invention performs center correction, scale reconstruction and attitude transformation operations on the three-dimensional mesh model after obtaining it, so as to avoid the target deviating from the imaging field of view due to the original coordinate offset. Attached Figure Description

[0024] Figure 1 A flowchart of a single-photon lidar imaging simulation method for space debris provided in an embodiment of the present invention.

[0025] Figure 2 This is a ground truth distance image of spatial debris provided in an embodiment of the present invention.

[0026] Figure 3 The simulated reconstructed depth map is provided in this embodiment of the invention, obtained from active single-photon imaging simulation of space debris over a distance of 10 kilometers.

[0027] Figure 4 The intensity map is obtained from the simulation of active single-photon imaging of space debris over a distance of 10 kilometers, as provided in an embodiment of the present invention. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0029] To achieve the above objectives, in a first aspect, the present invention provides a single-photon lidar imaging simulation method for space debris, applicable to single-photon lidar detection systems, such as... Figure 1 As shown, it includes: A 3D mesh model of the space debris target is obtained; a 2D image with the same resolution as the SPAD (single-photon lidar detector) array in the system is rendered from the 3D mesh model as the target. A ray array is generated based on the array resolution and projected onto the surface of the 3D mesh model for intersection detection; for each pixel in the 2D image, the radial distance from the optical center of the system along the ray direction to the corresponding object point, as well as the equivalent reflectivity and effective reflective area of ​​the object point, are calculated based on the ray hit results within the pixel. After the transmitting unit in the system emits a laser pulse toward the array, it targets each SPAD detector s in the array at the [missing information - likely a specific time or position]. k unit time period Perform the following calculations: The expected echo photon received by s under a single laser pulse ;based on dark count rate of s and the corresponding background photon count rate ,calculate Triggering expectation of the next s Substituting this into the Poisson model yields... The probability of being triggered (including noise triggering and target triggering) once in the next s is then calculated. The probability that the next s will be triggered at least once. And s records the probability of detecting the event. ;in, The number of photons contained in a single emission pulse; for Normalized time response; , , These represent the pixel positions corresponding to the system's optical center along the ray direction to point s. The radial distance to the corresponding object point, the equivalent reflectivity of the object point, and the effective reflective area; and These are the optical transmittance and emission solid angle of the emitting unit, respectively. Let s be the effective receiving area; Atmospheric transmittance; The optical transmittance of the receiving unit in the system; ; This refers to the number of unit time periods during the entire detection process; probability , … and Nonnegation and normalization constitute the first discrete probability distribution corresponding to s; For each SPAD detector, each laser pulse emitted by the transmitting unit is independently sampled according to the corresponding first discrete probability distribution to obtain the TCSPC histogram of each SPAD detector, and then the depth map and intensity map of the space debris target are reconstructed.

[0030] It should be noted that the method for reconstructing depth maps based on TCSPC histograms can be any existing method, such as peak method, first photon method, statistical modeling method, etc. There is no limitation here, but the peak method is preferred. For each SPAD, the time corresponding to the peak of the corresponding TCSPC histogram is obtained as the flight time, and then the physical distance corresponding to the corresponding pixel is calculated using the beam, thereby obtaining the depth map of the space debris target.

[0031] It should be noted that any existing method can be used to reconstruct the intensity map based on the TCSPC histogram, such as the direct integration method or the traditional model-driven method. There is no limitation here. The direct integration method is preferred. The TCSPC histogram of each SPAD detector is integrated over time to obtain the intensity value at the pixel position corresponding to each SPAD detector, thereby obtaining the intensity map of the space debris target.

[0032] In one alternative implementation, the above Normalized time response ;in, unit time period The starting time; Speed ​​of light; composite standard deviation ; The pulse width of the laser pulse; This refers to the time jitter of the SPAD; The above trigger expectation for: ;in, Let s be the photon detection probability; The duration of a unit of time interval.

[0033] In one alternative implementation, The probability that the next s will be detected at least once and The probability of detecting the event is recorded in the next s. They are respectively: and ;in, For the first j unit time period The probability that the inner s is triggered by noise at least once.

[0034] Preferably, in an optional implementation, after obtaining the TCSPC histogram of the SPAD detector, the TCSPC histogram is further subjected to post-pulse correction and non-extended dead time statistical correction.

[0035] Preferably, in an optional implementation, the above-mentioned single-photon lidar imaging simulation method further includes: after obtaining the intensity map of the space debris target, performing a noise reduction operation on the obtained intensity map of the space debris target: For each SPAD detector s, calculate the mean noise value of its inner group interval from the TCSPC histogram. The values ​​are then substituted into the Poisson model for calculation. The resulting probabilities are used as the probability that s is triggered by noise once within each unit time period. This allows for the calculation of the probability that s is triggered by noise at least once within each unit time period and the probability that it is mistakenly recorded as a detection event after being triggered by noise. Let s be the photon detection probability; The duration of a unit of time; probability , … and Nonnegation and normalization constitute the second discrete probability distribution corresponding to s; and The first k unit time period The probability that s is triggered by noise at least once and the probability that s is mistakenly recorded as a detection event after being triggered by noise; For each SPAD detector, each laser pulse emitted by the transmitting unit is independently sampled according to the corresponding second discrete probability distribution to obtain the TCSPC histogram of each SPAD detector due to noise, and then the estimated intensity map of the noise is reconstructed. The intensity value at each pixel location in the intensity map of the space debris target is subtracted from the intensity value at the corresponding pixel location in the estimated intensity map. When the difference is less than 0, the intensity value at that pixel location in the intensity map of the space debris target is set to 0, thereby achieving noise reduction of the intensity map of the space debris target.

[0036] In one alternative implementation, The probability that the next s is triggered by noise at least once. and The probability that the event 's' is mistakenly recorded as a probe event after being triggered by noise. They are respectively: and ;in, For the first j unit time period The probability that the inner s is triggered by noise at least once.

[0037] Preferably, in an optional implementation, after obtaining the TCSPC histogram of the SPAD detector due to noise, the TCSPC histogram is further subjected to post-pulse correction and non-extended dead time statistical correction.

[0038] In one alternative implementation, the aforementioned background photon count rate for: Among them, the spectral radiance formed by the sun directly shining at time s, the sunlight reflected by the Earth at time s, and the sunlight reflected by the moon at time s is in the band. Average value within ; Let be the bandwidth of the filter in s; For the wavelength band of the filter in s; , , All are coefficients; The spectral radiance formed when the sun shines directly on you at wavelength s The value at; The spectral radiance formed by the Earth reflecting sunlight up to s at wavelength The value at; The spectral radiance formed by sunlight reflected from the moon at wavelength s The value at; Solar spectral irradiance at Earth's orbit in wavelength The value at; For Earth at wavelength Spectral albedo at that location; For the moon at wavelength Spectral albedo at that location; The receiving solid angle of the receiving unit; It is Planck's constant; The speed of light; The wavelength at which the laser is emitted by the emitting unit.

[0039] Preferably, in one optional implementation, after obtaining the three-dimensional mesh model, center correction, scale reconstruction, and attitude transformation operations are also performed on it.

[0040] Preferably, in one optional implementation, the above aims to render the three-dimensional mesh model into a two-dimensional image with the same resolution as the SPAD array in the system. A ray array is generated based on the array resolution and projected onto the surface of the three-dimensional mesh model for intersection detection. For each pixel in the two-dimensional image, based on the ray hit results within it, the radial distance from the system's optical center along the ray direction to the corresponding object point, as well as the equivalent reflectivity and effective reflective area of ​​that object point, are calculated, including: With the goal of rendering the 3D mesh model into a 2D image with the same resolution as the SPAD array in the system, an oversampling strategy is used to generate a ray array based on the array resolution, and the array is projected onto the surface of the 3D mesh model for intersection detection. For each pixel in the 2D image, the hit results of all sub-pixel rays within it are aggregated, and the radial distance from the optical center of the system along the ray direction to the corresponding object point, as well as the equivalent reflectivity and effective reflective area of ​​the object point, are calculated.

[0041] In one optional implementation, the system optical center is located at the pixel position corresponding to s along the ray direction. radial distance of the corresponding object point The equivalent reflectance of the object point and effective reflective area They are respectively: , and ;in, The pixel position corresponding to s The set of subpixel rays that hit the surface of the 3D mesh model at the location; For the system optical center along the sub-pixel ray l To pixel position The distance to the corresponding object point (i.e., the hit distance); The pixel position corresponding to s The number of sub-pixel ray hits at a given location; when the pixel position corresponding to s... When the target is hit, the corresponding base reflectivity For space debris targets at pixel locations The material reflectivity at the location; when the pixel position corresponding to s When the target is not hit, the corresponding base reflectivity ; Directional scattering solid angle ; pixel position The normal vector at the corresponding object point; Let be the direction vector of the optical axis of s in the world coordinate system; Preset to positive numbers; The hit rate; This is the oversampling factor; The number of all sub-pixel rays in the pixel corresponding to s; and These are the focal length parameters in the horizontal and vertical directions of the camera intrinsic parameter matrix, respectively.

[0042] In a second aspect, the present invention provides a single-photon lidar detection system, comprising: a SPAD array, a transmitting unit, a receiving unit, and a processing unit; The processing unit is used to execute the single-photon lidar imaging simulation method for space debris provided in the first aspect of the present invention.

[0043] The related technical solutions are the same as the single-photon lidar imaging simulation method for space debris provided in the first aspect of this invention, and will not be described in detail here.

[0044] To further illustrate the single-photon lidar imaging simulation method for space debris provided by this invention, a specific embodiment is described in detail below: Example 1 This embodiment provides a single-photon lidar imaging simulation method for space debris, involving lidar simulation imaging of non-cooperative targets in space. The active single-photon imaging simulation of space debris is completed in six steps, namely: simulation of theoretical range image and reflectivity of space debris, expected modeling of echo photons of space targets by SPAD lidar, modeling of space background photons, dark count modeling, macro-pixel detection TCSPC histogram statistics and depth map imaging, and intensity map imaging.

[0045] 1. Simulation of theoretical distance image and reflectivity of space debris This step obtains a 3D mesh model of the space debris target, and sequentially completes center correction, real-scale reconstruction, target attitude transformation, oversampling ray tracing, and sub-pixel result aggregation to generate a theoretical distance image and the physical true value of reflectivity.

[0046] A 3D mesh model of the space debris target is acquired, and center correction, scale reconstruction, and attitude transformation operations are performed on it to simulate the size and attitude of the real space debris target. The goal is to project the 3D mesh model of the space debris into a 2D image with the same resolution as the SPAD array in the system. Based on pinhole imaging, the pixel positions corresponding to each SPAD detector in the array are determined according to a preset oversampling factor. Multiple sub-pixel sampling positions are determined internally. Starting from the optical center of the system (the center point of the small hole in this embodiment), sampling rays are projected onto the three-dimensional mesh model of the space fragment through each sub-pixel sampling position, and intersection detection is performed between each sampling ray and the surface of the three-dimensional mesh model.

[0047] Specifically, taking triangular mesh elements as the mesh unit of the 3D mesh model as an example, the set of triangular mesh vertices in the 3D mesh model of the space debris target is first read. The original mesh may use arbitrary modeling units and its geometric center may be off-center from the origin. First, calculate the mean center of all vertices: Translate the vertex to - Calculate the bounding box (the smallest cuboid that can completely enclose the 3D mesh model) aligned with the mesh axes after translation. , and directional span , , ,make The scaling factor for the actual scale is: , The actual size of the space debris target; The maximum bounding box span of the 3D mesh model after center correction along the three coordinate axes is used to characterize the maximum scale of the original mesh model; the updated vertices are... When it is necessary to change the attitude of a space debris target, the rotation angle is used as the basis for the change. and Construction around shaft and Rotation matrix of axis , and in accordance with Update vertices. The above processing performs scale transformation and pose rotation of the target around its own geometric center, avoiding the original coordinate offset that would cause the target to deviate from the imaging field of view.

[0048] After scaling and orientation transformation of the 3D mesh, oversampled ray tracing is performed on the target. For Output array (i.e., SPAD array resolution), according to the oversampling factor L structure Ray array. Let the focal length of the original intrinsic parameter be... , , light center is , Then the oversampling intrinsic parameter is: .

[0049] The world-to-camera transformation is obtained by inverting the camera-to-world coordinate transformation matrix. Pinhole camera rays are generated using the oversampling intrinsic parameters and the world-to-camera transformation, and their intersection with the triangular mesh is determined. For each SPAD detector, the following calculations are performed based on the ray hit results within its corresponding pixel: the target coverage ratio is calculated based on the number of sampled rays hitting the 3D mesh model and the total number of sampled rays within that pixel. Based on the hit distance of each hit sampling ray within the pixel, calculate the radial distance from the system optical center along the ray direction to the surface of the debris target, and the distance from the system optical center along the ray direction to the pixel position corresponding to the SPAD. radial distance of the corresponding object point Based on the target coverage ratio and the single-pixel projected area at that radial distance, calculate the effective area of ​​the target actually participating in reflection within that pixel. The normal vectors of the triangular facets corresponding to each hit sampling ray within the pixel are averaged and normalized to obtain the surface orientation information of the fragment target corresponding to the pixel, i.e., the pixel position. The normal vector at the corresponding object point Based on the target surface orientation information and the system optical axis orientation Calculate the scattering solid angle corresponding to the pixel based on the included angle relationship. When the pixel position corresponding to SPAD... When the target is hit, the corresponding base reflectivity For space debris targets at pixel locations The material reflectivity at the location; when the pixel position corresponding to s When the target is not hit, the corresponding base reflectivity Directional scattering solid angle ; pixel position The normal vector at the corresponding object point; Let $\mathbf{ ...

[0050] The pixel position corresponding to SPAD The set of sub-pixel rays that hit the surface of the 3D mesh model at the point of impact, i.e., the output pixel. Internal A subpixel ray, the number of rays within it is . For the first l The hit distance corresponding to each sampling ray; and These are the focal length parameters in the horizontal and vertical directions of the camera intrinsic parameter matrix, respectively; For the first l The normal vector of the triangular facet corresponding to the hit sampling ray; To prevent the scattering solid angle from being a preset positive number of zero, this embodiment uses 0.01; , and The positions of the SPAD detectors are matched one by one, serving as the input for subsequent calculations of the expected echo photons received by each SPAD detector.

[0051] The radial distance of the missed pixels is set to the preset background depth. And it distinguishes itself from the effective target pixels using a target mask. (Compared to a conventional graphics rendering output camera) The different shaft components in this embodiment The Euclidean propagation distance along the ray can be directly used. Flight time calculation.

[0052] After obtaining the radial distance and sub-pixel hit ratio, the target area and reflectivity are further generated to obtain the true physical values. Under the small-angle approximation of a pinhole, the distance... The projected footprint area of ​​a single output pixel is After considering sub-pixel coverage, the target area within a pixel is .right The mean value of the normal vectors of the corresponding triangular faces is calculated and normalized to obtain the unit normal vector of the pixel surface. Let the direction of the camera's optical axis in the world coordinate system be... The following directional scattering solid angle is then used: .

[0053] In this embodiment A value of 0.01 sr is used to avoid the denominator being zero when the normal vector is parallel to the optical axis; for missed pixels, their directional scattering solid angle is calculated. Set as The base reflectance is set to 0. For a hit pixel, the base reflectance is taken as the input target material reflectance. The effective reflectance is obtained based on the basic reflectance and the scattering solid angle. .

[0054] Through the above processing, a radial distance map, an intra-pixel target area map, a normal vector map, a scattering solid angle map, a basic reflectivity map, and an equivalent reflectivity map are generated. Among them, the radial distance, effective reflective area, and effective reflectivity are used as inputs to the subsequent space target echo photon expectation model, thereby connecting the three-dimensional geometric features of space debris with laser echo photon calculations.

[0055] 2. SPAD LiDAR Space Target Echo Photon Expectation Modeling The number of photons contained in a single emission pulse is: ; The energy of a single laser pulse; The wavelength of the laser; It is Planck's constant; The speed of light; with the divergence angle Construct a small-angle launch solid angle Radial distance The emitted photon density at that location is The geometric footprint of a single SPAD field of view on the target plane is approximately: To prevent a pixel from receiving an echo exceeding the actual target coverage area within that pixel, let: ; The field of view is the full angle.

[0056] Combining basic reflectivity and surface directionality, the pixel-level laser cross-section is The number of photons intercepted and scattered by the target is equal to the photon density emitted at the target location and The product of the receiving aperture and the proportion of light scattered from all directions of space is... Two-way atmospheric transmittance .

[0057] The entire detection process was divided into multiple time segments, in the first segment... k unit time period The expected echo photons received by a SPAD under a single laser pulse are: ;in, Normalized time response ;in, unit time period The start time; the laser pulse width SPAD time jitter Treating them as independent Gaussian broadenings, we obtain the composite standard deviation. ; The pulse width of the laser pulse; This refers to the time jitter of the SPAD.

[0058] The above formula includes both the outbound beam spread and the squared distance attenuation caused by the return receiving solid angle, and therefore the overall attenuation is on the order of the fourth power of the distance, which is suitable for describing the link between distant point targets in space debris.

[0059] 3. Spatial background photon modeling This step constructs a space background photon model, weights the spectral radiance formed by direct sunlight, Earth reflection, and lunar reflection, and calculates the space background photon rate in conjunction with the receiving system parameters.

[0060] Specifically, this step models the spectral radiation formed by direct sunlight, Earth reflection, and lunar reflection, and calculates the spatial background photon rate received by a single SPAD from the comprehensive spectral radiance within the observation band.

[0061] This implementation method uses solar temperature Constructing the Planck spectral radiance: ; where k is the Boltzmann constant.

[0062] Based on the solar radius Earth-Sun Distance The geometric proportions are used to obtain the solar spectral irradiance at Earth's orbit, specifically, Solar spectral irradiance at Earth's orbit in wavelength The value at; .

[0063] Earth and Moon, respectively, according to wavelength-dependent albedo , The Lambertian reflector was processed to obtain the spectral radiance of sunlight reflected from Earth up to s at wavelengths of 1000 nm. The value at the location The spectral radiance formed by the Earth reflecting sunlight up to s at wavelength The value at the location Let the pixel fill or visibility weights for the Sun, Earth, and Moon be respectively... , and The combined spectral radiance within the filter band is the average value, i.e., the spectral radiance formed by direct sunlight up to s, sunlight reflected from the Earth up to s, and sunlight reflected from the Moon up to s, within the filter band. Average value within ; Let be the bandwidth of the filter in s; For the wavelength band of the filter in s.

[0064] Full field of view The conical solid angle adopts a numerically stable form, that is, the receiving solid angle of the receiving unit. Background power is obtained from the receiver aperture, field solid angle, and filter bandwidth. Divide by the single-photon energy at the center wavelength to obtain the background incident photon rate. .

[0065] 4. Dark Count Rate Modeling SPAD dark counting is considered as a Poisson arrival process that is independent of the target echo photons and the spatial background photons.

[0066] 5. TCSPC Histogram Statistics and Depth Map Imaging This step constructs a TCSPC detection statistical model, converting target echo photons, spatial background photons, and dark counts into Poisson arrival means for each time interval, and forming a discrete distribution based on the first photon survival probability; multiple laser pulses are independently sampled, and statistical corrections for subsequent pulses and non-extended dead time are added to form a TCSPC histogram, and the depth map is reconstructed based on the peak time of the histogram.

[0067] Specifically, this step first establishes a TCSPC detection statistical model using a single SPAD and a single laser pulse as the basic detection unit, and then extends the detection statistics to macro pixels containing multiple SPADs.

[0068] Based on the gating start time Gating width and time group interval (Bin) width Establishing time series Only those that meet the requirements are retained. The time binning interval (Bin). The two-way flight time of space debris at the corresponding SPAD pixel location. .

[0069] In the k unit time period When the signal, background, and dark counts of the SPAD under a single pulse reach the mean, i.e., the trigger expectation is achieved. for: ;in, The photon detection probability of SPAD; The speed of light; The duration of a unit of time interval.

[0070] For the Poisson arrival process, the expectation will be triggered. Substituting this into the Poisson model yields... Next (i.e., the first) The probability of a SPAD being triggered for one probe within a time period (Bin), thus obtaining... The probability that SPAD will be triggered at least once. Since a SPAD only records the first primary probe event within the gating window of a single pulse, The probability of SPAD recording a probe event is: .

[0071] The probability that no Bin is detected in all unit time periods (i.e., all time Bin) is .Will , … and Nonnegation and normalization yield The discrete probability distribution of each mutually exclusive result is used as the first discrete probability distribution of the SPAD.

[0072] For each SPAD detector, each laser pulse emitted by the transmitting unit is independently sampled according to the corresponding first discrete probability distribution to obtain the TCSPC histogram of each SPAD detector. Specifically, for each SPAD, independent sampling is performed for each laser pulse: for each laser pulse, one sample is taken from the discrete probability distribution, and when the sampling result is... , … When any probability corresponds to a unit time interval (time Bin), it is recorded as the first main detection event of this laser pulse, and the count is incremented by 1 within that unit time interval; thus, the corresponding SPAD TCSPC histogram is constructed. When the sampling result is not... , … When any probability in the time interval (time Bin) corresponds to a given probability, it is determined that the current laser pulse will not generate a main detector count.

[0073] To improve the imaging rate, a single output pixel can be expanded to include... Each output pixel is a macro cell of an independent SPAD. One cumulative pulse and A total of [number] SPAD dimensions were executed. Sub-independent discrete result sampling. This space-for-time sampling method makes increasing the number of macro-pixel SPADs statistically comparable to increasing the number of independent cumulative samplings.

[0074] Furthermore, after obtaining the TCSPC histogram, post-pulse correction and non-extended dead time statistical correction are performed on the TCSPC histogram.

[0075] For sampling that has already generated the main probe event, with probability... Determine if a follow-pulse has been generated. Follow-pulse delay. Follow the mean The exponential distribution will Take the integer group interval (Bin) offset and superimpose it on the original main detector group interval; only add the post-pulse count to the histogram if the delayed group interval is still within the gating window.

[0076] This implementation uses a statistical approximation without extended dead time. Let the total pixel count before dead time correction be... Complex frequency is The time of death is The non-extended dead time statistical correction factor is obtained. and with The time group intervals of this pixel are scaled uniformly.

[0077] For each pixel histogram, search along the time dimension for the time group interval with the maximum count, and let it represent the time. The reconstruction depth is If the sum of the counts for all time intervals in the histogram is zero, the depth of that pixel is marked as null; if a count exists, the peak depth is output, thus forming a depth map.

[0078] 6. Intensity map imaging For any SPAD, its corresponding strength The intensity map of the space debris target is obtained by summing the counts at each time interval in its corresponding TCSPC histogram.

[0079] This embodiment denoises the intensity map of the obtained space debris target: For noise estimation, let the noise mean of a single group be the noise mean. The probability is then substituted into the Poisson model for calculation. The obtained probability is used as the probability that the SPAD is triggered by noise once in each unit time period. Then, the probability that the SPAD is triggered by noise at least once in each unit time period and the probability that it is mistakenly recorded as a detection event after being triggered by noise are calculated. Specifically, The probability that the next SPAD is triggered by noise at least once. and The probability that the event 's' is mistakenly recorded as a probe event after being triggered by noise. They are respectively: and ;in, For the first j unit time period The probability that the inner s is triggered by noise at least once.

[0080] probability , … and Nonnegation and normalization constitute the second discrete probability distribution corresponding to s; and The first k unit time period The probability that SPAD is triggered by noise to conduct a single probe and the probability that s is mistakenly recorded as a probe event after being triggered by noise; For each SPAD sensor, each laser pulse emitted by the transmitting unit is independently sampled according to the corresponding second discrete probability distribution to obtain the TCSPC histogram of each SPAD sensor due to noise, and then the estimated intensity map of the noise is reconstructed.

[0081] Furthermore, after obtaining the TCSPC histogram, post-pulse correction and non-extended dead time statistical correction are performed on the TCSPC histogram.

[0082] The intensity value at each pixel location in the intensity map of the space debris target is subtracted from the intensity value at the corresponding pixel location in the estimated intensity map. When the difference is less than 0, the intensity value at that pixel location in the intensity map of the space debris target is set to 0, thereby achieving noise reduction of the intensity map of the space debris target.

[0083] This embodiment first generates a theoretical range image and the physical truth value of reflectivity based on a 3D model of space debris. Then, based on a photon transport model, it establishes the expected photon echo of the space target, the photon background in space, and the dark counting model of single-photon avalanche diodes (SPADs). On this basis, Monte Carlo simulation is performed to simulate the Poisson distribution. Subsequently, macro-pixel time-correlated single-photon counting (TCSPC) is performed to detect and statistically analyze the data and reconstruct the depth map. Finally, the target intensity map is obtained based on the statistical analysis of the echo photon data. This invention achieves unified physical modeling of space debris geometry, photon links, detection statistics, and depth and intensity imaging, providing simulation data for parameter verification, performance prediction, and algorithm validation of active single-photon imaging systems for space debris.

[0084] In this embodiment, physical parameters such as the laser, receiving unit, space environment, SPAD, and TCSPC acquisition can all be adjusted, and the physical true value, TCSPC histogram, depth map, and target intensity map can be output simultaneously. It has high simulation flexibility and can provide a data foundation for parameter demonstration, performance prediction, and algorithm verification of space debris detection schemes.

[0085] Example 2 This embodiment presents the process of generating physical truth values ​​for 10km space debris using the physical truth value generation method provided in Embodiment 1.

[0086] In this embodiment, the following is adopted: Array, single SPAD full-angle field of view Given a target distance of 10,000 m, a target maximum size of 2.5 m, a base reflectivity of 0.6, and an oversampling factor of 4, input a cubic triangular mesh and make the target orbit around... shaft and The axes rotate respectively and After real-scale reconstruction and ray casting, 16 effective target pixels were obtained; the average radial distance of the effective pixels was 9998.75 m, and the maximum target area was approximately [missing value]. The effective pixel scattering solid angle is approximately 0.010003 sr, and the base reflectance map is set to 0.6 for target pixels and 0 for background pixels. This result demonstrates that oversampling aggregation can preserve the hit ratio and actual pixel coverage area even when a small, distant target only covers a small number of output pixels.

[0087] Example 3 Based on the physical truth value of Example 2, a single pulse energy is set. repetition frequency The background photon rate is 0, the gating start time is 65000 ns, the gating width is 4000 ns, the time grouping width is 10 ns, and the cumulative pulse number is 50. The photon detection probability of the SPAD is set to 0.2, and a single macro-pixel contains 9 SPADs. The radial distance, target area, basic reflectivity, and scattering solid angle are substituted into the spatial link, and sampling and histogram reconstruction are completed according to the discrete distribution of the first photon. Figures 2-4 The image shown is a schematic diagram of the simulation results of active single-photon imaging of space debris at a range of 10 kilometers; among which, Figure 2 For the true distance image of spatial debris; Figure 3 To simulate and reconstruct the depth map; Figure 4 This is the intensity map (the original intensity map before denoising). The 16 target pixels are in... The depth accuracy within the error range is 100%, and the average target reconstruction depth is 9999.0 m, consistent with the true average depth of 9998.75 m. This embodiment illustrates the physical closed loop between the geometric truth, the spatial photon link, and the first photon reconstruction.

[0088] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A single-photon lidar imaging simulation method for space debris, characterized in that, Applications in single-photon lidar detection systems include: Obtain a 3D mesh model of the space debris target; with the goal of rendering the 3D mesh model into a 2D image with the same resolution as the SPAD array in the system, generate a ray array based on the array resolution and project it onto the surface of the 3D mesh model for intersection detection; for each pixel position in the 2D image, calculate the radial distance from the optical center of the system along the ray direction to the corresponding object point, as well as the equivalent reflectivity and effective reflective area of ​​the object point, based on the ray hit results within it; After the transmitting unit in the system emits a laser pulse toward the array, it targets each SPAD detector s in the array at the [missing information - likely a specific time or position]. k unit time period Perform the following calculations: Calculate The expected echo photon received by s under a single laser pulse ;based on dark count rate of s and the corresponding background photon count rate ,calculate Triggering expectation of the next s Substituting this into the Poisson model yields... The probability of a probe being triggered in the next s is then calculated. The probability that the next s will be triggered at least once. And s records the probability of detecting the event. ;in, The number of photons contained in a single emission pulse; for Normalized time response; , , These represent the pixel positions corresponding to the system's optical center along the ray direction to point s. The radial distance to the corresponding object point, the equivalent reflectivity of the object point, and the effective reflective area; and These are the optical transmittance and emission solid angle of the emitting unit, respectively. Let s be the effective receiving area; Atmospheric transmittance; The optical transmittance of the receiving unit in the system; ; This refers to the number of unit time periods during the entire detection process; probability , … and Nonnegation and normalization constitute the first discrete probability distribution corresponding to s; For each SPAD detector, each laser pulse emitted by the transmitting unit is independently sampled according to the corresponding first discrete probability distribution to obtain the TCSPC histogram of each SPAD detector, and then the depth map and intensity map of the space debris target are reconstructed.

2. The single-photon lidar imaging simulation method for space debris according to claim 1, characterized in that, The Normalized time response ;in, unit time period The starting time; Speed ​​of light; composite standard deviation ; The pulse width of the laser pulse; This refers to the time jitter of the SPAD; The trigger expectation for: ;in, Let s be the photon detection probability; The duration of a unit of time; The probability that the next s will be detected at least once and The probability of detecting the event is recorded in the next s. They are respectively: and ;in, For the first j unit time period The probability that the inner s is triggered by noise at least once.

3. The single-photon lidar imaging simulation method for space debris according to claim 1, characterized in that, Also includes: After obtaining the intensity map of the space debris target, a noise reduction operation is performed on the obtained intensity map of the space debris target: For each SPAD detector s, calculate the mean noise value of its inner group interval from the TCSPC histogram. The values ​​are then substituted into the Poisson model for calculation. The resulting probabilities are used as the probability that s is triggered by noise once within each unit time period. This allows for the calculation of the probability that s is triggered by noise at least once within each unit time period and the probability that it is mistakenly recorded as a detection event after being triggered by noise. Let s be the photon detection probability; The duration of a unit of time; probability , … and Nonnegation and normalization constitute the second discrete probability distribution corresponding to s; and The first k unit time period The probability that s is triggered by noise at least once and the probability that s is mistakenly recorded as a detection event after being triggered by noise; For each SPAD detector, each laser pulse emitted by the transmitting unit is independently sampled according to the corresponding second discrete probability distribution to obtain the TCSPC histogram of each SPAD detector due to noise, and then the estimated intensity map of the noise is reconstructed. The intensity value at each pixel location in the intensity map of the space debris target is subtracted from the intensity value at the corresponding pixel location in the estimated intensity map. When the difference is less than 0, the intensity value at that pixel location in the intensity map of the space debris target is set to 0, thereby achieving noise reduction of the intensity map of the space debris target.

4. The single-photon lidar imaging simulation method for space debris according to claim 3, characterized in that, The probability that the next s is triggered by noise at least once. and The probability that the event 's' is mistakenly recorded as a probe event after being triggered by noise. They are respectively: and ;in, For the first j unit time period The probability that the inner s is triggered by noise at least once.

5. The single-photon lidar imaging simulation method for space debris according to claim 3, characterized in that, After obtaining the TCSPC histogram of the SPAD detector or the TCSPC histogram of the SPAD detector due to noise, the TCSPC histogram is further corrected by post-pulse correction and non-extended dead time statistical correction.

6. The single-photon lidar imaging simulation method for space debris according to any one of claims 1-5, characterized in that, The background photon count rate for: Among them, the spectral radiance formed by the sun directly shining at time s, the sunlight reflected by the Earth at time s, and the sunlight reflected by the moon at time s is in the band. Average value within ; Let be the bandwidth of the filter in s; For the wavelength band of the filter in s; , , All are coefficients; The spectral radiance formed when the sun shines directly on you at wavelength s The value at; The spectral radiance formed by the Earth reflecting sunlight up to s at wavelength The value at; The spectral radiance formed by sunlight reflected from the moon at wavelength s The value at; Solar spectral irradiance at Earth's orbit in wavelength The value at; For Earth at wavelength Spectral albedo at that location; For the moon at wavelength Spectral albedo at that location; The receiving solid angle of the receiving unit; It is Planck's constant; The speed of light; The wavelength at which the laser is emitted by the emitting unit.

7. The single-photon lidar imaging simulation method for space debris according to any one of claims 1-5, characterized in that, After obtaining the 3D mesh model, center correction, scale reconstruction, and attitude transformation operations are performed on it.

8. The single-photon lidar imaging simulation method for space debris according to any one of claims 1-5, characterized in that, The goal is to render the 3D mesh model into a 2D image with the same resolution as the SPAD array in the system. A ray array is generated based on the array resolution and projected onto the surface of the 3D mesh model for intersection detection. For each pixel in the 2D image, based on the ray hit results within it, the radial distance from the system's optical center along the ray direction to the corresponding object point, as well as the equivalent reflectivity and effective reflective area of ​​that object point, are calculated, including: With the goal of rendering the 3D mesh model into a 2D image with the same resolution as the SPAD array in the system, an oversampling strategy is used to generate a ray array based on the array resolution, and the array is projected onto the surface of the 3D mesh model for intersection detection. For each pixel in the 2D image, the hit results of all sub-pixel rays within it are aggregated, and the radial distance from the optical center of the system along the ray direction to the corresponding object point, as well as the equivalent reflectivity and effective reflective area of ​​the object point, are calculated.

9. The single-photon lidar imaging simulation method for space debris according to claim 8, characterized in that, The system optical center along the ray direction to the pixel position corresponding to s radial distance of the corresponding object point The equivalent reflectance of the object point and effective reflective area They are respectively: , and ;in, The pixel position corresponding to s The set of subpixel rays that hit the surface of the 3D mesh model at the location; For the system optical center along the sub-pixel ray l To pixel position The distance to the corresponding object point; The pixel position corresponding to s The number of sub-pixel ray hits at a given location; when the pixel position corresponding to s... When the target is hit, the corresponding base reflectivity For space debris targets at pixel locations The material reflectivity at the location; when the pixel position corresponding to s When the target is not hit, the corresponding base reflectivity ; Directional scattering solid angle ; pixel position The normal vector at the corresponding object point; Let be the direction vector of the optical axis of s in the world coordinate system; Preset to positive numbers; The hit rate; This is the oversampling factor; The number of all sub-pixel rays in the pixel corresponding to s; and These are the focal length parameters in the horizontal and vertical directions of the camera intrinsic parameter matrix, respectively.

10. A single-photon lidar detection system, comprising: SPAD array, transmitting unit, receiving unit, and processing unit; The processing unit is used to execute the single-photon lidar imaging simulation method for space debris as described in any one of claims 1-9.