High-speed moving target trajectory reconstruction system and method based on space-time correlation
By combining correlation imaging and quantum compression perception, the space-time correlation of single-photon array cameras is used to solve the problem that traditional technology cannot restore the moving target trajectory in extremely low light environments, and high-speed moving target trajectory in extremely low light environments is achieved Reconstruction and noise resistance improve imaging accuracy and efficiency.
Patent Information
- Application Number
- CN202411317627.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-20
- Publication Date
- 2024-11-15
AI Technical Summary
Traditional correlation imaging technology cannot effectively restore the trajectory of moving targets in extremely low light environments, and is susceptible to noise interference in high background noise, so it cannot effectively track and reconstruct the trajectory of high-speed moving targets.
Using a spatial and temporal correlation method, correlation imaging is combined with quantum compression perception, and the spatial and temporal correlation of data is detected by a single photon array camera. By analyzing the spatial and temporal correlation characteristics between adjacent frames, the three-dimensional motion characteristics of high-speed moving objects are extracted, and The pulse generation module and the current modulation module generate extremely narrow pulse lasers to achieve three-dimensional trajectory recovery of the moving target.
It significantly improves the trajectory recovery ability in extremely low-light environments, has good noise resistance, can accurately track and reconstruct the three-dimensional trajectory of high-speed moving targets, and improves the accuracy and efficiency of imaging.
Smart Images

Figure CN118962718A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of optical imaging, and in particular relates to a system and method for reconstructing the trajectory of a high-speed moving target based on time-space correlation. Background Art
[0002] Tracking and reconstructing the trajectory of moving objects are of great significance in daily applications. Tracking and analyzing moving vehicles can predict traffic jams and optimize autonomous driving technology. Recording the movement of cells or microorganisms in biological tissues helps to explore the laws of microbial movement. Analysis of athletes' movement trajectory behavior in sports competitions can improve athlete training methods, etc. Traditional methods of reconstructing object trajectories, such as "instant replay systems", mainly use high-speed cameras to continuously shoot objects, and through algorithm processing and operation, achieve the results of trajectory recovery prediction. However, traditional imaging technology is limited by complex optical design and expensive optical devices, which is not conducive to the integrated design of the system. And with the expansion and demand for detection information dimensions in scientific research, traditional imaging devices cannot meet the multi-dimensional image information detection capabilities.
[0003] Correlation imaging is an active imaging technology that obtains spatial information of objects based on the second-order coherence of light fields. The spatially correlated illumination light field is divided into two beams. The intensity distribution of one beam after free propagation is recorded by a planar array detector, and the other beam is collected by a point detector without spatial resolution after reflection or transmission by the object. Neither detector can image the object alone, but the information of the object can be obtained by the correlation of the signals of the two detectors. Compared with traditional imaging, correlation imaging uses light field intensity correlation imaging. A point detector can be used to collect the light field intensity from the object to perform two-dimensional or three-dimensional imaging of the object, breaking the limitation of the complex optical structure of traditional imaging. The use of point detectors can greatly improve the imaging band and sensitivity, and expand the application scenarios of imaging. With the optimization of the correlation imaging structure, the computational correlation imaging system has been proposed. The light field modulated by the spatial light modulator is used to illuminate the target. The traditional correlation imaging system can be evolved into a computational correlation imaging system with only an object arm, so that a point detector can be used to collect the object arm information to realize the imaging of two-dimensional or three-dimensional objects. Computational correlation imaging significantly improves image quality and information acquisition capabilities by integrating advanced algorithms, optical path design and multiple imaging technologies, expands the scope of application, reduces costs, and brings more flexibility and innovation possibilities. By combining computational correlation imaging with compressed sensing theory, the number of samples is reduced, and the speed of reconstructing images is increased to achieve trajectory reconstruction of simulated moving cars. Using the time-correlated single photon counting (TCSPC) mode of the SPAD camera, the three-dimensional trajectory of the object is restored under non-line-of-sight conditions. However, the above correlation imaging technology is not applicable to the application of target trajectory reconstruction in extremely weak light environments. The existing correlation imaging scheme at the photon level uses computational correlation imaging and compressed sensing schemes to suppress the influence of shot noise and improve the signal-to-noise ratio of imaging, but there are still problems such as long data acquisition time, long algorithm optimization time and poor imaging quality caused by photon counting, so it is impossible to track high-speed moving targets and detect trajectory reconstruction.
[0004] In the past decade, correlation imaging, as a new imaging method, has broken through the limitation that the signal light must directly illuminate the detector in imaging based on the second-order correlation of the light field. The image of the object is reconstructed through two spatially separated light paths, a reference light path containing the object information and a detection light path that does not directly illuminate the object. By calculating the second-order correlation function of the two-path light intensity, objects that are obscured or not directly illuminated can be indirectly "seen", showing unique imaging capabilities and application potential. By utilizing the second-order correlation characteristics, correlation imaging has the advantages of strong anti-interference ability, super-resolution imaging and simple architecture. These advantages make correlation imaging a new imaging technology with unique charm and wide application value. Summary of the invention
[0005] In view of the problem that traditional correlation imaging schemes cannot restore the trajectory of moving targets under extremely weak light levels and that traditional imaging is easily interfered by noise under high background conditions, the present invention provides a high-speed moving target trajectory reconstruction system and method based on spatiotemporal correlation. In this method, the correlation imaging method is combined with the principle of quantum compressed sensing, and the compressed sensing relationship is constructed by utilizing the second-order correlation properties of time and space that exist simultaneously between pixels when the photons radiated by the measured object are detected by a single-photon array camera. By analyzing the spatiotemporal correlation characteristics between adjacent frames of single-photon array detection data and entering the spatiotemporal correlation domain, the three-dimensional motion characteristics of high-speed moving objects at any time are directly extracted, and the three-dimensional trajectory of the target is further restored. Since the background noise photons exhibit completely random characteristics in time and space, the noise correlation intensity in the spatiotemporal correlation process is much lower than that of the moving object, so it shows good anti-noise performance.
[0006] In order to achieve the above object, the present invention adopts the following technical solutions:
[0007] A high-speed moving target trajectory reconstruction system based on time-space correlation, the system comprising a signal source, an optical transmitting end and an optical collecting end;
[0008] The optical transmitting end comprises a pulse generating module, a current modulating module, a laser diode and a flat-top beam which are sequentially arranged on the laser output line, and the pulse generating module is connected to a signal source;
[0009] The pulse generating module receives a square wave signal as a trigger signal, and adjusts the amplitude and pulse width of the input electric pulse by adjusting the amplitude and pulse width knob of the module. The amplitude and pulse width of the electric pulse determine the pulse width of the final output optical pulse.
[0010] The current modulation module is used to power the laser and modulate the electrical pulse waveform output by the pulse generation module to the output current;
[0011] The laser diode receives the modulated current waveform from the current modulation module and modulates it onto the output laser, so that the laser diode outputs pulsed laser;
[0012] The flat-top beam shapes the intensity Gaussian distribution pulse laser output by the laser diode into a laser with non-Gaussian distribution and uniform intensity;
[0013] The optical collection end includes an optical lens module, a single photon array camera and a computer for sequentially detecting and recording data, and the single photon array camera and the computer are connected via a USB 3.0 data cable;
[0014] The optical lens module achieves the goal of collecting target signals at different distances by assembling lens groups with different parameters;
[0015] The single photon array camera is connected to the optical lens module via a standard camera bayonet and detects and records the light signal collected by the optical lens module;
[0016] The computer records the data of the single photon signal detected by the single photon array camera for subsequent processing and analysis by the time correlation method;
[0017] The signal source is used to generate two square wave signals with the same frequency, one of which is input into the pulse generating module to trigger the pulse generating module, and the other is input into the single photon array camera as a clock signal;
[0018] The pulse generation module is powered by a 24V constant voltage source, the input trigger signal is a TTL signal, and the amplitude and pulse width knobs are used to adjust the amplitude and pulse width of the electric pulse output. The narrowest electric pulse output is 500ps.
[0019] The current modulation module can withstand a maximum current of 110mA and a frequency response range of 10kHz-1GHz;
[0020] The maximum operating current of the laser diode is 60mA, and the maximum output power is 10mW;
[0021] The single-photon array camera has a time accuracy of 55ps, a detectable wavelength range of 400nm-1000nm, and a chip pixel size of 32×32;
[0022] The square wave signal output by the signal source to the pulse generation module and the single-photon array camera has the characteristics of consistent amplitude and frequency. In the experiment, it is necessary to ensure that the pulse generator and the single-photon array camera are consistent from beginning to end. However, due to the time deviation introduced in the circuit, the relative phase of the two outputs is adjusted by the signal source to eliminate it.
[0023] The computer uses the sampling time parameters set by the single-photon array camera and the time-space correlation imaging method to analyze the time-space correlation characteristics between the recorded data. Since the background noise follows the Poisson distribution in the time domain, there is no correlation characteristic in the time-space correlation domain. However, the continuously moving target has the time and space correlation characteristics in the field of view of the single-photon array camera, so that the motion trajectory can be extracted from the background, and the position and time information of the moving target can be obtained.
[0024] A high-speed moving target trajectory reconstruction method based on time-space correlation, the method comprising the following steps:
[0025] Step 1: The signal source generates a first square wave signal and sends it to the pulse generation module to trigger the pulse generation module; the signal source generates another square wave signal and sends it to the single-photon array camera to be used as the clock signal of the single-photon array camera; the two square wave signals are generated at the same time, ensuring the clock synchronization of the optical transmitting end and the optical collecting end.
[0026] Step 2: The pulse generation module generates an electric pulse after being triggered by the square wave signal generated by the signal source. The modulated current outputted by the electric pulse input into the current modulation module is used to trigger the laser diode to output a pulse light signal. The pulse laser signal with Gaussian intensity distribution is shaped into a laser spot with uniform intensity distribution after passing through the flat-top beam. The laser spot illuminates the scene where the moving target is moving.
[0027] Step 3: The optical lens module collects the scene of the moving target illuminated by the laser spot and images it onto the chip of the single-photon array camera. The chip of the single-photon array camera records the photon signal of the moving target in a continuous time. Finally, the single-photon array camera saves the data transmission to the computer via USB 3.0, and then uses the correlation imaging algorithm to process the data on the computer to obtain the correlation imaging result.
[0028] The specific steps of correlation imaging are as follows:
[0029] Step 3.1: Define the second-order spatiotemporal correlation function between two spatiotemporal pixel points p1 and p1 of the single-photon array camera in the correlation imaging field of view as:
[0030] G (2) (t,t+τ;D,D+d)
[0031] =G (2) (p1(t1,D1(x1,y1,z1)),p2(t2,D2(x2,y2,z2)))
[0032] =<I1(t1,D1(x1,y1,z1))·I2(t2,D2(x2,y2,z2))>
[0033] Among them, t2=t1+τ represents the temporal correlation between two pixels, τ is the fixed sampling interval set by the single-photon array camera; D1(x1,y1,z1)=D2(x2,y2,z2)+d(x,y,z) represents the spatial correlation between two pixels, d(x,y,z) is the spatial interval between two pixels; I1 represents the number of photons measured at a certain pixel within time; I2 represents the number of photons measured at another pixel within time;
[0034] Before the photons radiated by a moving object are detected by a single-photon array camera, they are regarded as a photon number state light field, and its density operator is written as: n= |n><n|, so the intensity value measured at a certain pixel of the single-photon array camera is equivalent to the number of photons measured within the sampling interval τ:
[0035]
[0036] I i (t i ,D i (x i ,y i ,z i )) represents the number of photons collected at time i; <n>represents the average number of photons; represents the photon number operator; represents taking the trace of the product of the density operator and the photon number operator; |n> represents the right eigenvector of the photon number state; <n| represents the left eigenvector of the photon number state;
[0037] Step 3.2: Obtain the second-order spatio-temporal correlation coherence function according to Step 3.1:
[0038]
[0039] When there is second-order spatio-temporal coherence between two pixel points, then g (2) (t, t + τ; D, D + d) = 1 and g (1) (t, t + τ; D, D + d) = 1 both hold;
[0040] G (1) (t, D)G (1) (t + τ, D + d) represents the product of the first-order correlation functions between two pixel points;
[0041] Step 3.3: Set the time for the single-photon array camera to acquire an image as τ; when photons radiated by a moving object are detected at the pixel position (i, j) in the m-th frame image, due to the short integration time of the single-photon array camera, in the (m + 1)-th frame, the photons radiated by the moving object are in the coordinate (i, j) and the eight pixels in the surrounding area; by performing the operation of correlation dot product on adjacent frame images, the background noise is suppressed, and at the same time, the trajectory of the object's movement at this moment is extracted;
[0042] Step 3.4: According to the trajectory obtained in Step 3.3, due to the influence of dark noise on the single-photon array camera, the spatio-temporal correlation imaging results between adjacent two frames will have a problem of non-coincidence with the actual trajectory. To solve the problem of the influence of randomly appearing dark noise on the pixel points on the spatio-temporal correlation imaging results, in actual experiments, the influence of random noise is usually averaged by using the correlation results of consecutive multiple frames;
[0043] Step 3.5: Through continuous spatio-temporal correlation operations between data frames, the trajectory of the object's movement is obtained from the single-photon quantum data; the scene is illuminated by pulsed laser output by a laser diode, the pulse generation module and the single-photon array camera are clock-synchronized, and using the time-of-flight method, the distance information of the imaging object is obtained from the TCSPC data of the single-photon array camera.
[0044] Compared with the prior art, the present invention has the following advantages:
[0045] 1. The present invention uses a single-photon array camera with time resolution capability. The single-photon array camera is a 32×32 array composed of 1024 single-photon sensitive SPAD pixels. This parallel design can significantly improve the efficiency and accuracy of data acquisition; and has an ultra-fast 55ps time resolution, which ensures the accuracy of the single-photon array camera in measuring high-speed moving targets. The frame rate of the single-photon array camera is as high as 150kfps, which can quickly capture and process a large number of photon signals and is suitable for application scenarios that require high-speed imaging. This shows that the present invention has extremely wide application value in the field of high-speed moving target trajectory recovery.
[0046] 2. The present invention proposes a spatiotemporal correlation imaging method, which enters the spatiotemporal correlation domain by analyzing the spatiotemporal correlation characteristics between adjacent frames of single-photon array detection data, directly extracts the three-dimensional motion characteristics of high-speed moving objects at any time, and further restores the three-dimensional trajectory of the target. Since background noise photons show completely random characteristics in time and space, the correlation intensity of noise in the spatiotemporal correlation process is much lower than that of moving objects, so it shows good anti-noise performance.
[0047] 3. The present invention uses a combination of a pulse generation module and a current modulation module to drive a laser diode. The pulse generation module generates an extremely narrow electrical pulse, which is then used to light up the laser diode through the current modulation module, thereby generating a pulse laser with a high extinction ratio. The pulse width of the pulse laser can reach hundreds of ps. This improves the extremely high distance measurement accuracy for the three-dimensional trajectory recovery of the moving target and ensures the accuracy of the measured and restored trajectory.
[0048] 4. In the field of correlation imaging and motion target trajectory recovery, the present invention has the ability to measure high-speed targets and resist noise, and plays an important role in recovering the real motion target trajectory in practical applications. BRIEF DESCRIPTION OF THE DRAWINGS
[0049] Figure 1 The system connection block diagram of the present invention; wherein: 101-signal source; 102-optical transmitting end; 1021-pulse generating module; 1022-current modulation module; 1023-laser diode; 1024-flat top beam; 103-optical collecting end; 1031-optical lens module; 1032-single photon array camera; 1033-computer;
[0050] Figure 2 A schematic diagram of a scene including a moving target for using the present invention, the scene including high-intensity background noise and a moving target and associated imaging results, demonstrating the ability to resist background noise in this aspect;
[0051] Figure 3 To obtain the Figure 2 Schematic diagram of the three-dimensional trajectory of the moving target recovered from the background noise. DETAILED DESCRIPTION
[0052] In order to have a deeper understanding of the present invention, we will describe it in a comprehensive and detailed manner. However, the present invention has multiple implementations and is not limited to the specific examples listed in the present invention. The presentation of these examples is intended to deepen the comprehensive understanding of the disclosure of the present invention.
[0053] A high-speed moving target trajectory reconstruction system based on time-space correlation, the system includes a signal source 101, an optical transmitting end 102 and an optical collecting end 103;
[0054] The optical transmitting end 102 includes a pulse generating module 1021, a current modulating module 1022, a laser diode 1023 and a flat-top beam 1024 which are sequentially arranged on the laser output line, and the pulse generating module 1021 is connected to the signal source 101;
[0055] The pulse generating module 1021 receives a square wave signal as a trigger signal, and adjusts the amplitude and pulse width of the input electric pulse by adjusting the amplitude and pulse width knob of the module. The amplitude and pulse width of the electric pulse determine the pulse width of the final output optical pulse.
[0056] The current modulation module 1022 is used to power the laser and modulate the electrical pulse waveform output by the pulse generation module 1021 to the output current;
[0057] The laser diode 1023 receives the modulated current waveform from the current modulation module 1022 and modulates it onto the output laser, so that the laser diode 1023 outputs a pulsed laser;
[0058] The flat-top beam 1024 shapes the intensity Gaussian distribution pulse laser output by the laser diode 1023 into a laser with non-Gaussian distribution and uniform intensity;
[0059] The optical collection end 103 includes an optical lens module 1031, a single photon array camera 1032 and a computer 1033 for sequentially detecting and recording data, and the single photon array camera 1032 and the computer 1033 are connected via a USB 3.0 data cable;
[0060] The optical lens module 1031 achieves the goal of collecting target signals at different distances by assembling lens groups with different parameters;
[0061] The single photon array camera 1032 is connected to the optical lens module through a standard camera bayonet and detects and records the light signal collected by the optical lens module 1031;
[0062] The computer 1033 records the data of the single photon signal detected by the single photon array camera 1032 for subsequent processing and analysis by the time correlation method;
[0063] The signal source 101 is used to generate two square wave signals with the same frequency, one of which is input to the pulse generating module 1021 to trigger the pulse generating module 1021, and the other is input to the single photon array camera 1032 as a clock signal.
[0064] The pulse generating module 1021 is powered by a 24V constant voltage source, the input trigger signal is a TTL signal, and the amplitude and pulse width knobs are used to adjust the amplitude and pulse width of the electric pulse output. The narrowest electric pulse output is 500ps.
[0065] The current modulation module 1022 can withstand a maximum current of 110 mA and a frequency response range of 10 kHz-1 GHz;
[0066] The maximum operating current of the laser diode 1023 is 60mA, and the maximum output power is 10mW;
[0067] The single photon array camera 1032 has a time accuracy of 55ps, a detectable wavelength range of 400nm-1000nm, and a chip pixel size of 32×32;
[0068] The square wave signal output by the signal source 101 to the pulse generating module 1021 and the single photon array camera 1032 has the characteristics of consistent amplitude and frequency. In the experiment, it is necessary to ensure that the pulse generator and the single photon array camera are consistent from beginning to end. However, due to the time deviation introduced in the circuit, the relative phase of the two outputs is adjusted by the signal source to eliminate it.
[0069] The computer 1033 uses the sampling time parameters set by the single photon array camera 1032 and the time-space correlation imaging method to analyze the time-space correlation characteristics between the recorded data. Since the background noise follows the Poisson distribution in the time domain, there is no correlation characteristic in the time-space correlation domain. However, the continuously moving target has the time and space correlation characteristics in the field of view of the single photon array camera, so that the motion trajectory can be extracted from the background, and the position and time information of the moving target can be obtained.
[0070] A high-speed moving target trajectory reconstruction method based on time-space correlation, the method comprising the following steps:
[0071] Step 1: The signal source 101 generates a first square wave signal and sends it to the pulse generating module 1021 to trigger the pulse generating module 1021; the signal source 101 generates another square wave signal and sends it to the single photon array camera 1032 to be used as the clock signal of the single photon array camera 1032; the two square wave signals are generated simultaneously, thereby ensuring the clock synchronization of the optical transmitting end 102 and the optical collecting end 103.
[0072] Step 2: The pulse generating module 1021 generates an electric pulse after being triggered by the square wave signal generated by the signal source 101. The modulated current outputted from the electric pulse input to the current modulating module 1022 is used to trigger the laser diode 1023 to output a pulse light signal. The pulse laser signal with Gaussian intensity distribution is shaped into a laser spot with uniform intensity distribution after passing through the flat-top beam 1024. The laser spot illuminates the scene where the moving target is moving.
[0073] Step 3: The optical lens module 1031 collects the scene of the moving target illuminated by the laser spot, and images it onto the chip of the single-photon array camera 1032. The chip of the single-photon array camera 1032 records the photon signal of the moving target in a continuous time; finally, the single-photon array camera 1032 saves the data transmission to the computer 1033 via USB 3.0, and then processes the data using the correlation imaging algorithm on the computer 1033 to obtain the correlation imaging result;
[0074] The specific steps of correlation imaging are as follows:
[0075] Step 3.1: Define the second-order spatiotemporal correlation function between two spatiotemporal pixel points p1 and p1 of the single-photon array camera 1032 in the correlation imaging field of view as follows:
[0076] G (2) (t,t+τ;D,D+d)
[0077] =G (2) (p1(t1,D1(x1,y1,z1)),p2(t2,D2(x2,y2,z2)))
[0078] =<I1(t1,D1(x1,y1,z1))·I2(t2,D2(x2,y2,z2))>
[0079] Wherein, t2=t1+τ represents the temporal correlation between two pixels, τ is the fixed sampling interval set by the single photon array camera 1032; D1(x1,y1,z1)=D2(x2,y2,z2)+d(x,y,z) represents the spatial correlation between two pixels, d(x,y,z) is the spatial interval between two pixels; I1 represents the number of photons measured at a certain pixel within a time; I2 represents the number of photons measured at another pixel within a time;
[0080] Before the photons radiated by the moving object are detected by the single-photon array camera 1032, they are regarded as a photon-number state light field, and its density operator is written as: ρ n = |n><n|. Therefore, the intensity value measured at a certain pixel of the single-photon array camera 1032 is equivalent to the number of photons measured within the sampling interval τ:
[0081]
[0082] I i (t i , D i (x i , y i , z i )) represents the number of photons collected at time i; <n>represents the average number of photons; represents the photon number operator; represents taking the trace of the product of the density operator and the photon number operator; |n> represents the right eigenvector of the photon number state; <n| represents the left eigenvector of the photon number state;
[0083] Step 3.2: Obtain the second-order spatio-temporal correlation coherence function according to Step 3.1:
[0084]
[0085] When there is second-order spatio-temporal coherence between two pixel points, then g (2) (t, t + τ; D, D + d) = 1 and g (1) (t, t + τ; D, D + d) = 1 both hold;
[0086] G (1) (t, D)G (1) (t + τ, D + d) represents the product of the first-order correlation functions between two pixel points;
[0087] Step 3.3: Set the time for the single-photon array camera 1032 to acquire an image as τ; When photons radiated by a moving object are detected at the pixel position (i, j) in the m-th frame image, due to the short integration time of the single-photon array camera 1032, in the (m + 1)-th frame, the photons radiated by the moving object are in the coordinate (i, j) and the eight pixels in the surrounding area; By performing the operation of correlation dot product on adjacent frame images, the background noise can be suppressed, and at the same time, the trajectory of the object's movement at this moment can be extracted;
[0088] Step 3.4: According to the trajectory obtained in Step 3.3, due to the influence of dark noise on the single-photon array camera 1032, the spatio-temporal correlation imaging results between adjacent two frames will have a problem of not coinciding with the actual trajectory. To solve the problem of the influence of randomly appearing dark noise in pixel points on the spatio-temporal correlation imaging results, in actual experiments, the influence of random noise is usually averaged by using the correlation results of consecutive multiple frames;
[0089] Step 3.5: Through continuous spatio-temporal correlation operations between data frames, the trajectory of the object's movement is obtained from the single-photon quantum data; The scene is illuminated by the pulsed laser output by the laser diode 1023, the pulse generation module 1021 and the single-photon array camera 1032 are clock-synchronized, and the distance information of the imaging object is obtained from the TCSPC data of the single-photon array camera by using the time-of-flight method.
[0090] As Figure 2 shown, it shows a schematic diagram of a scene including a moving target in the present invention. This scene includes high-intensity background noise, a moving target, and the correlation imaging results, demonstrating the anti-background noise ability in this aspect.
[0091] Figure 3 To obtain the Figure 2 The three-dimensional trajectory of the moving target is recovered from the background noise.
[0092] The present invention proposes a high-speed moving target trajectory reconstruction system and method based on spatiotemporal correlation; it realizes the direct extraction of sparse trajectories with spatiotemporal correlation characteristics in the spatiotemporal correlation domain; the method embodies good anti-noise performance by extracting the characteristics of the spatiotemporal correlation characteristics of the target. This imaging technology can be used as a new type of high-speed imaging technology in extremely weak light environments, which is of great significance for promoting the practical application of single-photon imaging.
[0093] In summary, the image results verify the correctness, effectiveness and reliability of the present invention.
[0094] The contents not described in detail in the specification of the present invention belong to the prior art known to the professional and technical personnel in the field. Although the illustrative specific embodiments of the present invention are described above to facilitate the understanding of the present invention by the technical personnel in the field, it should be clear that the present invention is not limited to the scope of the specific embodiments. For the ordinary technical personnel in the field, as long as various changes are within the spirit and scope of the present invention defined and determined by the attached claims, these changes are obvious, and all inventions and creations using the concept of the present invention are protected.< / n> < / n>
Claims
1. A high-speed moving target trajectory reconstruction system based on time-space correlation, characterized in that: The system comprises a signal source (101), an optical transmitting end (102) and an optical collecting end (103); The optical emission end (102) comprises a pulse generation module (1021), a current modulation module (1022), a laser diode (1023) and a flat-top beam (1024) which are sequentially arranged on the laser emission line, and the pulse generation module (1021) is connected to the signal source (101); The pulse generating module (1021) receives a square wave signal as a trigger signal, and adjusts the amplitude and pulse width of the input electric pulse by adjusting the amplitude and pulse width knobs of the module. The amplitude and pulse width of the electric pulse determine the pulse width of the final output light pulse. The current modulation module (1022) is used to power the laser and modulate the electrical pulse waveform output by the pulse generation module (1021) onto the output current; The laser diode (1023) receives the modulated current waveform from the current modulation module (1022) and modulates the waveform onto the output laser, so that the laser diode (1023) outputs pulsed laser; The flat-top beam (1024) shapes the intensity Gaussian distribution pulse laser output by the laser diode (1023) into a laser with non-Gaussian distribution and uniform intensity; The optical collection end (103) comprises an optical lens module (1031) for sequentially detecting and recording data, a single photon array camera (1032) and a computer (1033), and the single photon array camera (1032) and the computer (1033) are connected via a USB 3.0 data cable; The optical lens module (1031) achieves the goal of collecting target signals at different distances by assembling lens groups with different parameters; The single photon array camera (1032) is connected to the optical lens module via a standard camera bayonet and detects and records the light signal collected by the optical lens module (1031); The computer (1033) records the data of the single photon signal detected by the single photon array camera (1032) for subsequent processing and analysis by a time correlation method; The signal source (101) is used to generate two square wave signals with consistent frequencies, wherein one square wave signal is input into a pulse generating module (1021) to trigger the pulse generating module (1021), and the other square wave signal is input into a single photon array camera (1032) as a clock signal.
2. A high-speed moving target trajectory reconstruction system based on time-space correlation according to claim 1, characterized in that: The pulse generation module (1021) is powered by a 24V constant voltage source, the input trigger signal is a TTL signal, and the amplitude and pulse width knobs are used to adjust the amplitude and pulse width of the electric pulse output, and the narrowest electric pulse output is 500ps; The current modulation module (1022) can withstand a maximum current of 110 mA and a frequency response range of 10 kHz-1 GHz; The maximum operating current of the laser diode (1023) is 60mA, and the maximum output power is 10mW; The single photon array camera (1032) has a time accuracy of 55 ps, a detectable wavelength range of 400 nm-1000 nm, and a chip pixel size of 32×32; The square wave signal output by the signal source (101) to the pulse generation module (1021) and the single photon array camera (1032) has the characteristics of consistent amplitude and consistent frequency; The computer (1033) uses the sampling time parameters set by the single photon array camera (1032) and the time-space correlation imaging method to analyze the time-space correlation characteristics between the recorded data.
3. A high-speed moving target trajectory reconstruction method based on spatiotemporal correlation, characterized in that: The method comprises the following steps: Step 1: The signal source (101) generates a first square wave signal and sends it to the pulse generating module (1021) to trigger the pulse generating module (1021); the signal source (101) generates another square wave signal and sends it to the single photon array camera (1032) to be used as the clock signal of the single photon array camera (1032); the two square wave signals are generated simultaneously, thereby ensuring the clock synchronization of the optical transmitting end (102) and the optical collecting end (103); Step 2: The pulse generating module (1021) generates an electric pulse after being triggered by the square wave signal generated by the signal source (101), and the modulated current outputted after the electric pulse is inputted into the current modulating module (1022) is used to trigger the laser diode (1023) to output a pulsed light signal; the pulsed laser signal with Gaussian intensity distribution is shaped into a laser spot with uniform intensity distribution after passing through the flat-top beam (1024); the laser spot illuminates the scene where the moving target is moving; Step 3: The optical lens module (1031) collects the scene of the moving target illuminated by the laser spot and images it onto the chip of the single-photon array camera (1032). The chip of the single-photon array camera (1032) records the photon signal of the moving target in a continuous time. Finally, the single-photon array camera (1032) transmits and saves the data to the computer (1033) via USB 3.0, and then processes the data using the correlation imaging algorithm on the computer (1033) to obtain the correlation imaging result.
4. The method for reconstructing high-speed moving target trajectory based on spatiotemporal correlation according to claim 3, characterized in that: In step 3, the optical lens module (1031) collects the scene of the moving target illuminated by the laser spot, and images it onto the chip of the single-photon array camera (1032), and the chip of the single-photon array camera (1032) records the photon signal of the moving target in a continuous time; finally, the single-photon array camera (1032) transmits and saves the data to the computer (1033) via USB 3.0, and then processes the data using the correlation imaging algorithm on the computer (1033) to obtain the correlation imaging result, wherein the specific steps of the correlation imaging are: Step 3.1: Define the second-order spatiotemporal correlation function between two spatiotemporal pixel points p1 and p1 of the single-photon array camera (1032) in the imaging field of view as: G (2) (t,t+τ;D,D+d) <h2 style=";text-align:left;direction:ltr">=G<h2 style=";text-align:left;direction:ltr"> (2) <h2 style=";text-align:left;direction:ltr"> (p1(t1,D1(x1,y1,z1)),p2(t2,D2(x2,y2,z2))) =<I1(t1,D1(x1,y1,z1))·I2(t2,D2(x2,y2,z2))> Wherein, t2=t1+τ represents the temporal correlation between two pixels, τ is a fixed sampling interval set by the single photon array camera (1032); D1(x1, y1, z1)=D2(x2, y2, z2)+d(x, y, z) represents the spatial correlation between two pixels, d(x, y, z) is the spatial interval between two pixels; I1 represents the number of photons measured at a certain pixel within a time; I2 represents the number of photons measured at another pixel within a time; Before the photons radiated by a moving object are detected by a single-photon array camera (1032), they are regarded as a photon number state optical field, and its density operator is written as: ρ n = |n><n|. Therefore, the intensity value measured at a certain pixel of the single-photon array camera (1032) is equivalent to the number of photons measured within the sampling interval τ: I i (t i ,D i (x i ,y i ,z i )) represents the number of photons collected at time i; <n>represents the average number of photons; represents the photon number operator; represents the trace of the product of the density operator and the photon number operator; |n> represents the right eigenvector of the photon number state; <n| represents the left eigenvector of the photon number state;< / n> Step 3.2: According to step 3.1, the second-order space-time correlation function is obtained: When there is a second-order spatiotemporal coherence between two pixels, then g (2) (t,t+τ;D,D+d)=1 and g (1) (t, t+τ; D, D+d) = 1; G (1) (t,D)G (1) (t+τ,D+d) represents the product of the first-order correlation function between two pixels; Step 3.3: The time for the single-photon array camera (1032) to collect an image is set to τ; when the photons radiated by the moving object are detected at the pixel position (i, j) in the m-th frame image, in the m+1-th frame, the photons radiated by the moving object are located at the coordinates (i, j) and eight pixels in a circle around it; by performing a correlation dot product operation on the images of adjacent frames, the background noise is suppressed, and the trajectory of the object's movement at that moment is extracted; Step 3.4: Based on the trajectory obtained in step 3.3, the influence of random noise is averaged using the correlation results of multiple consecutive frames; Step 3.5: The trajectory of the object's motion is obtained from the single-photon quantum data by performing continuous spatiotemporal correlation operations between data frames; the pulsed laser output by the laser diode (1023) is used to illuminate the scene, the pulse generation module (1021) and the single-photon array camera (1032) are clock-synchronized, and the distance information of the imaging object is obtained from the TCSPC data of the single-photon array camera using the time-of-flight method.
Citation Information
Cited By
X-ray diode signal reconstruction method
CN121098288A
X-ray diode signal reconstruction method
CN121098288B
Photoelectric conversion and signal processing method and system of eddy current aircraft detection device
CN121829575A
