Low-noise spatial heterodyne hyper-spectral detection simulation and evaluation method

Abstract

The invention discloses a low-noise spatial heterodyne hyper-spectral detection simulation and evaluation method. The method comprises the following steps: simulating a satellite platform to obtain a satellite platform model; simulating a camera to obtain a camera model; setting a camera model on the satellite platform model; incident light passes through the camera model and then an interference image is output; performing spectrum restoration processing on the interference image to obtain a restored spectrum signal, and obtaining a signal-to-noise ratio of the image according to the restored spectrum signal, electronic system noise in the camera model and the number N of pixels merged by different limb heights in the satellite platform model; and obtaining a root-mean-square error RMSE and an average spectral similarity MSS according to the incident light and the recovered spectral signal. According to the method, the problem of near space detection data set missing can be solved, a high-reliability simulation data set is formed, technical support is provided for comprehensive system analysis and optimization of a near space atmospheric parameter hyper-spectrum detection scheme, and quantitative guarantee is provided for comprehensive verification and improvement of application capability.

Description

Technical Field

[0001] The invention belongs to the technical field of optical remote sensing quantification, and in particular relates to a low-noise spatial heterodyne hyperspectral detection simulation and evaluation method. Background Art

[0002] The detection of greenhouse gas concentrations and multiple factors such as temperature, humidity, and pressure in the middle and upper atmosphere is of great significance to the study of the global climate change process. Spatial heterodyne spectroscopy technology uses diffraction gratings to replace the plane mirrors of the Michelson interferometer. It has the characteristics of ultra-high spectral resolution, high light flux, and high stability, and can achieve vertical high-resolution observation of narrow-band spectra. The spatial heterodyne observation system of horizontal wide-band uniform modulation and profile high-resolution simultaneous imaging further realizes the hyperspectral, high signal-to-noise ratio, and high-temporal and spatial resolution layered observation of atmospheric elements. Compared with the previous observation system, the new observation system improves the spatial resolution, and the corresponding pixel receives lower radiation energy, which reduces the system signal-to-noise ratio under the same noise level. Signal-to-noise ratio is one of the core indicators of spectrometer design, and it is also a key indicator for evaluating the accuracy of spectral restoration and parameter inversion capabilities. Therefore, it is necessary to propose a spatial heterodyne hyperspectral detection simulation and data quality assessment method that meets the low noise requirements, so as to support system design and performance prediction, and facilitate the subsequent uncertainty transfer process analysis and error allocation.

[0003] At present, simulations are mainly carried out on principle prototypes built in the laboratory, using ZEMAX optical software to simulate interference signals to optimize optical system design parameters, or using Matlab for numerical simulation, but the numerical model lacks modeling of low-noise electronic systems. Data quality assessment mainly involves qualitative comparison of the morphological similarity between the restored spectrum and the input spectrum, and quantitative analysis of the residuals of the two, which cannot achieve a dynamic association between evaluation indicators and design parameters. Summary of the invention

[0004] The technical problem solved by the present invention is: to overcome the shortcomings of the prior art and provide a low-noise spatial heterodyne hyperspectral detection simulation and evaluation method, which can solve the problem of missing near-space detection data sets and form a highly reliable simulation data set, providing technical support for the comprehensive systematic analysis and optimization of near-space atmospheric parameter hyperspectral detection solutions, and also providing quantitative guarantees for the comprehensive verification and improvement of application capabilities.

[0005] The object of the present invention is achieved through the following technical scheme: a low-noise spatial heterodyne hyperspectral detection simulation and evaluation method, comprising: simulating a satellite platform to obtain a satellite platform model; simulating a camera to obtain a camera model; setting the camera model on the satellite platform model; outputting an interference image after the incident light passes through the camera model; spectrally restoring the interference image to obtain a restored spectral signal, and obtaining the signal-to-noise ratio of the image according to the restored spectral signal, the electronic system noise in the camera model and the number of pixels N merged at different edge heights in the satellite platform model; and obtaining a root mean square error RMSE and a mean spectral similarity MSS according to the incident light and the restored spectral signal.

[0006] In the above-mentioned low-noise spatial heterodyne hyperspectral detection simulation and evaluation method, the camera model obtained by simulating the camera includes: simulating the front optical system to obtain the front optical system model; simulating the interferometer system to obtain the interferometer system model; simulating the rear optical system to obtain the rear optical system model; simulating the electronics system to obtain the electronics system model.

[0007] In the above-mentioned low-noise spatial heterodyne hyperspectral detection simulation and evaluation method, the incident light passes through the front optical system model to output a collimated light signal, and the collimated light signal passes through the interferometer system model to output an interference pattern; the interference pattern is converged to the electronic system model through the rear optical system model, and the electronic system model samples the converged interference pattern to obtain an interference image.

[0008] In the above low-noise spatial heterodyne hyperspectral detection simulation and evaluation method, the satellite platform model is obtained by the following formula:

[0009]

[0010] Among them, S is the distance between the satellite and the limb observation point, R is the equatorial radius, H is the satellite orbit height, h is the height of the optical axis tangent point, W is the instrument imaging width, and W v is the vertical component of the instrument imaging width, FOV is the instrument field of view, FOV v is the vertical component of the instrument’s field of view, FOV H is the horizontal component of the instrument's field of view, N_pixel v is the number of pixels of effective imaging in the vertical direction of the array detector, N_pixel is the number of pixels of effective imaging in the horizontal direction of the array detector, K is the profile resolution of the instrument, and N is the number of pixels merged in the vertical direction.

[0011] In the above low-noise spatial heterodyne hyperspectral detection simulation and evaluation method, the signal-to-noise ratio of the image is obtained by the following formula:

[0012]

[0013] Among them, SNR is the signal-to-noise ratio of the images corresponding to the ultraviolet, visible, near-infrared and short-wave infrared spectral bands, NETD is the signal-to-noise ratio of the images corresponding to the medium- and long-wave infrared spectral bands, L is the restored spectral signal, σ is the electronic system noise, N is the number of pixels merged at different limb heights, T0 is the background temperature, and T is the equivalent blackbody temperature of the target at different heights.

[0014] In the above low-noise spatial heterodyne hyperspectral detection simulation and evaluation method, the electronic system noise σ is obtained by the following formula:

[0015]

[0016] Where σ is the electronic system noise, σ shot is the photon noise, σ dark is the dark current, σ quant is the quantization noise, σ read is the preamplifier read noise, σ bit is the bit error rate, A pixel is the detector pixel area, Tint is the integration time, h is Planck's constant, c is the speed of light, σ mean is the average wavelength in the spectral range, QE is the quantum efficiency, e cal is the calibration error rate, I m (x m ) is the interference image, x m is the wave number position of the interference image corresponding to the detector pixel, A a is the entrance pupil area, D is the detection rate of the device, Sd is the pixel size, F is the detector frame rate, ψ1 and ψ2 are the transmission rates of the front optical system model and the rear optical system model respectively, e satur is the number of saturated electrons, M is the number of quantization bits, W is the grating length, θ L is the Littrow angle of interference, R is the detector response, NPSD is the preamplifier noise power spectral density, bit error is the bit error rate, SR is the spectral resolution, σ m is the wave number of the grating interference position.

[0017] In the above low-noise spatial heterodyne hyperspectral detection simulation and evaluation method, the root mean square error RMSE is obtained by the following formula:

[0018]

[0019] Where L is the restored spectral radiance signal, L input is the incident light signal, and n is the number of sampling points in the spectral range.

[0020] In the above low-noise spatial heterodyne hyperspectral detection simulation and evaluation method, the average spectral similarity MSS is obtained by the following formula:

[0021]

[0022] Where L is the restored spectral radiance signal, L input is the incident light signal, is the average value of the incident light signal, corr is the Pearson correlation coefficient, RMSE is the root mean square error, and mean() is the average value of the spectral dimension signal.

[0023] In the above low-noise spatial heterodyne hyperspectral detection simulation and evaluation method, the electronics system model is obtained by the following formula:

[0024] I m (x m )=R(σ,x)×I(N_pixel)+I σ ;

[0025] I(N_pixel)=f(I′(x),N_pixel);

[0026]

[0027] Among them, I m (x m ) is the interference image, I(N_pixel) is the interference signal after being sampled by the detector, N_pixel is the number of pixels in the horizontal direction of the detector, f() is the one-dimensional linear interpolation function, R(σ,x) is the spectral response function of the detector, x m is the wave number position of the interference signal corresponding to the detector pixel, FWHM is the spectral bandwidth, Sd is the pixel size, I σ is the total noise intensity of the detector, σ is the noise of the electronic system, A pixel is the detector pixel area, Tint is the integration time, h is Planck's constant, c is the speed of light, σ m is the wave number of the grating interference position, QE is the quantum efficiency, x is the grating interference position, and I′(x) is the output interference intensity value of the post-optical system.

[0028] A low-noise spatial heterodyne hyperspectral detection simulation and evaluation system comprises: a first module for simulating a satellite platform to obtain a satellite platform model; a second module for simulating a camera to obtain a camera model; the camera model is set on the satellite platform model; a third module is used to output an interference image after incident light passes through the camera model; a fourth module is used to perform spectral restoration processing on the interference image to obtain a restored spectral signal, and obtain the signal-to-noise ratio of the image according to the restored spectral signal, the electronic system noise in the camera model and the number of pixels N combined with different limb heights in the satellite platform model; and a fifth module is used to obtain a root mean square error RMSE and a mean spectral similarity MSS according to the incident light and the restored spectral signal.

[0029] Compared with the prior art, the present invention has the following beneficial effects:

[0030] (1) The present invention can solve the problem of missing near-space detection data sets, form a highly reliable simulation data set, provide technical support for the comprehensive system analysis and optimization of near-space atmospheric parameter hyperspectral detection schemes, and provide quantitative guarantee for comprehensive verification and improvement of application capabilities;

[0031] (2) The low-noise spatial heterodyne hyperspectral detection simulation method proposed in the present invention starts with the analysis of the detector noise characteristics, analyzes and models the detector output response and the preamplifier output noise, and realizes the correlation between the noise and the design indicators such as the integration time, the entrance pupil diameter, and the spectral resolution, which is helpful for the subsequent quantitative analysis of the system performance;

[0032] (3) The present invention takes into account the impact of the low-noise detection system on the signal-to-noise ratio, establishes a system performance characterization and evaluation method based on the core indicator of the signal-to-noise ratio, and realizes the identification of the detection data quality at different edge heights. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0034] Figure 1 It is a flow chart of a low-noise spatial heterodyne hyperspectral detection simulation and evaluation method provided by an embodiment of the present invention;

[0035] FIG. 2( a ) is a schematic diagram of a low-noise spatial heterodyne hyperspectral detection simulation visible light interference signal provided by an embodiment of the present invention;

[0036] FIG. 2( b ) is a schematic diagram of imaging at different profile heights of a detector provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0037] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to be able to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features described in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0038] Figure 1 1 is a flow chart of a low-noise spatial heterodyne hyperspectral detection simulation and evaluation method provided by an embodiment of the present invention. Figure 1 As shown, the low-noise spatial heterodyne hyperspectral detection simulation and evaluation method includes: simulating a satellite platform to obtain a satellite platform model; simulating a camera to obtain a camera model; setting the camera model on the satellite platform model; outputting an interference image after the incident light passes through the camera model; spectrally restoring the interference image to obtain a restored spectral signal, and obtaining the signal-to-noise ratio of the image according to the restored spectral signal, the electronic system noise in the camera model, and the number of pixels N combined at different edge heights in the satellite platform model; and obtaining the root mean square error RMSE and the average spectral similarity MSS according to the incident light and the restored spectral signal.

[0039] The camera model obtained by simulating the camera includes: simulating the front optical system to obtain the front optical system model; simulating the interferometer system to obtain the interferometer system model; simulating the rear optical system to obtain the rear optical system model; simulating the electronics system to obtain the electronics system model.

[0040] The incident light passes through the front optical system model and outputs a collimated light signal, and the collimated light signal passes through the interferometer system model and outputs an interference pattern. The interference pattern is converged to the electronics system model through the rear optical system model, and the electronics system model samples the converged interference pattern to obtain an interference image.

[0041] The satellite platform model is obtained by the following formula:

[0042]

[0043] Among them, S is the distance between the satellite and the limb observation point, R is the equatorial radius, H is the satellite orbit height, h is the height of the optical axis tangent point, W is the instrument imaging width, and W v is the vertical component of the instrument imaging width, FOV is the instrument field of view, FOV v is the vertical component of the instrument’s field of view, FOV H is the horizontal component of the instrument's field of view, N_pixel v is the number of pixels of effective imaging in the vertical direction of the array detector, N_pixel is the number of pixels of effective imaging in the horizontal direction of the array detector, K is the profile resolution of the instrument, and N is the number of pixels combined in the vertical direction.

[0044] The signal-to-noise ratio of the image is obtained by the following formula:

[0045]

[0046] Among them, SNR is the signal-to-noise ratio of the images corresponding to the ultraviolet, visible, near-infrared and short-wave infrared spectral bands, NETD is the signal-to-noise ratio of the images corresponding to the medium- and long-wave infrared spectral bands, L is the restored spectral signal, σ is the electronic system noise, N is the number of pixels merged at different limb heights, T0 is the background temperature, and T is the equivalent blackbody temperature of the target at different heights.

[0047] The electronic system noise σ is obtained by the following formula:

[0048]

[0049]

[0050] Where, σ is the electronic system noise, σ shot is the photon noise, σ dark is the dark current, σ quant is the quantization noise, σ read is the preamplifier read noise, σ bit is the bit error rate, A pixel is the detector pixel area, Tint is the integration time, h is Planck's constant, c is the speed of light, σ mean is the average wavelength in the spectral range, QE is the quantum efficiency, e cal is the calibration error rate, I m (x m ) is the interference image, x m is the wave number position of the interference image corresponding to the detector pixel, A a is the entrance pupil area, D is the detection rate of the device, Sd is the pixel size, F is the detector frame rate, ψ1 and ψ2 are the transmission rates of the front optical system model and the rear optical system model respectively, e satur is the number of saturated electrons, M is the number of quantization bits, W is the grating length, θ L is the Littrow angle of interference, R is the detector response, NPSD is the preamplifier noise power spectral density, bit error is the bit error rate, SR is the spectral resolution, σ m is the wave number of the grating interference position.

[0051] The root mean square error RMSE is obtained by the following formula:

[0052]

[0053] Where L is the restored spectral radiance signal, L input is the incident light signal, and n is the number of sampling points in the spectral range.

[0054] The average spectral similarity MSS is obtained by the following formula:

[0055]

[0056] Where L is the restored spectral radiance signal, L input is the incident light signal, is the average value of the incident light signal, corr is the Pearson correlation coefficient, RMSE is the root mean square error, and mean() is the average value of the spectral dimension signal.

[0057] The electronics system model is obtained by the following formula:

[0058] I m (x m )=R(σ,x)×I(N_pixel)+I σ ;

[0059] I(N_pixel)=f(I′(x),N_pixel);

[0060]

[0061] Among them, I m (x m ) is the interference image, I(N_pixel) is the interference signal after being sampled by the detector, N_pixel is the number of pixels in the horizontal direction of the detector, f() is the one-dimensional linear interpolation function, R(σ,x) is the spectral response function of the detector, x m is the wave number position of the interference signal corresponding to the detector pixel, FWHM is the spectral bandwidth, Sd is the pixel size, I σ is the total noise intensity of the detector, σ is the noise of the electronic system, A pixel is the detector pixel area, Tint is the integration time, h is Planck's constant, c is the speed of light, σ m is the wave number of the grating interference position, QE is the quantum efficiency, x is the grating interference position, and I′(x) is the output interference intensity value of the post-optical system.

[0062] Specifically, the method comprises the following steps:

[0063] Step 1: Satellite platform simulation

[0064] The satellite platform's edge observation geometry determines the size of the instantaneous observation field of view, which in turn affects the number of effective pixels in the instrument's internal optical system. The number of effective signal pixels in the detector for vertical (spatial) field-of-view imaging is determined by the distance between the satellite and the observation point, the vertical imaging width, and the design redundancy.

[0065] The satellite platform modeling mainly considers the impact of orbital altitude and the tangent point height of the detection optical axis on the distance between the satellite and the observation position, and finally simulates the size of the detector's effective observation area and the number of pixel merging during satellite observation.

[0066]

[0067] Where S is the distance between the satellite and the limb observation point, R is the equatorial radius, H is the satellite orbit height, h is the height of the optical axis tangent point, W is the instrument imaging width (the subscript v represents the vertical direction, and H represents the horizontal direction), FOV is the instrument field of view (the subscript indicates the same width), N_pixel v is the number of pixels of effective imaging in the vertical direction of the array detector, N_pixel is the number of pixels of effective imaging in the horizontal direction of the array detector (usually the total number of pixels in the horizontal direction of the detector), K is the profile resolution of the instrument, and N is the number of pixels combined in the vertical direction.

[0068] Step 2: Pre-optical system simulation

[0069] The front optical system realizes vertical field imaging and horizontal uniform illumination. The optical channel includes a cylindrical telescope system and a collimation system. The product of the field of view and the optical transmittance constitutes a transmission coefficient to reduce the signal intensity. The thermal radiation of the instrument has a negligible effect on the visible and near-infrared spectrum, but will have an impact on the infrared spectrum. Therefore, the unit blackbody radiation simulation at a temperature of T is used and calculated by Planck's formula.

[0070] The front optical system takes into account the thermal radiation, field of view and incident light transmission efficiency of the instrument. Incident radiance spectrum L input After (σ) passes through the front optical system, its output spectrum brightness value is L′(σ).

[0071]

[0072] Where Ω1 is the field of view of the incident optical system, ψ1 is the transmission rate of the optical system, and L self,1 is the thermal radiation of the front optical system, which is calculated by Planck's formula:

[0073]

[0074] In the above formula, σ is the wave number, h is Planck's constant, c is the speed of light, and k is B is Boltzmann's constant, and T is the blackbody temperature.

[0075] Step 3: Interferometer system simulation

[0076] The optical signal collimated by the front optical system enters the interferometer system to interfere. The interferometer system consists of a beam splitter and a diffraction grating. The two arms of the grating are equal in length and have different diffraction efficiencies. The effects of phase error, grating angle error, and offset error on the interference signal are considered in the interference.

[0077] The incident spectrum is split into two beams of light with equal energy after passing through the interferometer beam splitter, and then diffracted by the two arm gratings respectively. The diffracted light is combined into one beam of light again through the beam splitter. The grating efficiency in diffraction, the angle measurement deviation of the two arm gratings, and the fluctuation of the spectral optical path difference caused by instrument vibration will affect the accuracy of the interference signal. Therefore, the interference intensity I(x) after passing through the interference system is:

[0078]

[0079] Where η A and η B are the diffraction efficiencies of the two arm gratings, x max is the strongest interference position of the grating, W is the grating length, θ L is the interference Littrow angle, θ error is the Littrow angle error, x is the grating interference position, x error is the grating interference position offset error, p error is the phase error, σ is the wavelength, and σ0 is the Littrow wavelength.

[0080] Step 4: Post-optical system simulation

[0081] The high-resolution interference pattern output by the interferometer system will eventually be imaged on the detector after passing through the post-optical system. The post-optical system impact modeling also takes into account the light transmission efficiency and instrument thermal radiation. The light transmission efficiency mainly considers the field of view and optical transmittance, and the instrument thermal radiation value is still simulated by unit blackbody radiation at temperature T.

[0082] The post-optical system takes into account the thermal radiation, field of view and output light transmission rate of the instrument. After the interference signal passes through the output optical system, its output interference intensity value is I'(σ).

[0083] I′(x)=ψ2I(x)+Ω2·ψ2·L self,2 (σ,T) (9)

[0084] Where ψ2 is the transmission rate of the output optical system, Ω2 is the field of view of the output optical system, and L self,2 (σ,T) is the thermal radiation of the instrument at this location, and the calculation formula is the same as (6).

[0085] Step 5: Low-noise electronics system simulation

[0086] The high-resolution interference signal is output as the final interference pattern after detector resampling, detector response and noise influence. The low-noise electronics system modeling mainly includes detector noise and signal circuit noise, and the total noise obeys the normal distribution. As shown in Figure 2(a) and Figure 2(b).

[0087] Interference signal I imaged on the detectorm (x m ) intensity is related to the detector spectral response function and noise:

[0088] I(N_pixel)=f(I′(x),N_pixel) (10)

[0089] Where I(N_pixel) is the interference signal after being sampled by the detector, N_pixel is the number of pixels in the horizontal direction of the detector, and f is a one-dimensional linear interpolation function.

[0090] I m (x m )=R(σ,x)×I(N_pixel)+I σ (11)

[0091]

[0092] In the formula, I m (x m ) is the interference signal finally output by the detector, R(σ,x) is the spectral response function of the detector, and x m is the wave number position of the interference signal corresponding to the detector pixel, FWHM is the spectral bandwidth, Sd is the pixel size, I σ is the total noise intensity of the detector.

[0093] The key aspects of low noise electronics include detector noise and circuit noise. Detector noise includes photon noise σ shot , dark current σ dark .

[0094] Photon noise:

[0095] Temperature fluctuations, especially blackbody temperature fluctuations during medium and long wave observations, can cause calibration errors in photon noise. An approximate calibration error rate is used to describe photon noise:

[0096]

[0097] In the formula, A pixel is the detector pixel area, Tint is the integration time, h is Planck's constant, c is the speed of light, σ mean is the average wavelength in the spectral range, QE is the quantum efficiency, e cal is the calibration error rate.

[0098] Dark Current:

[0099]

[0100] In the formula, A ais the entrance pupil area, D is the detection rate of the device, Sd is the pixel size, F is the detector frame rate, ψ1 and ψ2 are the transmission rates of the front and rear optical systems respectively, and A pixel ,Tint,h,c,σ mean Same meaning as QE.

[0101] Circuit noise includes quantization noise σ quant , preamplifier read noise σ read , bit error rate σ bit The quantization noise model is:

[0102]

[0103] In the formula, e satur is the number of saturated electrons, and M is the number of quantization bits.

[0104] Preamplifier read noise:

[0105]

[0106] Where W is the grating length, θ L is the Littrow angle of interferometry, F is the detector frame rate, R is the detector response, and NPSD is the preamplifier noise power spectral density.

[0107] Bit error rate noise:

[0108]

[0109]

[0110] In the formula, bit error is the bit error rate, SR is the spectral resolution, σ m is the wave number of the grating interference position.

[0111] Finally, the noise of the entire electronics system is:

[0112]

[0113] Noise is in the form of photons and needs to be converted into intensity signals when superimposed with the signal:

[0114]

[0115] Step 6: Data processing

[0116] The interference image output by the electronics system is restored through spectral restoration processes such as flat field correction, error correction, apodization, and Fourier transform to obtain the atmospheric radiance signal (i.e., restored spectral signal). Considering the pixel merging characteristics of the profile direction in the array detector, the signal-to-noise ratio of the image is calculated based on the noise σ of the electronics system, the restored spectral signal, and the number of pixels N merged at different limb heights in the satellite platform numerical model.

[0117] The interference signal is flat-field corrected, phase corrected, offset removed, interferometric toe-cut, and Fourier transformed to obtain the radiance signal. In order to evaluate the detection and inversion capabilities of the system in different spectral ranges and different layer heights, the effect of detector pixel merging on the system signal-to-noise ratio is considered in the modeling. For the ultraviolet, visible, near-infrared, and short-wave infrared spectral bands, the signal-to-noise ratio is modeled as follows:

[0118]

[0119] Where SNR is the signal-to-noise ratio at different limb heights, L is the restored spectral radiance at different limb heights, σ ​​is the noise at different limb heights, and N is the number of pixels merged at different limb heights.

[0120] For mid- and long-wave infrared limb observations, background thermal radiation is also a factor affecting the performance index calculation:

[0121]

[0122] Where T is the target equivalent blackbody temperature at different heights, T0 is the system background temperature, L is the medium and long wave spectral radiance at different limb heights, σ ​​is the noise at different limb heights, and N is the number of pixels merged at different limb heights.

[0123] Step 7: Image evaluation

[0124] Image evaluation includes the evaluation of the quality and application capability of the restored spectrum. The quality evaluation of the restored spectrum uses the mean spectral similarity (MSS) and the root mean square error (RMSE) as quantitative criteria for the similarity between the restored spectrum and the original input spectrum to ensure the accuracy of the restored spectrum and thus the accuracy of the signal-to-noise ratio calculation. The application capability of the restored spectrum uses the signal-to-noise ratio as an evaluation indicator. If it is better than the maximum signal-to-noise ratio required for the inversion of atmospheric parameters, it is considered that the design indicators at this time meet the inversion requirements. If it does not meet the indicator requirements, the parameters within each module are readjusted for optimization.

[0125] The application capability is evaluated using the signal-to-noise ratio as an indicator. When the calculated signal-to-noise ratio is greater than the maximum signal-to-noise ratio required for atmospheric parameter inversion, the simulation data quality has the inversion capability and the matching instrument design indicators are given.

[0126]

[0127] Where L is the restored spectral radiance signal, L input is the original input signal, n is the number of sampling points in the spectral range, is the average value of the original input signal, and corr is the Pearson correlation coefficient.

[0128] The present embodiment also provides a low-noise spatial heterodyne hyperspectral detection simulation and evaluation system, which includes: a first module, used to simulate a satellite platform to obtain a satellite platform model; a second module, used to simulate a camera to obtain a camera model; the camera model is set on the satellite platform model; a third module, used to output an interference image after the incident light passes through the camera model; a fourth module, used to perform spectral restoration processing on the interference image to obtain a restored spectral signal, and obtain the signal-to-noise ratio of the image according to the restored spectral signal, the electronic system noise in the camera model and the number of pixels N merged at different edge heights in the satellite platform model; a fifth module, used to obtain the root mean square error RMSE and the average spectral similarity MSS according to the incident light and the restored spectral signal.

[0129] This embodiment proposes a low-noise simulation and data quality assessment method integrating platform-payload-spectral restoration according to the mechanism and requirements of hyperspectral detection of near-space atmospheric parameters and the characteristics of space heterodyne instruments, providing technical support for the comprehensive system analysis and optimization of hyperspectral detection solutions for near-space atmospheric parameters, and providing quantitative guarantee for the comprehensive verification and improvement of application capabilities.

[0130] This embodiment can solve the problem of missing near-space detection data sets, form a highly reliable simulation data set, provide technical support for the comprehensive system analysis and optimization of near-space atmospheric parameter hyperspectral detection solutions, and provide quantitative guarantees for comprehensive verification and improvement of application capabilities; the low-noise spatial heterodyne hyperspectral detection simulation method proposed in this embodiment starts with the analysis of detector noise characteristics, analyzes and models the detector output response and preamplifier output noise, and realizes the correlation between noise and design indicators such as integration time, entrance pupil aperture, and spectral resolution, which is helpful for subsequent quantitative analysis of system performance; this embodiment takes into account the impact of the low-noise detection system on the signal-to-noise ratio, establishes a system performance characterization and evaluation method based on the core indicator of the signal-to-noise ratio, and realizes the identification of detection data quality at different edge heights.

[0131] Although the present invention has been disclosed as above in the form of a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art may make possible changes and modifications to the technical solution of the present invention by using the methods and technical contents disclosed above without departing from the spirit and scope of the present invention. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall fall within the protection scope of the technical solution of the present invention.

Claims

1. A low-noise spatial heterodyne hyperspectral detection simulation and evaluation method, characterized in that include: Simulating the satellite platform to obtain a satellite platform model; Simulate the camera to obtain a camera model; The incident light passes through the camera model and then outputs an interference image; The interference image is processed by spectral restoration to obtain a restored spectral signal, and the signal-to-noise ratio of the image is obtained according to the restored spectral signal, the electronic system noise in the camera model and the satellite platform model; The root mean square error and average spectral similarity are obtained based on the incident light and the restored spectral signal.

2. The low-noise spatial heterodyne hyperspectral detection simulation and evaluation method according to claim 1, characterized in that: The camera model obtained by camera simulation includes: The front optical system is simulated to obtain a front optical system model; The interferometer system is simulated to obtain an interferometer system model; The rear optical system is simulated to obtain a rear optical system model; The electronic system is simulated to obtain an electronic system model.

3. The low-noise spatial heterodyne hyperspectral detection simulation and evaluation method according to claim 2, characterized in that: The incident light passes through the front optical system model and outputs a collimated light signal, and the collimated light signal passes through the interferometer system model and outputs an interference pattern. The interference pattern is converged to the electronics system model through the rear optical system model, and the electronics system model samples the converged interference pattern to obtain an interference image.

4. The low-noise spatial heterodyne hyperspectral detection simulation and evaluation method according to claim 1, characterized in that: The satellite platform model is obtained by the following formula: Among them, S is the distance between the satellite and the limb observation point, R is the equatorial radius, H is the satellite orbit height, h is the height of the optical axis tangent point, W is the instrument imaging width, and W v is the vertical component of the instrument imaging width, FOV is the instrument field of view, FOV v is the vertical component of the instrument’s field of view, FOV H is the horizontal component of the instrument's field of view, N_pixel v is the number of pixels of effective imaging in the vertical direction of the array detector, N_pixel is the number of pixels of effective imaging in the horizontal direction of the array detector, K is the profile resolution of the instrument, and N is the number of pixels combined in the vertical direction.

5. The low-noise spatial heterodyne hyperspectral detection simulation and evaluation method according to claim 1, characterized in that: The signal-to-noise ratio of the image is obtained by the following formula: Among them, SNR is the signal-to-noise ratio of the images corresponding to the ultraviolet, visible, near-infrared and short-wave infrared spectral bands, NETD is the signal-to-noise ratio of the images corresponding to the medium- and long-wave infrared spectral bands, L is the restored spectral signal, σ is the electronic system noise, N is the number of pixels merged at different limb heights, T0 is the background temperature, and T is the equivalent blackbody temperature of the target at different heights.

6. The low-noise spatial heterodyne hyperspectral detection simulation and evaluation method according to claim 5, characterized in that: The electronic system noise σ is obtained by the following formula: Where σ is the electronic system noise, σ shot is the photon noise, σ dark is the dark current, σ quant is the quantization noise, σ read is the preamplifier read noise, σ bit is the bit error rate, A pixel is the detector pixel area, Tint is the integration time, h is Planck's constant, c is the speed of light, σ mean is the average wavelength in the spectral range, QE is the quantum efficiency, e cal is the calibration error rate, I m (x m ) is the interference image, x m is the wave number position of the interference image corresponding to the detector pixel, A a is the entrance pupil area, D is the detection rate of the device, Sd is the pixel size, F is the detector frame rate, ψ1 and ψ2 are the transmission rates of the front optical system model and the rear optical system model respectively, e satur is the number of saturated electrons, M is the number of quantization bits, W is the grating length, θ L is the Littrow angle of interference, R is the detector response, NPSD is the preamplifier noise power spectral density, bit error is the bit error rate, SR is the spectral resolution, σ m is the wave number of the grating interference position.

7. The low-noise spatial heterodyne hyperspectral detection simulation and evaluation method according to claim 1, characterized in that: The root mean square error RMSE is obtained by the following formula: Where L is the restored spectral radiance signal, L input is the incident light signal, and n is the number of sampling points in the spectral range.

8. The low-noise spatial heterodyne hyperspectral detection simulation and evaluation method according to claim 1, characterized in that: The average spectral similarity MSS is obtained by the following formula: Where L is the restored spectral radiance signal, L input is the incident light signal, is the average value of the incident light signal, corr is the Pearson correlation coefficient, RMSE is the root mean square error, and mean() is the average value of the spectral dimension signal.

9. The low-noise spatial heterodyne hyperspectral detection simulation and evaluation method according to claim 1, characterized in that: The electronics system model is obtained by the following formula: I m (x m )=R(σ,x)×I(N_pixel)+I σ ; I(N_pixel)=f(I′(x),N_pixel); Among them, I m (x m ) is the interference image, I(N_pixel) is the interference signal after being sampled by the detector, N_pixel is the number of pixels in the horizontal direction of the detector, f() is the one-dimensional linear interpolation function, R(σ,x) is the spectral response function of the detector, x m is the wave number position of the interference signal corresponding to the detector pixel, FWHM is the spectral bandwidth, Sd is the pixel size, I σ is the total noise intensity of the detector, σ is the noise of the electronic system, A pixel is the detector pixel area, Tint is the integration time, h is Planck's constant, c is the speed of light, σ m is the wave number of the grating interference position, QE is the quantum efficiency, x is the grating interference position, and I′(x) is the output interference intensity value of the post-optical system.

10. A low-noise spatial heterodyne hyperspectral detection simulation and evaluation system, characterized in that include: The first module is used to simulate the satellite platform to obtain a satellite platform model; The second module is used to simulate the camera to obtain a camera model; and set the camera model on the satellite platform model; The third module is used to output an interference image after the incident light passes through the camera model; The fourth module is used to perform spectral restoration processing on the interference image to obtain a restored spectral signal, and obtain the signal-to-noise ratio of the image according to the restored spectral signal, the electronic system noise in the camera model, and the number of pixels N combined at different limb heights in the satellite platform model; The fifth module is used to obtain the root mean square error RMSE and the average spectral similarity MSS according to the incident light and the restored spectral signal.

Images