Space-based remote sensing imaging characteristic-based space-air image equivalent conversion method

Abstract

An air-space image equivalent conversion method based on space-based remote sensing imaging characteristics, comprising: collecting and preprocessing the air-based image; taking the preprocessed air-based image as an input data source, converting the air-based measured image radiance to the detector response electron number through the energy transmission model; based on the detector response electron number output by the energy transmission model, superimposing noise through the noise model to simulate the noise characteristics of the space-based imaging, obtaining the electron number distribution with noise; Fourier transforming the electron number distribution with noise, multiplying the total MTF determined through the MTF model, and then inverse transforming to obtain the degraded electron number image; quantizing and format converting the degraded electron number image to obtain the equivalent space-based remote sensing image. The application realizes the equivalent conversion of space-based remote sensing images under different scenes.

Description

Technical Field

[0001] This invention relates to a method for equivalent conversion of space-based remote sensing images based on the characteristics of space-based remote sensing imaging, belonging to the field of optical remote sensing image processing technology. Background Technology

[0002] Efficient and accurate equivalent simulation of low-altitude UAV aerial images and space-based satellite images can provide data support for target identification and detection from different perspectives. Based on high-resolution images acquired by UAVs, the input images are converted into equivalent satellite images through processes such as atmospheric parameters, optical systems, detectors and electronics, platform motion, and image sampling.

[0003] However, space-based satellite imagery, acquired at long distances using sensors carried by satellite platforms, typically has a resolution between 0.3 meters and 30 meters. Unmanned aerial vehicle (UAV) imagery, on the other hand, is acquired at close range using sensors carried by UAV platforms at low altitudes, capturing various ground targets with higher resolution, typically between 0.01 meters and 0.1 meters.

[0004] Due to factors such as atmospheric transmission loss, shooting altitude and angle, optical images acquired by UAVs differ significantly from real satellite images in terms of spatial scale, image resolution and color tone, making it difficult to accurately reflect the space-based satellite imaging characteristics of the target. Summary of the Invention

[0005] The technical problem solved by this invention is: addressing the multi-scale space-based remote sensing equivalence problem, this invention provides a space-based image equivalence conversion method based on the imaging characteristics of space-based remote sensing, realizing the equivalence conversion of space-based remote sensing images under different scenarios.

[0006] The technical solution of this invention is:

[0007] A method for equivalent transformation of space-based remote sensing images based on their imaging characteristics includes:

[0008] Acquire and preprocess spatial base images;

[0009] Using the preprocessed spaceborne image as the input data source, the energy transfer model is used to convert the measured radiance of the spaceborne image into the number of electrons in the detector response.

[0010] Based on the detector response electron count output by the energy transfer model, noise is superimposed through a noise model to simulate the noise characteristics of space-based imaging, thus obtaining the noisy electron count distribution.

[0011] The noisy electron number distribution is subjected to Fourier transform, multiplied with the total MTF determined by the MTF model, and then inversely transformed to obtain the degraded electron number image.

[0012] The degraded electronic digital image is quantized and format converted to obtain an equivalent space-based remote sensing image.

[0013] Furthermore, when acquiring airborne images, a low-altitude unmanned aerial vehicle (UAV) platform is used to acquire high-resolution optical images of the target area as airborne images. The image resolution is 0.01 meters to 0.1 meters, and the spectral range covers the visible to near-infrared bands.

[0014] Furthermore, the acquired airborne images undergo preprocessing, specifically including radiometric correction and geometric correction. Radiometric correction utilizes UAV payload calibration parameters to remove dark current noise from the images and converts the original image grayscale values ​​into radiance values ​​L. in (λ); Geometric correction is based on UAV GPS, IMU data and ground control point locations to eliminate geometric distortions in the image caused by platform jitter and terrain undulations, so that the coordinate system of the airborne image is consistent with the geographic coordinate system of the spaceborne remote sensing image.

[0015] Furthermore, the energy transfer model is used to simulate the energy attenuation and electron conversion of airborne radiance after passing through the atmosphere, optical system, and detector.

[0016] Furthermore, the energy transfer model is specifically as follows:

[0017]

[0018] Among them, S target A represents the total number of electrons in the response of the empty-base image after passing through the optical system and detector; detector ε is the detector pixel area; ε is the optical aperture area obstruction ratio, which is 0 for an unobstructed optical system; F is the optical system f-number; t int η(λ) represents the exposure time of the detector in the space-based imaging system; η(λ) is the optical responsivity, which is the average number of photoelectrons generated per incident photon with wavelength λ, also known as quantum efficiency; λ max and λ min τ represents the upper and lower limits of the spectral response range; optics (λ) represents the transmittance of the optical system; L target denoted as radiance of the empty-base image; h is Planck's constant, h = 6.63 × 10⁻⁶. -34 Js, where c is the speed of light, c = 3 × 10 8 m / s.

[0019] Furthermore, the noise model is specifically as follows:

[0020] Where, n sys Let n be the standard deviation of the camera system noise. shot Represents shot noise, n dark Indicates dark current noise, nquantizatio Indicates quantization noise, n det This represents the random noise of the detector.

[0021] Furthermore, for shot noise, the variance of shot noise is equal to the sum of the root mean squares of photoelectrons and dark current electrons, that is: <n 2 shot >= <n 2 pe >+ <n 2 dark >=n pe +n dark ;

[0022] in, <n 2 shot > is the variance of the equivalent electron number in shot noise, n pe It is the average number of signal electrons, n dark It is the number of electrons in the dark current;

[0023]

[0024] A D J is the effective area of ​​the depletion layer. D t represents the dark current density. int Exposure time, R e (λ) is the spectral response. It is the incident amount of the detector, M e (λ) is the monochromatic radiance, A s R1 is the ideal light source area, q is the distance between the array and the light source, λ1 and λ2 are the upper and lower limits of the detector response wavelength.

[0025] Furthermore, for dark current noise, we have n 2 pd =n dark , where n pd n represents the number of dark signal electrons. pd =t int ·v d v d This indicates the rate at which dark signals are generated.

[0026] Furthermore, regarding quantization noise, there is

[0027] Where, n well_depth denoted as the number of electrons in the full trap, and b represents the number of quantization bits.

[0028] Furthermore, the MTF model is as follows:

[0029] MTF sensor =MTF optics ×MTFdetector ×MTF cir ×MTF atm

[0030] Among them, MTF sensor Indicates the total MTF of the space-based imaging system.

[0031] The specific MTF of the optical system is: MTF optics =MTF diff ×MTF aberratio × n MTF defo

[0032] Among them, MTF diff For diffraction MTF, MTF aberration For aberration MTF, MTF defocus This is the defocused MTF;

[0033] The detector's MTF is specifically: MTF detector =MTF sl ×MTF diffusio × n MTF tra_nsc

[0034] Among them, MTF sl Let MTF be the detector's geometric transfer function. diffusion MTF is the photoelectron diffusion transfer function. trans_ccd The charge conversion efficiency transfer function;

[0035] The circuit MTF is specifically: MTF cir =MTF amp ×MTF filter

[0036] Among them, MTF amp MTF is the modulation transfer function of the amplifier. filter The filter modulation transfer function;

[0037] MTF atm This represents the atmospheric MTF, which is estimated based on the atmospheric turbulence, aerosol, and water vapor content at the time of imaging in the imaging area. It is usually taken as an empirical value of 0.80.

[0038] Furthermore, the quantization and format conversion specifically refers to:

[0039] The degraded electron number image is converted into DN values ​​using analog-to-digital conversion. The conversion formula is as follows:

[0040] DN = clip((N) sig+noise / N well )×2 b ,0,2 b-1)

[0041] Where clip represents a function operation, N sig+noise For the number of noisy electrons, N well is the full-well capacity, and b is the number of bits for quantization.

[0042] The clip(a, a_min, a_max) function replaces elements in the input array a that are outside the range [a_min, a_max] with the corresponding boundary values ​​by setting minimum and maximum values.

[0043] The advantages of this invention compared to the prior art are:

[0044] To address the issue of multi-scale space-based remote sensing equivalence, this invention combines the influencing factors of optical remote sensing imaging during the imaging process to establish a multi-scale space-based remote sensing equivalent imaging model. Using space-based measured image data as the input data source, the model is converted into entrance pupil radiance, and imaging environment parameters are used as variable conditions to equivalently convert the model into space-based remote sensing images under different scenarios. Attached Figure Description

[0045] Figure 1 This is a schematic diagram illustrating the conversion process of aerospace images;

[0046] Figure 2 This is a schematic diagram of a multi-scale space-based remote sensing equivalent imaging model. Detailed Implementation

[0047] The specific embodiments of the present invention will now be described in further detail with reference to the accompanying drawings.

[0048] To address the problem of equivalent conversion of space-based images, a multi-scale space-based remote sensing equivalent imaging model is established to characterize the imaging radiative transfer and detector response processes. This model simulates various stages such as the attenuation of the remote sensor's optical system, the photoelectric conversion of the detector, and sampling quantization. Based on the space-based image as the radiance at the entrance pupil, and combined with non-ideal characteristics such as optical system distortion, detector noise, and circuit noise, an equivalent space-based remote sensing image is generated.

[0049] The multi-scale space-based remote sensing equivalent imaging model mainly includes the energy transfer model, the noise model, and the MTF model. The energy transfer model includes processes such as the imaging optical system and the detector response, which respectively describe the energy attenuation of imaging data by the imaging optical system and the detector electronics system. The noise model and the MTF model mainly describe the blurring and degradation of energy.

[0050] (1) Energy transfer model

[0051] The energy transfer model is used to quantitatively describe the physical process by which the radiance of a space-based image is converted into the number of electrons in the detector response via the space-based remote sensing imaging link. Its core is to establish a precise mapping from space-based input to space-based detector output by integrating energy attenuation and photoelectric conversion mechanisms in multiple stages.

[0052] The total number of electrons S in the response of the empty-base image after passing through the optical system and detector. target for:

[0053]

[0054] Among them, A detector ε is the detector pixel area; ε is the optical aperture area obstruction ratio, which is 0 for an unobstructed optical system; F is the system f-number; t int η(λ) represents the exposure time of the imaging system; η(λ) is the optical responsivity, which is the average number of photoelectrons generated per incident photon with wavelength λ, also known as quantum efficiency; λ max and λ min τ represents the upper and lower limits of the spectral response range; optics (λ) represents the transmittance of the optical system; L target denoted as radiance of the image acquired using the airborne method; h is Planck's constant, h = 6.63 × 10⁻⁶. -34 Js, where c is the speed of light, c = 3 × 10 8 m / s.

[0055] (2) Noise Model

[0056] The noise model of an imaging system is a mathematical description of various interference signals during image acquisition, and it quantitatively characterizes the sources and distribution characteristics of noise.

[0057] In practical simulations, detector noise can be calculated based on the parameters provided in the detector manual, while circuit noise is fitted using laboratory test data, such as establishing Gaussian white noise. The main noise types considered are shot noise, mode noise, dark current noise, and system noise.

[0058] The standard deviation of the camera system noise is n sys The calculation formula is as follows:

[0059]

[0060] Among them, shot noise n shot Noise caused by signals during photoelectric conversion; detector random noise n det It is the noise generated during charge readout, consisting of detector readout noise, reset noise, 1 / f noise caused by the amplifier circuit, and white noise, etc.; dark current noise n dark This is mainly caused by the dark signal from the detector; n quantizationThis represents quantization noise.

[0061] The process of generating signal charges by light injection into the photosensitive area of ​​a CCD can be regarded as an independent, uniform, and continuous random process. The number of signal charges generated by light per unit time is not absolutely constant, but fluctuates slightly randomly around an average value; this is shot noise. Shot noise is independent of frequency and has a uniform power distribution over a wide frequency range, often referred to as white noise. It determines the noise limit of the device. Both electrons and dark current will generate shot noise, which can be described by a Poisson distribution. The variance of shot noise is equal to the sum of the root mean squares of photoelectrons and dark current electrons, as shown in formula (3).

[0062] n 2 shot >= <n 2 pe >+ <n 2 dark >=n pe +n dark (3)n 2 shot > represents the variance of the equivalent electron number in the ejection noise, n pe It is the average number of signal electrons, n DARK The number of dark current electrons is calculated using formulas (4) and (5), respectively.

[0063]

[0064] A D The effective area of ​​the depletion layer, J D t represents the dark current density. int Exposure time, R e (λ) is the spectral response. It is the incident amount on the detector. M e (λ) is the monochromatic radiance, A s R1 is the ideal light source area, and q is the distance between the array and the light source. λ1 and λ2 are the upper and lower limits of the detector response wavelength.

[0065] The n mentioned above pe n dark Both are affected by charge transfer efficiency (CTE), and therefore have

[0066] <n 2 shot >= <n 2 pe >+ <n 2 dark >=(CTE) N n pe +(CTE) N n dark(6)

[0067] Detector random noise is the noise generated during charge readout. It consists of detector readout noise, reset noise, 1 / f noise caused by the amplifier circuit, and white noise.

[0068] Dark current noise is mainly caused by the dark signal from the detector, and n 2 pd =n dark , where n pd n represents the number of dark signal electrons. pd =t int ·v d v d This indicates the rate at which dark signals are generated.

[0069] n quantization The standard deviation of the quantization noise is calculated using the following formula:

[0070]

[0071] Where, n well_depth denoted as b, representing the number of electrons in the full-well of the detector and the number of quantization bits.

[0072] (3) MTF model

[0073] The MTF model quantifies the transmission capability of signals at different spatial frequencies, accurately characterizing the impact of various processes in remote sensing imaging on image quality. MTF degradation models are established for the four components of the remote sensor: optical system, detector, circuitry, and atmosphere, as described below:

[0074] MTF sensor =MTF optics ×MTF detector ×MTF cir ×MTF atm (8)

[0075] The MTF model of an optical system includes MTF for diffraction, aberration, defocus, and other aspects, namely:

[0076] MTF optics =MTF diff ×MTF aberration ×MTF defocus (9)

[0077] Among them, MTF diff For diffraction MTF, MTF aberration For aberration MTF, MTF defocus This is the defocused MTF.

[0078] The detector's MTF model is as follows:

[0079] MTFdetector =MTF sl ×MTF diffusion ×MTF trans_ccd (10)

[0080] Among them, MTF sl MTF is the detector's geometric transfer function. diffusion MTF is the photoelectron diffusion transfer function. trans_ccd This is the charge conversion efficiency transfer function.

[0081] The circuit MTF model is as follows:

[0082] MTF cir =MTF amp ×MTF filter (11)

[0083] Among them, MTF amp MTF is the modulation transfer function of the amplifier. filter This is the filter modulation transfer function.

[0084] MTF atm This represents the atmospheric MTF, which is estimated based on the atmospheric turbulence, aerosol, and water vapor content at the time of imaging in the imaging area. It is usually taken as an empirical value of 0.80.

[0085] Example:

[0086] The space-based image equivalent conversion method based on space-based remote sensing imaging characteristics described in this invention converts space-based images into equivalent space-based remote sensing images through multi-scale space-based remote sensing equivalent imaging models (energy transfer model, noise model, MTF model). The specific implementation steps are as follows: Figure 1 , 2 As shown:

[0087] (1) Acquisition and preprocessing of spatial images.

[0088] a. Spaceborne image acquisition

[0089] High-resolution optical images of the target area were acquired using a low-altitude unmanned aerial vehicle (UAV) platform. The image resolution ranged from 0.01 meters to 0.1 meters (consistent with the typical resolution range of airborne images), and the spectral range covered the visible to near-infrared band (0.4 μm-0.9 μm). During acquisition, auxiliary information such as flight altitude, attitude angle, and atmospheric conditions (e.g., visibility, aerosol optical thickness) at the time of imaging were recorded.

[0090] b. Preprocessing of spaceborne images

[0091] Airborne image preprocessing includes radiometric correction and geometric correction. Radiometric correction uses UAV payload calibration parameters (such as dark current and gain) to remove dark current noise from the image and converts the original image grayscale values ​​(DN values) into radiance values ​​L. in (λ). Geometric correction, based on UAV GPS / IMU data and ground control points, eliminates geometric distortions caused by platform jitter and terrain undulations, ensuring that the image coordinate system is consistent with the geographic coordinate system of the space-based remote sensing image, thus laying the foundation for subsequent scale transformation.

[0092] (2) Parameter configuration and calculation of energy transfer model: The preprocessed space-based image is used as the input data source. The energy transfer model is used to convert the measured radiance of the space-based image into the number of electrons in the detector response.

[0093] The energy transfer model is used to simulate the energy attenuation and electron conversion of airborne radiance after passing through the atmosphere, optical system, and detector. The specific implementation is as follows:

[0094] a. Parameter Input

[0095] Input atmospheric parameters, optical system parameters, and detector parameters.

[0096] b. Pixel-by-pixel energy conversion calculation

[0097] For each pixel of the preprocessed empty-base image, substitute it into formula (1) to calculate the total number of electrons S in the detector response. target Where, the integration interval λ min =0.4μm, λ max =0.9μm, and the electron number distribution of each pixel is obtained by numerical integration (such as the trapezoidal method).

[0098] (3) Noise superposition of noise model: Based on the number of detector response electrons output by the energy transfer model, noise is superimposed through the noise model to simulate the noise characteristics of space-based imaging and obtain the distribution of electrons with noise.

[0099] The number of electrons S output based on the energy transfer model target By superimposing various types of noise using a noise model, the noise characteristics of space-based imaging are simulated.

[0100] a. Noise parameter configuration

[0101] Number of electrons in the full trap n well_depth =10 5 e - The quantization bit depth b = 12 bits.

[0102] b. Calculation of each noise component

[0103] Shot noise:

[0104] <n 2 shot >=n pe +n dark (12)

[0105] Dark current noise:

[0106]

[0107] Quantization noise:

[0108]

[0109] c. Total noise superposition

[0110] The formula for calculating the standard deviation of total noise is as follows:

[0111]

[0112] A noise matrix is ​​generated by a random number generator (such as a Gaussian distribution), and then superimposed on the electron number matrix to obtain a noisy electron number distribution.

[0113] (4) Image degradation of MTF model: Perform Fourier transform on the noisy electron number distribution, multiply it with the total MTF determined by the MTF model, and then perform inverse transform to obtain the degraded electron number image.

[0114] The imaging quality degradation of a space-based imaging system is simulated using an MTF model. The specific steps are as follows:

[0115] a. MTF parameter configuration

[0116] Optical System MTF optics Detector MTF detector and circuit MTF cir The calculation formulas are (9), (10), and (11), respectively.

[0117] b. Total MTF Calculation and Degradation

[0118] System Total MTF sensor The calculation formula is as follows:

[0119] MTF sensor =MTF optics ×MTF detector ×MTF cir ×MTF atm (16)

[0120] A Fourier transform is performed on the noisy electron count image, which is then multiplied by the MTF curve and inversely transformed to obtain the degraded electron count image.

[0121] (5) Post-processing and output: Quantize and convert the degraded electron image to obtain an equivalent space-based remote sensing image.

[0122] a. Quantization and format conversion

[0123] The degraded electron number image is converted to DN values ​​using an analog-to-digital converter (ADC). The conversion formula is as follows:

[0124] DN = clip((N) sig+noise / N well )×2 b ,0,2 b -1) (17)

[0125] The output is a 12-bit grayscale image.

[0126] Where clip represents a function operation, N sig+noise For the number of noisy electrons, N well is the full-well capacity, and b is the number of bits for quantization.

[0127] The clip(a, a_min, a_max) function replaces elements in the input array a that are outside the range [a_min, a_max] with the corresponding boundary values ​​by setting minimum and maximum values.

[0128] b. Result Verification

[0129] Compare and analyze the output equivalent space-based image with the actual space-based satellite image (same region, same band):

[0130] Radiation characteristics: Calculate the average radiance difference (≤5%) between the two on the same ground features (such as vegetation and water bodies);

[0131] Noise characteristics: Signal-to-noise ratio (SNR) difference ≤ 3dB;

[0132] Spatial characteristics: The deviation of the MTF curve at 0.5 lp / mm is ≤0.1, ensuring that the equivalence meets the application requirements.

[0133] The parts of this invention not described in detail are common knowledge to those skilled in the art.

Claims

1. A method for equivalent conversion of space-based remote sensing images based on their imaging characteristics, characterized in that... include: Acquire and preprocess spatial base images; Using the preprocessed spaceborne image as the input data source, the energy transfer model is used to convert the measured radiance of the spaceborne image into the number of electrons in the detector response. Based on the detector response electron count output by the energy transfer model, noise is superimposed through a noise model to simulate the noise characteristics of space-based imaging, thus obtaining the noisy electron count distribution. The noisy electron number distribution is subjected to Fourier transform, multiplied with the total MTF determined by the MTF model, and then inversely transformed to obtain the degraded electron number image. The degraded electronic digital image is quantized and format converted to obtain an equivalent space-based remote sensing image.

2. The method for equivalent transformation of aerospace images based on space-based remote sensing imaging characteristics according to claim 1, characterized in that: When acquiring airborne images, a low-altitude unmanned aerial vehicle (UAV) platform is used to acquire high-resolution optical images of the target area as airborne images. The image resolution is 0.01 meters to 0.1 meters, and the spectral range covers the visible to near-infrared bands.

3. The method for equivalent transformation of aerospace images based on space-based remote sensing imaging characteristics according to claim 1, characterized in that: The acquired airborne images undergo preprocessing, specifically including radiometric correction and geometric correction. Radiometric correction utilizes UAV payload calibration parameters to remove dark current noise from the image and converts the original image grayscale values ​​into radiance values ​​L. in (λ); Geometric correction is based on UAV GPS, IMU data and ground control point locations to eliminate geometric distortions in the image caused by platform jitter and terrain undulations, so that the coordinate system of the airborne image is consistent with the geographic coordinate system of the spaceborne remote sensing image.

4. The method for equivalent transformation of aerospace images based on space-based remote sensing imaging characteristics according to claim 1, characterized in that: The energy transfer model is used to simulate the energy attenuation and electron conversion of airborne radiance after passing through the atmosphere, optical system, and detector.

5. The method for equivalent transformation of aerospace images based on space-based remote sensing imaging characteristics according to claim 1, characterized in that: The energy transfer model is as follows: Among them, S target A represents the total number of electrons in the response of the empty-base image after passing through the optical system and detector; detector ε is the detector pixel area; ε is the optical aperture area obstruction ratio, which is 0 for an unobstructed optical system; F is the optical system f-number; t int η(λ) represents the exposure time of the detector in the space-based imaging system; η(λ) is the optical responsivity, which is the average number of photoelectrons generated per incident photon with wavelength λ, also known as quantum efficiency; λ max and λ min τ represents the upper and lower limits of the spectral response range; optics (λ) represents the transmittance of the optical system; L target denoted as radiance of the empty-base image; h is Planck's constant, h = 6.63 × 10⁻⁶. -34 Js, where c is the speed of light, c = 3 × 10 8 m / s.

6. The method for equivalent transformation of aerospace images based on space-based remote sensing imaging characteristics according to claim 5, characterized in that: The noise model is specifically as follows: Where, n sys Let n be the standard deviation of the camera system noise. shot Represents shot noise, n dark Indicates dark current noise, n quantizatio Indicates quantization noise, n det This represents the random noise of the detector.

7. The method for equivalent transformation of aerospace images based on space-based remote sensing imaging characteristics according to claim 6, characterized in that: For shot noise, the variance of shot noise is equal to the sum of the root mean squares of photoelectrons and dark current electrons, that is: <n 2 shot >= <n 2 pe >+ <n 2 dark >=n pe +n dark ; in, <n 2 shot > is the variance of the equivalent electron number in shot noise, n pe It is the average number of signal electrons, n dark It is the number of electrons in the dark current; A D J is the effective area of ​​the depletion layer. D t represents the dark current density. int Exposure time, R e (λ) is the spectral response. It is the incident amount of the detector, M e (λ) is the monochromatic radiance, A s R1 is the ideal light source area, q is the distance between the array and the light source, λ1 and λ2 are the upper and lower limits of the detector response wavelength.

8. The method for equivalent transformation of aerospace images based on space-based remote sensing imaging characteristics according to claim 7, characterized in that: For dark current noise, we have n 2 pd =n dark , where n pd n represents the number of dark signal electrons. pd =t int ·v d v d This indicates the rate at which dark signals are generated.

9. The method for equivalent transformation of aerospace images based on space-based remote sensing imaging characteristics according to claim 7, characterized in that: For quantization noise, there is Where, n well_depth denoted as the number of electrons in the full trap, and b represents the number of quantization bits.

10. The method for equivalent transformation of aerospace images based on space-based remote sensing imaging characteristics according to claim 1, characterized in that: The MTF model is specifically as follows: MTF sensor =MTF optics ×MTF detector ×MTF cir ×MTF atm Among them, MTF sensor Indicates the total MTF of the space-based imaging system. The specific MTF of the optical system is as follows: Among them, MTF diff For diffraction MTF, MTF aberration For aberration MTF, MTF defocus This is the defocused MTF; The detector's MTF is as follows: Among them, MTF sl MTF is the detector's geometric transfer function. diffusion MTF is the photoelectron diffusion transfer function. trans_ccd The charge conversion efficiency transfer function; The circuit MTF is specifically: MTF cir =MTF amp ×MTF filter Among them, MTF amp MTF is the modulation transfer function of the amplifier. filter The filter modulation transfer function; MTF atm The atmospheric MTF is estimated based on atmospheric turbulence, aerosol and water vapor content at the time of imaging in the imaging region.

11. The method for equivalent transformation of aerospace images based on space-based remote sensing imaging characteristics according to claim 1, characterized in that: The quantization and format conversion are specifically as follows: The degraded electron number image is converted into DN values ​​using analog-to-digital conversion. The conversion formula is as follows: DN=clip((N sig+noise / N well )×2 b ,0,2 b -1) Where clip represents a function operation, N sig+noise For the number of noisy electrons, N well b represents the full-well capacity, and b represents the number of bits used for quantization.

Images