Satellite platform flutter influence optical load push-broom imaging simulation test system and method

Abstract

The invention relates to a satellite platform flutter influence optical load push-broom imaging simulation test system and method. The system comprises a satellite platform flutter simulation assembly, a ground object push-broom imaging simulation assembly, a flutter measurement assembly and an image processing and motion control computer. The ground feature push-broom imaging simulation assembly is used for simulating push-broom imaging of a satellite to the ground under different satellite orbit heights; the flutter measurement assembly mainly comprises an accelerometer installed on the zoom optical load and an image processing and motion control computer connected with the zoom optical load, and is used for acquiring amplitude and frequency parameters of the flutter of the zoom optical load in three spatial directions, and the amplitude and frequency parameters are compared and analyzed by the image processing and motion control computer. According to the method, disturbance of flutters of different amplitudes and frequencies generated when a satellite platform works in orbit on push-broom imaging of the remote sensing optical load is simulated, and then support is provided for how to invert remote sensing satellite flutters and how to remove flutter fuzziness of imaging.

Description

Technical Field

[0001] The present invention relates to a system and method for simulating and testing optical payload push-broom imaging influenced by satellite platform vibration, and belongs to the technical field of push-broom remote sensing satellite vibration testing. Background Art

[0002] Microsatellites have been widely used in space remote sensing both domestically and internationally. Currently, they are primarily used for emergency observation of hotspots and emergencies, as well as for high-resolution and long-term observation of specific areas. Microsatellites often use flywheels as attitude control components. These flywheels achieve attitude control by adjusting rotor speed and exchanging angular momentum with the satellite itself. However, these structures can suffer from rotor imbalance and bearing defects, which can generate interference torques and, in turn, cause periodic flutter in the satellite platform. This is also the primary source of microvibration affecting optical payloads, particularly on microsatellites lacking vibration isolation equipment. Microsatellites, such as Skysat and Gaojing-1, have entered the sub-meter range, and the impact of microvibration on payloads cannot be ignored. Due to cost and mass constraints, microsatellites lack vibration isolation platforms to isolate the payload from the effects of flywheel rotation. Flutter above the imaging frequency generally causes image blur, while flutter below the imaging frequency causes image distortion.

[0003] On-orbit experiments on remote sensing satellite platforms for push-broom imaging under the influence of flutter are extremely expensive. Existing simulation systems for optical payload imaging under the influence of flutter are mostly designed for staring imaging rather than push-broom imaging, or they mount the target image on a vibration platform and are unable to directly simulate satellite flutter imaging. The lack of a simulation system specifically for the effects of satellite platform flutter on push-broom imaging of TDI-CMOS optical payloads makes it difficult to provide effective guidance for analyzing the imaging quality of high-resolution optical micro-remote sensing satellites. Therefore, it is necessary to develop a simulation test system for the effects of satellite platform flutter on push-broom imaging of remote sensing optical payloads. This system can simulate the disturbances of satellite platform flutter of varying amplitudes and frequencies on push-broom imaging on remote sensing optical payloads on the ground, providing support for flutter inversion and flutter blur removal for push-broom remote sensing satellites. Summary of the Invention

[0004] The technical problem solved by the present invention is to overcome the deficiencies of the prior art and provide a satellite platform vibration-affected optical payload push-broom imaging simulation test system and method.

[0005] The technical solution of the present invention is:

[0006] The present invention discloses a simulation test system for the effect of satellite platform flutter on remote sensing optical payload push-broom imaging, comprising: a flutter signal generator, a vibration isolation platform, a vibration excitation platform, a zoom optical payload, an air-floating vibration isolation platform, a linear guide rail, a target image, an accelerometer, and an image processing and motion control computer; wherein,

[0007] The vibration excitation platform is mounted on the vibration isolation platform, and the vibration excitation platform generates vibration under the excitation of the signal of the vibration signal generator to simulate the vibration of the satellite platform;

[0008] The zoom optical payload is mounted on the excitation platform to image the target and send the imaging result to the image processing and motion control computer;

[0009] The accelerometer is mounted on the zoom optical payload, collects vibration information of the zoom optical payload, and sends the information to the image processing and motion control computer;

[0010] The linear guide rail is mounted on the air-floating vibration isolation platform and provides uniform linear motion in one dimension;

[0011] The image processing and motion control computer analyzes and processes the imaging results and the vibration information of the accelerometer, and controls the linear guide rail to perform uniform linear motion;

[0012] The target image is installed on the linear guide rail to simulate target ground object information.

[0013] Furthermore, in the above system, the vibration platform provides vibration in the vertical direction and the camera push-scan direction; the maximum output frequency of the vibration platform is greater than or equal to 2 kHz.

[0014] Furthermore, in the above system, the zoom optical payload adopts a TDI-CMOS sensor. The line of sight of the zoom optical payload after installation is perpendicular to the direction of movement along the linear guide rail, and is coordinated with the target image moving in the same direction to simulate the on-orbit push-scan imaging of the remote sensing satellite; the target image is a real remote sensing image or a black and white checkerboard image.

[0015] Furthermore, in the above system, the accelerometer measurement range is -19.6m / s 2 ~19.6m / s 2 The frequency measurement range is 1Hz~2kHz, and the measurement accuracy is greater than 1μm.

[0016] The present invention discloses a test method for a simulation test system for satellite platform flutter affecting remote sensing optical payload push-broom imaging, comprising:

[0017] S1: Move the target image to the center of the linear guide and align the lens of the zoom optical payload with the target image;

[0018] S sets the focal length of the zoom optical payload and the distance between the lens and the target image on the linear guide rail, satisfying L1 / L2=L3 / L4; L1 is the ground simulation focal length, L2 is the ground simulation image distance, L3 is the actual satellite focal length, and L4 is the actual satellite image distance;

[0019] S3: Switch the CMOS sensor of the zoom optical payload to the standard surface imaging mode and focus so that the zoom optical payload can clearly image the target image to obtain the non-flutter image f(x, y);

[0020] S4: Move the target image to one end of the linear guide rail and calculate the motion speed v of the linear guide rail to satisfy Where p is the pixel size, f is the focal length of the optical payload, l is the imaging distance, and T is the integration time of the optical payload;

[0021] S5: Set the vibration frequency and amplitude of the vibration signal generator, turn on the vibration platform, amplify the vibration signal, and the zoom optical payload begins to vibrate along with the vibration platform.

[0022] S6: The target image moves along the linear guide at the speed calculated in step S4, and the zoom optical load images the target image during vibration to obtain a flutter image g(x′, y′);

[0023] S7: The accelerometer collects the vibration information of the zoom optical load and uses the frequency domain analysis method to calculate the frequency and amplitude of the vibration;

[0024] S8: Using the zoom optical payload to image the target, the low-frequency chatter frequency and amplitude are calculated based on sub-pixel matching. The frequency and amplitude are then subtracted from the vibration frequency and amplitude obtained by the accelerometer to obtain the error of the inverted low-frequency chatter frequency and amplitude.

[0025] S9: Using the zoom optical payload to image the target, the high-frequency flutter frequency and amplitude are calculated based on Radon transform. The frequency and amplitude of the high-frequency flutter are subtracted from the frequency and amplitude of the vibration obtained by the accelerometer to obtain the error of the inverted high-frequency flutter frequency and amplitude.

[0026] S10: Obtain a test result of the impact of the flutter according to the inverted low-frequency flutter frequency and amplitude errors and the inverted high-frequency flutter frequency and amplitude errors.

[0027] Furthermore, in the above method, in step S7, the frequency domain analysis method is specifically:

[0028] S7-1: Filter the accelerometer signal during imaging using a 6th-order phase-shift-free Butterfly filter to obtain a filtered signal;

[0029] S7-2: Perform fast Fourier transform on the filtered signal and estimate the vibration frequency based on the spectrum peak;

[0030] S7-3: When the absolute value of the Fourier spectrum amplitude corresponding to each vibration frequency f is A, the acceleration amplitude a=A / (N / 2), where N is the number of sampling points in the fast Fourier transform in step S7-2;

[0031] S7-4: Based on each vibration frequency f and acceleration amplitude, obtain the vibration amplitude D = a / (2πf) 2 .

[0032] Furthermore, in the above method, in step S8, the amplitude of the low-frequency chatter is calculated based on sub-pixel matching, and the specific method is:

[0033] S8-1: Perform sub-pixel matching based on surface fitting on the flutter image g(x′, y′) and the non-flutter image f(x, y) to obtain a rough estimate of the amplitude;

[0034] S8-2: Based on the rough estimate of the amplitude, bilinear interpolation is performed on the vibrated image g(x′, y′) and the unvibrated image f(x, y), and then gradient-based sub-pixel matching is performed within the amplitude range to obtain the low-frequency vibrated frequency and amplitude.

[0035] Furthermore, in the above method, in step S8-1, a rough estimate of the amplitude is obtained based on sub-pixel matching of surface fitting, and the specific method is:

[0036] S8-1-1: Based on the integer pixel displacement search algorithm, find the position of the extreme value point of the correlation coefficient (x0, y0);

[0037] S8-1-2: Taking the extreme value point of the correlation coefficient (x0, y0) as the center, select the correlation coefficient matrix composed of the surrounding neighborhood points, perform binary quadratic surface fitting, and obtain the coordinates of the extreme value point (x, y). The rough estimate of the amplitude is Δx = x-x0 and Δy = y-y0.

[0038] Furthermore, in the above method, in step S8-2, the low-frequency dither frequency and amplitude are obtained based on the gradient sub-pixel matching, and the specific method is as follows:

[0039] S8-2-1: Assuming that the image subregion undergoes rigid body displacement before and after image deformation, establish the optimization function f(x,y)=g(x',y')=g(x+u+σ,y+v+β), where u and v are the integer parts of the rough amplitude estimate; σ and β are the decimal parts of the displacement before and after image deformation.

[0040] S8-2-2: Perform a first-order Taylor series expansion on the optimization function g(x',y')=g(x+u+σ,y+v+β), ignore high-order small quantities above order 2, and use the Barron operator to calculate the grayscale gradient;

[0041] S8-2-3: Based on the grayscale gradient, the least squares method is used to solve the sub-pixel displacements σ and β, and the amplitudes of the low-frequency vibrations in the two directions are obtained as (u+σ) and (u+β), respectively.

[0042] Furthermore, in the above method, in step S9, the method for calculating the high-frequency dither frequency and amplitude based on Radon transform is:

[0043] S9-1: The image is transformed into the frequency domain F(m,n) through Fourier transform, and histogram equalization and binarization are performed to obtain the processed image;

[0044] S9-2: Perform Radon transform in all directions on the processed image to obtain the transformation result;

[0045] S9-3: taking the maximum value among the transformed values as the main vibration direction θ and image frequencies m and n;

[0046] S9-4: Perform Radon transform again in the vibration direction θ to obtain the first kind of zero-order Bezier curve;

[0047] S9-5: Based on the zero point position μ of the first-order zero-order Bezier curve, obtain the amplitude of the high-frequency dither A = μ / (2π(mcosθ+nsinθ)), where μ is the zero point position of the first-order zero-order Bezier curve, θ is the main vibration direction, and m and n are the frequency values in the horizontal and vertical directions of the image, respectively.

[0048] The beneficial effects of the present invention and the prior art are:

[0049] (1) The present invention provides a simulation test system for the effects of satellite platform flutter on push-broom imaging of optical payloads. In this system, the optical payload is fixed while the target image moves along a linear guide rail, enabling push-broom imaging of the optical payload using a TDI-CMOS sensor. This system can simulate, on the ground, the effects of satellite platform flutter of varying amplitudes, ranging from 1 Hz to 2 kHz, on push-broom imaging of remote sensing optical payloads. This system provides support for flutter inversion and flutter blur removal for push-broom remote sensing satellites, significantly reducing costs.

[0050] (2) The present invention can perform multi-party verification of the vibration information: after the vibration signal frequency of the vibration signal generator is set, the excitation platform starts to vibrate, the target image moves with the linear guide rail, and the optical payload obtains the target image. At the same time, the accelerometer obtains the vibration information of the optical payload. The frequency and amplitude of the vibration are inverted from the image through the independently developed inversion algorithm, and can be mutually verified with the information obtained by the acceleration and the parameters set by the vibration signal generator, thereby providing a multi-party verification analysis for the simulation test of the vibration impact. The analysis results have very important guiding significance for the impact of satellite platform vibration on optical payload push-broom imaging.

[0051] (3) The present invention adopts a sub-pixel matching-based scheme and a Radon transform-based scheme to achieve accurate inversion of the amplitudes of low-frequency and high-frequency vibrations, respectively, supporting the verification of the satellite vibration characteristic inversion algorithm based on remote sensing images and the verification of the remote sensing satellite vibration image deblurring algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 2. It is a schematic structural diagram of a push-broom imaging simulation test system for remote sensing optical payloads affected by satellite platform flutter according to the present invention;

[0053] Figure 2 It is a schematic diagram of a local black and white checkerboard target image obtained by the remote sensing optical payload of the present invention during vibration. DETAILED DESCRIPTION

[0054] The present invention will be further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0055] like Figure 1 As shown, the present invention discloses a simulation test system for the effect of satellite platform flutter on push-broom imaging of remote sensing optical payload, comprising: a flutter signal generator 1, a vibration isolation platform 2, a vibration excitation platform 3, a zoom optical payload 4, an air-floating vibration isolation platform 5, a linear guide rail 6, a target image 7, an accelerometer 8, and an image processing and motion control computer 9; wherein,

[0056] The vibration excitation platform 3 is installed on the vibration isolation platform 2, and the vibration excitation platform 3 generates vibration under the signal excitation of the vibration signal generator 1 to simulate the vibration of the satellite platform;

[0057] The zoom optical payload 4 is mounted on the excitation platform 3 to image the target image 7 and send the imaging result to the image processing and motion control computer 9;

[0058] The accelerometer 8 is mounted on the zoom optical payload 4 to collect vibration information of the zoom optical payload 4 and send the information to the image processing and motion control computer 9;

[0059] The linear guide rail 6 is installed on the air-floating vibration isolation platform 5 to provide uniform linear motion in one dimension;

[0060] The image processing and motion control computer 9 analyzes and processes the imaging results and the vibration information of the accelerometer 8, and controls the linear guide rail 6 to perform uniform linear motion;

[0061] The target image 7 is mounted on the linear guide rail 6 to simulate target ground object information.

[0062] Preferably, the vibration platform 3 provides vibration in the vertical direction and the camera push-scan direction; the maximum output frequency of the vibration platform 3 is greater than or equal to 2 kHz.

[0063] Preferably, the zoom optical payload 4 adopts a TDI-CMOS sensor. The line of sight of the zoom optical payload 4 after installation is perpendicular to the direction of movement along the linear guide rail 6, and cooperates with the target image 7 moving in the same direction to simulate the on-orbit push-scan imaging of the remote sensing satellite; the target image 7 is a real remote sensing image or a black and white checkerboard image.

[0064] Preferably, the measurement range of the accelerometer 8 is -19.6 m / s 2 ~19.6m / s 2 The frequency measurement range is 1Hz~2kHz, and the measurement accuracy is greater than 1μm.

[0065] The present invention discloses a test method for a simulation test system for satellite platform flutter affecting remote sensing optical payload push-broom imaging, comprising:

[0066] S1: Move the target image 7 to the center of the linear guide 6, and align the lens of the zoom optical payload 4 with the target image 7;

[0067] S sets the focal length of the zoom optical payload 4 and the distance between the lens and the target image 7 on the linear guide 6, satisfying L1 / L2=L3 / L4; L1 is the ground simulation focal length, L2 is the ground simulation image distance, L3 is the actual satellite focal length, and L4 is the actual satellite image distance;

[0068] S3: Switch the CMOS sensor of the zoom optical payload 4 to the standard surface imaging mode and focus the zoom optical payload 4 so that the target image 7 is clearly imaged to obtain a non-flutter image f(x, y);

[0069] S4: Move the target image 7 to one end of the linear guide rail 6, and calculate the movement speed v of the linear guide rail 6 to satisfy Where p is the pixel size, f is the focal length of the optical payload, l is the imaging distance, and T is the integration time of the optical payload;

[0070] S5: Setting the vibration frequency and amplitude of the vibration signal generator 1, turning on the vibration platform 3, amplifying the vibration signal, and causing the zoom optical payload 4 to vibrate along with the vibration platform 3;

[0071] S6: The target image 7 moves along the linear guide rail 6 at the speed calculated in step S4, and the zoom optical payload 4 images the target image 7 during vibration to obtain a flutter image g(x′, y′);

[0072] S7: The accelerometer 8 collects vibration information of the zoom optical payload 4 and calculates the frequency and amplitude of the vibration using a frequency domain analysis method;

[0073] S8: Using the zoom optical payload 4 to image the target image 7, based on sub-pixel matching, calculate the low-frequency chatter frequency and amplitude, and subtract them from the vibration frequency and amplitude obtained by the accelerometer 8 to obtain the inverted low-frequency chatter frequency and amplitude error;

[0074] S9: Using the zoom optical payload 4 to image the target image 7, the high-frequency chatter frequency and amplitude are calculated based on Radon transform, and the frequency and amplitude of the high-frequency chatter are subtracted from the frequency and amplitude of the vibration obtained by the accelerometer 8 to obtain the error of the inverted high-frequency chatter frequency and amplitude;

[0075] S10: Obtain a test result of the impact of the flutter according to the inverted low-frequency flutter frequency and amplitude errors and the inverted high-frequency flutter frequency and amplitude errors.

[0076] Preferably, in step S7, the frequency domain analysis method is specifically:

[0077] S7-1: Filter the accelerometer signal during imaging using a 6th-order phase-shift-free Butterfly filter to obtain a filtered signal;

[0078] S7-2: Perform fast Fourier transform on the filtered signal and estimate the vibration frequency based on the spectrum peak;

[0079] S7-3: When the absolute value of the Fourier spectrum amplitude corresponding to each vibration frequency f is A, the acceleration amplitude a=A / (N / 2), where N is the number of sampling points in the fast Fourier transform in step S7-2;

[0080] S7-4: Based on each vibration frequency f and acceleration amplitude, obtain the vibration amplitude D = a / (2πf) 2 .

[0081] Preferably, in step S8, the amplitude of the low-frequency chatter is calculated based on sub-pixel matching, and the specific method is:

[0082] S8-1: Perform sub-pixel matching based on surface fitting on the flutter image g(x′, y′) and the non-flutter image f(x, y) to obtain a rough estimate of the amplitude;

[0083] S8-2: Based on the rough estimate of the amplitude, bilinear interpolation is performed on the vibrated image g(x′, y′) and the unvibrated image f(x, y), and then gradient-based sub-pixel matching is performed within the amplitude range to obtain the low-frequency vibrated frequency and amplitude.

[0084] Preferably, in step S8-1, a rough estimate of the amplitude is obtained based on sub-pixel matching of surface fitting, specifically by:

[0085] S8-1-1: Based on the integer pixel displacement search algorithm, find the position of the extreme value point of the correlation coefficient (x0, y0);

[0086] S8-1-2: Taking the extreme value point of the correlation coefficient (x0, y0) as the center, select the correlation coefficient matrix composed of the surrounding neighborhood points, perform binary quadratic surface fitting, and obtain the coordinates of the extreme value point (x, y). The rough estimate of the amplitude is Δx = x-x0 and Δy = y-y0.

[0087] Preferably, in step S8-2, the low-frequency dither frequency and amplitude are obtained by gradient-based sub-pixel matching, specifically by:

[0088] S8-2-1: Assuming that the image subregion undergoes rigid body displacement before and after image deformation, establish the optimization function f(x,y)=g(x',y')=g(x+u+σ,y+v+β), where u and v are the integer parts of the rough amplitude estimate; σ and β are the decimal parts of the displacement before and after image deformation.

[0089] S8-2-2: Perform a first-order Taylor series expansion on the optimization function g(x',y')=g(x+u+σ,y+v+β), ignore high-order small quantities above order 2, and use the Barron operator to calculate the grayscale gradient;

[0090] S8-2-3: Based on the grayscale gradient, the least squares method is used to solve the sub-pixel displacements σ and β, and the amplitudes of the low-frequency vibrations in the two directions are obtained as (u+σ) and (u+β), respectively.

[0091] Furthermore, in the above method, in step S9, the method for calculating the high-frequency dither frequency and amplitude based on Radon transform is:

[0092] S9-1: The image is transformed into the frequency domain F(m,n) through Fourier transform, and histogram equalization and binarization are performed to obtain the processed image;

[0093] S9-2: Perform Radon transform in all directions on the processed image to obtain the transformation result;

[0094] S9-3: taking the maximum value among the transformed values as the main vibration direction θ and image frequencies m and n;

[0095] S9-4: Perform Radon transform again in the vibration direction θ to obtain the first kind of zero-order Bezier curve;

[0096] S9-5: Based on the zero point position μ of the first-order zero-order Bezier curve, obtain the amplitude of the high-frequency dither A = μ / (2π(mcosθ+nsinθ)), where μ is the zero point position of the first-order zero-order Bezier curve, θ is the main vibration direction, and m and n are the frequency values in the horizontal and vertical directions of the image, respectively.

[0097] Example

[0098] The simulation test system for the effect of satellite platform flutter on optical payload push-broom imaging proposed in this embodiment will be further described in detail below with reference to the accompanying drawings and specific implementation plans.

[0099] like Figure 1 As shown, the simulation test system for the impact of satellite platform flutter on optical payload push-broom imaging in this embodiment includes: a flutter signal generator, a vibration isolation platform, an excitation platform, a zoom optical payload, an air-floating vibration isolation platform, a linear guide, a target image, an accelerometer, and an image processing and motion control computer. The excitation platform is mounted on the vibration isolation platform and, in response to signals from the flutter signal generator, generates flutter, providing flutter in the vertical and along-track directions (i.e., the camera push-broom direction) to simulate satellite platform flutter. The zoom optical payload uses a TDI-CMOS sensor, mounted on the excitation platform, with its time delay integration direction corresponding to the along-track motion of the linear guide. Images of the zoom optical payload are sent to the image processing and motion control computer for processing. The accelerometer, mounted on the zoom optical payload, works in conjunction with the image processing and motion control computer to measure the frequency and amplitude of the camera's vibration in all directions. The linear guide is mounted on the air-floating vibration isolation platform to provide uniform linear motion in one dimension. The target image is mounted on a linear guide rail to simulate the target ground object information, and is combined with a zoom optical payload to simulate satellite push-broom imaging. In this example, the chessboard target uses a black and white chessboard image with a length and width of 400mm and 280mm respectively, and the side length of the square black and white grid is 40mm. Figure 2 The maximum output frequency range of the vibration platform is 2kHz. The accelerometer measurement range is ±19.6m / s. 2 , the frequency can be measured up to 2kHz, and the measurement accuracy is 1μm.

[0100] The specific operation process of the present invention includes the following steps:

[0101] Step S1: Move the target image to the center of the guide rail and align the lens of the zoom optical payload with the target image;

[0102] Step S2: According to the equivalent model, ground simulation focal length / ground simulation image distance = actual satellite focal length / actual satellite image distance, set the focal length of the optical payload and the distance between the lens and the target image on the guide rail;

[0103] Step S3: Switching the CMOS sensor of the optical payload to a surface imaging mode and focusing the sensor so that the target image can be clearly imaged;

[0104] Step S4: Calculate the movement speed of the guide rail according to the speed-to-height ratio, and move the target image to one end of the guide rail;

[0105] Step S5: Setting the vibration frequency and amplitude of the vibration signal generator, turning on the vibration platform, amplifying the vibration signal, and causing the camera to vibrate along with the platform;

[0106] Step S6: The target image moves along the guide rail at the speed set in step S4, and the camera images the moving target during vibration;

[0107] Step S7: The accelerometer collects vibration information of the optical load and uses the frequency domain analysis method to invert the frequency and amplitude of the vibration;

[0108] The method for calculating the motion speed v of the guide rail using the speed-to-height ratio is specifically as follows:

[0109]

[0110] Where p is the pixel size, f is the focal length of the optical payload, l is the imaging distance, and T is the integration time of the optical payload.

[0111] The frequency domain analysis method is specifically as follows:

[0112] First, the accelerometer signal during imaging is filtered with a 6th-order phase-shift-free Butterfly filter. Then, the filtered signal is subjected to a fast Fourier transform, and the vibration frequency is estimated based on the spectrum peak. Then, for each vibration frequency f, its acceleration amplitude is calculated:

[0113] a=A / (N / 2)

[0114] Where A is the absolute value of the corresponding Fourier spectrum amplitude, and N is the number of sampling points during fast Fourier transform;

[0115] For each vibration frequency f, the amplitude of the vibration is:

[0116] D=a / (2πf) 2

[0117] Step S8: Using the optical payload to image the target, the amplitude of the low-frequency vibration is calculated based on sub-pixel matching, and compared with the vibration information obtained by the accelerometer to test the impact of the vibration.

[0118] The amplitude calculation method based on sub-pixel matching is specifically as follows:

[0119] First, a sub-pixel match is performed on the flutter image g(x, y) and the unflutter image f(x, y) based on binary quadratic surface fitting to obtain a rough estimate of the amplitude: First, the position of the correlation coefficient extreme point (x0, y0) is found using an integer pixel shift search algorithm. Then, with (x0, y0) as the center, a binary quadratic surface fitting is performed on the correlation coefficient matrix consisting of points in the surrounding neighborhood (3×3 pixels). The coordinates of the extreme point (x, y) are obtained, and the amplitude is Δx = x-x0 and Δy = y-y0.

[0120] Then, bilinear interpolation is performed on the tremor image g(x′, y′) and the untremor image f(x, y), and gradient-based sub-pixel matching is performed within the amplitude range to estimate the final amplitude:

[0121] First, assume that the image subregion is subjected to rigid body displacement before and after image deformation, and establish the optimization function:

[0122] f(x,y)=g(x′,y′)=g(x+u+Δx,y+v+Δy)

[0123] Where u, v, Δx, and Δy are the pixel-level and sub-pixel-level displacements in the x and y directions before and after the deformation, respectively, and u and v are the integer parts of the coarse amplitude estimates obtained by sub-pixel matching based on binary quadratic surface fitting. Then, a first-order Taylor series expansion is performed on g(x′, y′), ignoring high-order small quantities:

[0124] g(x′,y′)=g(x+u,y+v)+Δxg x (x+u,y+v)+Δyg y (x+u,y+v)

[0125] The Barron operator is used to calculate the grayscale gradient:

[0126]

[0127] Finally, the least squares method is used to solve the sub-pixel displacement Δx and Δy, and the final amplitude estimates in the two directions are obtained as (u+Δx) and (v+Δy):

[0128]

[0129] Step S9: Using the optical payload to image the target, the amplitude of the high-frequency vibration is calculated based on Radon transform, and compared with the vibration information obtained by the accelerometer to test the impact of the vibration.

[0130] The method for calculating high-frequency chatter frequency and amplitude based on Radon transform is as follows:

[0131] First, the image is transformed into the frequency domain F(u, v) through Fourier transform, and then histogram equalization and binarization are performed. The processed image is then subjected to a Radon transform in all directions. The maximum value of the transform corresponds to the main vibration direction θ and the image frequencies u and v. Then, another Radon transform is performed in the main vibration direction θ to obtain a fluctuation curve, namely the first-class zero-order Bezier curve. The amplitude of the vibration is calculated based on its zero point position μ:

[0132]

[0133] In summary, the present invention utilizes a fixed TDI-CMOS optical payload combined with a target image moving along a linear guide to simulate optical payload push-broom imaging. The present invention's dither signal generator, the accelerometer mounted on the excitation platform's optical payload, and the optical payload push-broom imaging all contribute to platform vibration inversion. This allows for low-cost ground-based simulation of the effects of satellite platform dither of varying amplitudes and frequencies on push-broom imaging of remote sensing optical payloads, providing support for dither inversion and dither blur removal for push-broom remote sensing satellites.

[0134] It should be noted that the drawings are in a very simplified form and are not in precise proportions, and are only used to conveniently and clearly assist in explaining the purpose of the embodiments of the present invention. In order to make the purposes, features and advantages of the present invention more obvious and easy to understand, please refer to the drawings. It should be noted that the structures, proportions, sizes, etc. illustrated in the drawings of this specification are only used to match the contents disclosed in the specification for people familiar with this technology to understand and read, and are not used to limit the limiting conditions for the implementation of the present invention. Therefore, they have no technical significance. Any modification of the structure, change in the proportional relationship or adjustment of the size should still fall within the scope of the technical content disclosed in the present invention without affecting the efficacy and purpose that can be achieved by the present invention.

[0135] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0136] It should be noted that the devices and methods disclosed in the embodiments of this document may also be implemented in other ways. The device embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of the devices, methods, and computer program products according to the various embodiments of this document. In this regard, each box in the flowchart or block diagram may represent a module, program, or portion of code, wherein the module, program segment, or portion of code contains one or more executable instructions for implementing a specified logical function, and the module, program segment, or portion of code contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the boxes may also occur in an order different from that marked in the accompanying drawings. For example, two consecutive boxes may actually be executed substantially in parallel, or they may sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart, may be implemented by a dedicated hardware-based system for performing the specified function or action, or may be implemented by a combination of dedicated hardware and computer instructions.

[0137] Although the present invention has been described in detail through the above preferred embodiments, it should be understood that the above description is not intended to limit the present invention. After reading the above description, various modifications and substitutions of the present invention will become apparent to those skilled in the art. Therefore, the scope of protection of the present invention should be defined by the appended claims.

[0138] The contents not described in detail in the specification of the present invention belong to the common knowledge of professionals in this field.

Claims

1. A simulation test system for the effect of satellite platform flutter on push-broom imaging of remote sensing optical payloads, characterized in that: include: A vibration signal generator (1), a vibration isolation platform (2), an excitation platform (3), a zoom optical load (4), an air-floating vibration isolation platform (5), a linear guide rail (6), a target image (7), an accelerometer (8), and an image processing and motion control computer (9); wherein, The vibration excitation platform (3) is installed on the vibration isolation platform (2), and the vibration excitation platform (3) generates vibration under the signal excitation of the vibration signal generator (1) to simulate the vibration of the satellite platform; The zoom optical load (4) is mounted on the excitation platform (3), images the target image (7), and sends the imaging result to the image processing and motion control computer (9); The accelerometer (8) is mounted on the zoom optical payload (4), collects vibration information of the zoom optical payload (4), and sends the information to the image processing and motion control computer (9); The linear guide rail (6) is installed on the air-floating vibration isolation platform (5) to provide uniform linear motion in one-dimensional direction; The image processing and motion control computer (9) analyzes and processes the imaging results and the vibration information of the accelerometer (8), and controls the linear guide rail (6) to perform uniform linear motion; The target image (7) is mounted on the linear guide rail (6) to simulate target ground object information.

2. The simulation test system for satellite platform flutter affecting remote sensing optical payload push-broom imaging according to claim 1, characterized in that: The excitation platform (3) provides vibration in the vertical direction and the camera push-scan direction; the maximum output frequency of the excitation platform (3) is greater than or equal to 2 kHz.

3. The simulation test system for satellite platform flutter affecting remote sensing optical payload push-broom imaging according to claim 1, characterized in that: The zoom optical payload (4) adopts a TDI-CMOS sensor. The sight direction of the zoom optical payload (4) after installation is perpendicular to the along-track movement direction of the linear guide rail (6), and cooperates with the target image (7) moving in the same direction to simulate on-orbit push-scan imaging of a remote sensing satellite; the target image (7) is a real remote sensing image or a black and white checkerboard image.

4. The simulation test system for satellite platform flutter affecting remote sensing optical payload push-broom imaging according to claim 1, characterized in that: The measuring range of the accelerometer (8) is -19.6 m / s 2 ~19.6m / s 2 The frequency measurement range is 1Hz~2kHz, and the measurement accuracy is greater than 1μm.

5. A test method for a simulation test system for the effect of satellite platform flutter on push-broom imaging of remote sensing optical payloads according to any one of claims 1 to 4, characterized in that: include: S1: Move the target image (7) to the center of the linear guide rail (6), and align the lens of the zoom optical payload (4) with the target image (7); S2: Set the focal length of the zoom optical payload (4) and the distance between the lens and the target image (7) on the linear guide (6) to satisfy L1 / L2=L3 / L4; L1 is the ground simulation focal length, L2 is the ground simulation image distance, L3 is the actual satellite focal length, and L4 is the actual satellite image distance; S3: Switch the CMOS sensor of the zoom optical payload (4) to a standard surface imaging mode, perform focusing, and make the zoom optical payload (4) clearly image the target image (7), and obtain a non-flutter image f(x, y); S4: Move the target image (7) to one end of the linear guide rail (6), and calculate the motion speed ν of the linear guide rail (6) to satisfy v≤(p / f)×(l / T), where p is the pixel size, f is the focal length of the optical payload, l is the imaging distance, and T is the integration time of the optical payload; S5: setting the vibration frequency and amplitude of the vibration signal generator (1), turning on the excitation platform (3), amplifying the vibration signal, and causing the zoom optical load (4) to vibrate along with the excitation platform (3); S6: The target image (7) moves along the linear guide rail (6) at the speed calculated in step S4, and the zoom optical load (4) images the target image (7) during vibration to obtain a flutter image g(x′, y′); S7: The accelerometer (8) collects vibration information of the zoom optical load (4) and calculates the frequency and amplitude of the vibration using a frequency domain analysis method; S8: Using the zoom optical payload (4) to image the target image (7), based on sub-pixel matching, calculate the low-frequency vibration frequency and amplitude, and make a difference between the frequency and amplitude of the vibration obtained by the accelerometer (8) to obtain the error of the inverted low-frequency vibration frequency and amplitude; S9: Using the zoom optical load (4) to image the target image (7), based on Radon transform, calculate the high-frequency flutter frequency and amplitude, and make a difference between the frequency and amplitude of the vibration obtained by the accelerometer (8) to obtain the error of the inverted high-frequency flutter frequency and amplitude; S10: Obtain a test result of the impact of the flutter according to the inverted low-frequency flutter frequency and amplitude errors and the inverted high-frequency flutter frequency and amplitude errors.

6. The testing method according to claim 5, characterized in that: In step S7, the frequency domain analysis method is specifically: S7-1: Filter the accelerometer signal during imaging using a 6th-order phase-shift-free Butterfly filter to obtain a filtered signal; S7-2: Perform fast Fourier transform on the filtered signal and estimate the vibration frequency based on the spectrum peak; S7-3: When the absolute value of the Fourier spectrum amplitude corresponding to each vibration frequency f is A, the acceleration amplitude a=A / (N / 2), where N is the number of sampling points in the fast Fourier transform in step S7-2; S7-4: Based on each vibration frequency f and acceleration amplitude, obtain the vibration amplitude D = a / (2πf) 2 .

7. The testing method according to claim 5, characterized in that: In step S8, the amplitude of the low-frequency vibration is calculated based on sub-pixel matching. The specific method is: S8-1: Perform sub-pixel matching based on surface fitting on the flutter image g(x′, y′) and the non-flutter image f(x, y) to obtain a rough estimate of the amplitude; S8-2: Based on the rough estimate of the amplitude, bilinear interpolation is performed on the vibrated image g(x′, y′) and the unvibrated image f(x, y), and then gradient-based sub-pixel matching is performed within the amplitude range to obtain the low-frequency vibrated frequency and amplitude.

8. The testing method according to claim 7, characterized in that: In step S8-1, a rough estimate of the amplitude is obtained based on sub-pixel matching of surface fitting. The specific method is as follows: S8-1-1: Based on the integer pixel displacement search algorithm, find the position of the extreme value point of the correlation coefficient (x0, y0); S8-1-2: Taking the extreme value point of the correlation coefficient (x0, y0) as the center, select the correlation coefficient matrix composed of the surrounding neighborhood points, perform binary quadratic surface fitting, and obtain the coordinates of the extreme value point (x, y). The rough estimate of the amplitude is Δx = x-x0 and Δy = y-y0.

9. The testing method according to claim 8, characterized in that: In step S8-2, the low-frequency dither frequency and amplitude are obtained based on the gradient-based sub-pixel matching. The specific method is as follows: S8-2-1: Assuming that the image subregion undergoes rigid body displacement before and after image deformation, establish the optimization function f(x,y)=g(x',y')=g(x+u+σ,y+v+β), where u and v are the integer parts of the rough amplitude estimate; σ and β are the decimal parts of the displacement before and after image deformation. S8-2-2: Perform a first-order Taylor series expansion on the optimization function g(x',y')=g(x+u+σ,y+v+β), ignore high-order small quantities above order 2, and use the Barron operator to calculate the grayscale gradient; S8-2-3: Based on the grayscale gradient, the least squares method is used to solve the sub-pixel displacements σ and β, and the amplitudes of the low-frequency vibrations in the two directions are obtained as (u+σ) and (u+β), respectively.

10. The testing method according to claim 5, characterized in that: In step S9, the method for calculating the high-frequency dither frequency and amplitude based on Radon transform is: S9-1: The image is transformed into the frequency domain F(m,n) through Fourier transform, and histogram equalization and binarization are performed to obtain the processed image; S9-2: Perform Radon transform in all directions on the processed image to obtain the transformation result; S9-3: taking the maximum value among the transformed values as the main vibration direction θ and image frequencies m and n; S9-4: Perform Radon transform again in the vibration direction θ to obtain the first kind of zero-order Bezier curve; S9-5: Based on the zero point position μ of the first-order zero-order Bezier curve, obtain the amplitude of the high-frequency dither A = μ / (2π(mcosθ+nsinθ)), where μ is the zero point position of the first-order zero-order Bezier curve, θ is the main vibration direction, and m and n are the frequency values in the horizontal and vertical directions of the image, respectively.

Images