CN-114492240-B - Infrared pneumatic optical imaging simulation method, storage medium and computer equipment

CN114492240BCN 114492240 BCN114492240 BCN 114492240BCN-114492240-B

Abstract

The invention belongs to the technical field of optical imaging, and discloses an infrared pneumatic optical imaging simulation method, a storage medium and computer equipment, wherein wave front change values under different flight states are obtained through flow field dynamics and optical numerical simulation, a mapping relation between the wave front change values and flight state parameters is established, and a continuously-changing degradation model is constructed; the method comprises the steps of constructing an infrared scene simulation model, outputting a sequence infrared image of a detection stage and an imaging flight state corresponding to the sequence infrared image, inputting the imaging flight state into a continuously-changing Gaussian superposition model, and carrying out convolution operation on an obtained continuously-changing point spread function and a corresponding infrared image to realize real-time infrared pneumatic optical effect degradation imaging. The method establishes the pneumatic continuous degradation model coupled with the flight state, establishes the infrared detection visual simulation model, and can truly simulate the infrared detection imaging process in the high-speed motion state.

Inventors

RAO PENG
ZHANG SHUYUAN
CHEN XIN

Assignees

中国科学院上海技术物理研究所
中国科学院上海技术物理研究所

Dates

Publication Date: 20260421
Application Date: 20220125
Priority Date: 20220125

Claims (6)

1. The infrared pneumatic optical imaging simulation method is characterized by comprising the following steps of: Step one, obtaining a wavefront variation value sigma under different flight states through flow field dynamics and optical numerical simulation; Step two, establishing a mapping relation between a wavefront variation value sigma and flight state parameters, and constructing a continuously-varying degradation model; step three, constructing an infrared scene simulation model, and outputting a sequence infrared image of a detection stage and a corresponding imaging flight state; inputting the imaging flight state into a continuously-changing Gaussian superposition model to obtain a group of continuously-changing point spread functions, and performing convolution operation with a corresponding infrared image to realize real-time infrared pneumatic optical effect degradation imaging; the flow field dynamics numerical simulation method in the first step adopts a Reynolds average Navier-Stokes equation, a direct numerical simulation DNS and a large vortex simulation LES; The optical numerical simulation in the first step adopts geometrical optics, physical optics and Fourier optics to analyze and calculate the distortion of the light after passing through the flow field; The method for constructing the mapping relation in the second step specifically comprises the following steps: (1) Preparing a data set; (2) Fitting by training set, which can be polynomial fitting, gaussian fitting, linear regression based on least squares, nonlinear regression; (3) Performing regression diagnosis, finding out and removing abnormal points by looking up the residual image, and then performing regression analysis again; (4) Performing model inspection, and calculating an evaluation index value of each regression model; The preparation of the data set of step (1) comprises: Acquiring wavefront variation parameters sigma under N groups of flight state parameters through flow field simulation and optical simulation, randomly sequencing the flight state parameters and the corresponding wavefront variation parameters sigma, and dividing the flight state parameters and the corresponding wavefront variation parameters sigma into a training set and a testing set according to a proportion; and drawing a scatter diagram of the sigma value and each flight state parameter, carrying out correlation coefficient analysis and significance test to obtain correlation coefficients and confidence coefficients of the dependent variable sigma and each flight state parameter independent variable, judging correlation relations among the variables, and eliminating abnormal values.
2. The infrared pneumatic optical imaging simulation method as set forth in claim 1, wherein in the regression analysis performed in step (2), fitting is performed by using a training set, and the fitting method is polynomial fitting, gaussian fitting, linear regression or nonlinear regression based on least squares; the step (4) is to perform model inspection, and calculate the evaluation index value of each regression model, when the evaluation index value of the regression model exceeds the index threshold value, the corresponding regression model is used as a mapping formula of sigma by combining the prediction precision of the test set; The evaluation index value of the regression model in the step (4) comprises a fitting goodness R-square, a regression coefficient significance test index, a regression equation significance test index or a correlation coefficient significance test index.
3. The infrared pneumatic optical imaging simulation method as set forth in claim 1, wherein the degradation model and the parameter setting thereof in the second step are specifically as follows: ; M=μ·Ma/H where ωi is the weight parameter, σ is the intensity parameter, xm and ym are the offset control parameters, and M is the number of turbulence units.
4. The infrared pneumatic optical imaging simulation method as set forth in claim 1, wherein the method for constructing the infrared scene simulation model in the third step comprises: Firstly, three-dimensional modeling and surface temperature field setting are carried out, a three-dimensional geometric model of a target and a background is established by utilizing three-dimensional modeling software, and the surface temperature of each part of a scene is set according to an empirical value; then, according to the received flight state parameters, the field of view parameters and imaging parameters of the sensor, determining an imaging observation area and the flight state corresponding to each imaging frame; Then calculating the infrared radiation quantity, taking the whole scene as a gray body, and calculating the infrared radiation intensity by using the Planck law to obtain an infrared radiation characteristic image under observation; and finally, calculating a voltage value of the infrared radiation intensity after photoelectric conversion, and quantifying the output voltage into a gray value.
5. A computer device, characterized in that the computer device comprises a memory and a processor, the memory storing a computer program, which, when executed by the processor, causes the processor to perform the steps of the infrared aerodynamic optical imaging simulation method according to any one of claims 1-4.
6. A computer readable storage medium storing a computer program which, when executed by a processor, causes the processor to perform the steps of the infrared pneumatic optical imaging simulation method of any one of claims 1 to 4.

Description

Infrared pneumatic optical imaging simulation method, storage medium and computer equipment Technical Field The invention belongs to the technical field of optical imaging, and particularly relates to an infrared pneumatic optical imaging simulation method, a storage medium and computer equipment. Background Currently, on-board infrared optical systems play an important role in remote sensing. With the development of aerospace technology, image distortion caused by aerodynamic optical effects has become a serious problem in airborne infrared optical imaging systems. The large density gradient of turbulent compressible flow around the optical window is a direct cause of aerodynamic optical distortion. Flow fields outside the optical window may include complex flow structures such as free shear layers, expansion waves, turbulent boundary layers, shock waves. The density gradient profile is unstable and constantly changing. According to the Gladstone-Dale relationship, the air refractive index is proportional to its density. Due to the variation of the refractive index, the beam passing through the aerodynamic flow presents serious wavefront distortion, leading to blurring, light deflection and jitter. These aviation optical effects adversely affect the imaging quality of the on-board optical system. Therefore, in order to improve the quality and accuracy of the on-board infrared remote sensing task, it is necessary to study the imaging changes caused by high-speed flight conditions. At present, a common method for estimating imaging influence caused by aerodynamic optical transmission effect is to study based on numerical calculation of flow field and optical transmission, and the method uses Computational Fluid Dynamics (CFD) software, so that the calculation amount is large, and only analysis can be performed for limited flight states. Meanwhile, due to the fact that the cost is high, special cameras and optical elements are needed for infrared imaging, the wind tunnel test and the flight test are limited in use, and a large amount of continuous infrared pneumatic degradation image data cannot be obtained. While the simulation model established based on phenomenology at present can simulate the degradation phenomenon caused by aerodynamic optical effects such as image blurring, shaking, line-of-sight errors, saturation and the like, typical parameters of the model still need to be independently set according to experience values, so the model is only suitable for describing the image quality degradation law and cannot be simulated in real time according to the flight state. Therefore, in order to effectively and inexpensively study the aerodynamic optics and its real-time correction, it is necessary to design a simulation method of the infrared aerodynamic optical effect that can simulate the real-time scene simulation. Through the above analysis, the problems and defects existing in the prior art are as follows: (1) The existing method for estimating the imaging influence caused by the pneumatic optical transmission effect has large calculated amount, can only analyze the limited flight state, has higher cost, is limited in the use of wind tunnel tests and flight tests, and cannot obtain a large amount of continuous infrared pneumatic degradation image data. (2) The existing simulation model is only suitable for describing the image quality degradation law, and cannot simulate in real time according to the flight state. The difficulty of solving the problems and the defects is as follows: (1) The wind tunnel experiment can not simulate the real flight environment, and is only suitable for simulating the influence of a single environmental factor; (2) It is necessary to consider how to combine the results of a small number of numerical simulations, to construct a model that can be continuously varied, So as to meet the change trend of the aerodynamic mechanism under different influencing factors. The meaning of solving the problems and the defects is as follows: (1) Because of cost and effect limitations, constructing a dynamic infrared field Jing Chengxiang simulation is a simple method for verifying a correction algorithm; (2) The fuzzy model under different parameters can be used as priori information to guide more accurate and more efficient correction algorithm research; (3) The degradation simulation model of different flight states can be simulated, and a large amount of data can be provided for training advanced deep learning algorithm. Disclosure of Invention Aiming at the problems existing in the prior art, the invention provides an infrared pneumatic optical imaging simulation method, a storage medium and computer equipment. The invention is realized in such a way that the infrared pneumatic optical imaging simulation method comprises the following steps: obtaining a wave front change value sigma under different flight states through flow field dynamics and optical numerical simulation, obtaining