Multi-frame self-adaption optical image high resolution restoration method using wave front data

Abstract

The invention relates to a multiframe self-adaptive optical image high resolution restoration method by utilization of wavefront data, comprising the following steps: firstly, a multiframe short exposure degraded image gm (x, y) and a corresponding slope signal of a wavefront sensor are recorded continuously; secondly, the slope signal is restored into a point spread function hm (x, y) by utilization of wavefront residuals, and the function is taken as a guess starting point; thirdly, an estimated value f(x, y) of a real image is restored by utilization of the short exposure degraded image gm (x, y) and the point spread function hm (x, y), and an estimated value f(x, y) is obtained after addition of positivity limitation on the estimated value f(x, y); fourthly, an estimated value hm (x, y) of the point spread function is obtained by utilization of the short exposure degraded image gm (x, y) and the estimated value f(x, y), and an estimated value h(x, y) is obtained after addition of positivity limitation in the same way; fifthly, iteration stopping conditions are defined, if the iteration stopping conditions are not met, the third step is returned until convergence. The invention has the advantages of quick convergence speed, capability of being adaptive to different degraded images obtained under the condition of wavefront disturbances with different types and degrees, effective compensation of correction capability under hardware limitation of a self-adaptive optical system, and improvement of imaging quality.

Description

Technical field [0001] The present invention belongs to image processing technology, and proposes an image restoration method for an adaptive optical system. Background technique [0002] Image restoration is the estimation process of obtaining the real image from the observed degraded image. The imaging model can be expressed as: [0003] g(x,y)=f(x,y)h(x,y)+n(x,y) (1) [0004] In the formula, g(x, y) represents the observed degraded image; f(x, y) represents the real target, which is the target to be restored; h(x, y) represents the point spread function of the optical system, which is used to describe the wave Front phase difference; n(x, y) represents system noise; "" represents cyclic convolution. Due to factors such as insufficient freedom of the deformable mirror and inaccurate measurement of the wavefront sensor, the adaptive optics system often cannot achieve complete correction of the degraded image. Therefore, the partially corrected image results need to be restored...

AI Technical Summary

Problems solved by technology

[0005] Ayers et al. proposed an iterative blind image restoration method, which uses single-frame degraded images and positive constraints to restore the real target, but it can only deal with the case where the point spread function is small, and the convergence cannot be guaranteed; Conan et al. proposed a A Restoration Method for f(x, y) and h(x, y) Using the Zernike Multinomial Distribution Law under the Kolmogorov Atmospheric Turbulence Model

However, in actual observations, the atmospheric turbulence does not fit the Kolmogorov model well, so the method has poor adaptability

Images