Restoration method for reconstructing turbulence degraded image based on high and low frequency synthesis

The invention discloses a restoration method for reconstructing a turbulence degraded image based on high and low frequency synthesis. The method comprises a turbulence-caused image degradation analysis module, a neural network structure design module, a high and low frequency information separation module and a neural network turbulence removal module. According to the turbulence-caused image quality degradation module, after an optical imaging system is modeled, a Zernike polynomial under different wavefront turbulence intensities (D/r0) is set to obtain wavefront phase distribution of turbulence and a degraded image under the turbulence; the neural network structure design module adopts a double-U-Net network structure; the high-low frequency information separation module is used for separating high-frequency information and low-frequency information of the image through a high-pass filter; and the neural network turbulence removal module realizes the recovery effect of removing turbulence after multiple r