Atmospheric turbulence is an irregular form of motion in the atmosphere. Because of turbulence interference,when theoptical system through the atmosphere of the target imaging, the observed image will appear point intensity diffusion,image blur, image drift and other turbulence effects. Digital recovery of the turbulence-degraded images technique is aclassical ill-conditioned problem by removing the blurring effect and suppressing the noise. Traditional approaches relyingon image heuristics suffer from high frequency noise amplification and processing artifacts. In this paper, the imagedegradation models of the turbulent flow are given, the point spread function of turbulence is simulated by the similarGaussian function model, and investigated a general framework of neural networks for restoring turbulence-degradedimages. The blur and additive noise are considered simultaneously. Two solutions respectively exploiting fullyconvolutional networks (FCN) and conditional Generative Adversarial Networks (CGAN) are presented. The FCN basedon minimizing the mean squared reconstruction error (MSE) in pixel space gets high PSNR. On the other side, the CGANbased on perceptual loss optimization criterion retrieves more textures. We conduct comparison experiments todemonstrate the performance at different degree of turbulence intensity from the training configuration. The results indicatethat the proposed networks outperform traditional approaches for restoring high frequency details and suppressing noiseeffectively.
展开▼
