Restoration of atmospheric turbulence-degraded images based on multi-frame information

Atmospheric turbulence introduces random disturbances that degrade and distort images of observed targets as light propagates through the atmosphere. Although numerous algorithms have been developed to restore images degraded by turbulence, most of these algorithms lack sufficient generalization and are limited to specific application scenarios or fixed atmospheric turbulence intensities. In this paper, we propose an Atmospheric Turbulence Restoration Network (ATRN), a two-stage algorithm based on multi-frame information fusion. The first stage employs Transformers to comprehensively mitigate atmospheric turbulence, while the second stage performs multi-frame information fusion. Our approach effectively captures and processes both temporal and spatial information, extracting potentially clear image details from neighboring frames across multiple scenes and integrating this information to enhance restoration. The experimental results demonstrate that our algorithm has strong generalization capabilities across various atmospheric turbulence conditions, outperforming existing methods in restoring atmospheric turbulence-degraded images.