Atmospheric turbulence restoration by diffeomorphic image registration and blind deconvolution

TL;DR

This paper introduces two innovative algorithms combining blind deconvolution, elastic registration, and temporal filtering to enhance images degraded by atmospheric turbulence, validated on real desert images.

Contribution

The paper proposes new algorithms that integrate multiple techniques for improved atmospheric turbulence image restoration, demonstrating their effectiveness on real-world data.

Findings

Effective image restoration in atmospheric turbulence conditions

Successful application on real desert images

Improved image clarity and detail recovery

Abstract

A novel approach is presented in this paper to improve images which are altered by atmospheric turbulence. Two new algorithms are presented based on two combinations of a blind deconvolution block, an elastic registration block and a temporal filter block. The algorithms are tested on real images acquired in the desert in New Mexico by the NATO RTG40 group.