Computational Imaging Through Atmospheric Turbulence

TL;DR

This paper reviews the progress and challenges in computational imaging through atmospheric turbulence, emphasizing the role of deep learning in improving image reconstruction accuracy and speed.

Contribution

It highlights recent advancements in deep learning methods for turbulence mitigation and discusses future directions for faster, more accurate computational models.

Findings

Deep learning enhances image reconstruction through turbulence.

New models improve speed and accuracy of imaging.

The field is rapidly evolving with technological advances.

Abstract

Since the seminal work of Andrey Kolmogorov in the early 1940's, imaging through atmospheric turbulence has grown from a pure scientific pursuit to an important subject across a multitude of civilian, space-mission, and national security applications. Fueled by the recent advancement of deep learning, the field is further experiencing a new wave of momentum. However, for these deep learning methods to perform well, new efforts are needed to build faster and more accurate computational models while at the same time maximizing the performance of image reconstruction. The book is written primarily for image processing engineers, computer vision scientists, and engineering students who are interested in the field of atmospheric turbulence, statistical optics, and image processing. The book can be used as a graduate text, or advanced topic classes for undergraduates.