Image Reconstruction of Static and Dynamic Scenes Through Anisoplanatic Turbulence

@article{Mao2020ImageRO,
  title={Image Reconstruction of Static and Dynamic Scenes Through Anisoplanatic Turbulence},
  author={Zhiyuan Mao and Nicholas Chimitt and Stanley H. Chan},
  journal={IEEE Transactions on Computational Imaging},
  year={2020},
  volume={6},
  pages={1415-1428}
}
Ground based long-range passive imaging systems often suffer from degraded image quality due to a turbulent atmosphere. While methods exist for removing such turbulent distortions, many are limited to static sequences which cannot be extended to dynamic scenes. In addition, the physics of the turbulence is often not integrated into the image reconstruction algorithms, making the physics foundations of the methods weak. In this article, we present a unified method for atmospheric turbulence… 
Characterizing Atmospheric Turbulence and Removing Distortion in Long-range Imaging by Cameron Whyte A Thesis Presented in Partial Fulfillment of the Requirement for the Degree Master of Arts Approved April 2021 by the Graduate Supervisory Committee: Malena Espanol, Co-Chair
Atmospheric turbulence distorts the path of light passing through the air. When capturing images at long range, the effects of this turbulence can cause substantial geometric distortion and blur in
Accelerating Atmospheric Turbulence Simulation via Learned Phase-to-Space Transform
TLDR
A new concept known as the phase-to-space (P2S) transform is introduced to significantly speed up the simulation of imaging through atmospheric turbulence, and offers 300× – 1000× speed up compared to the mainstream split-step simulators while preserving the essential turbulence statistics.
Unsupervised Non-Rigid Image Distortion Removal via Grid Deformation
TLDR
A novel unsupervised network is presented to recover the latent distortion-free image by model non-rigid distortions as deformable grids and leveraging the positional encoding operator to simplify the network structure while maintaining fine spatial details in the recovered images.
Subspace modeling for fast and high-sensitivity X-ray chemical imaging
TLDR
This work introduces a simple and robust denoising approach to improve the image quality, which enables fast and high-sensitivity chemical imaging.
A New Blind Video Quality Metric for Assessing Different Turbulence Mitigation Algorithms
TLDR
Experimental results show that the proposed metrics correlate well with human subjective evaluations, and are compared with an existing blind video metric and two other blind image quality metrics, which performed consistently better.
TurbuGAN: An Adversarial Learning Approach to Spatially-Varying Multiframe Blind Deconvolution with Applications to Imaging Through Turbulence
—We present a self-supervised and self-calibrating multi-shot approach to imaging through atmospheric turbulence, called TurbuGAN. Our approach requires no paired training data, adapts itself to the

References

SHOWING 1-10 OF 51 REFERENCES
Deep learning for software-based turbulence mitigation in long-range imaging
TLDR
A turbulence mitigation algorithm which employs a deep neural network to recover high spatial frequency information from turbulence degraded video and shows that its deep learning based approach provides on semi-synthetic test data a substantial performance increase compared to the previous sharpening approach.
Restoration of Atmospheric Turbulence-distorted Images via RPCA and Quasiconformal Maps
TLDR
A novel algorithm is proposed in this paper to reduce geometric distortion as well as space-and-time-varying blur due to strong turbulence, effective in alleviating distortions and blur, restoring image details and enhancing visual quality.
Block matching and Wiener filtering approach to optical turbulence mitigation and its application to simulated and real imagery with quantitative error analysis
TLDR
The proposed block-matching and Wiener filtering approach to atmospheric turbulence mitigation for long-range imaging of extended scenes has excellent performance in comparison with state-of-the-art benchmark methods in this study.
Removing Atmospheric Turbulence via Space-Invariant Deconvolution
  • Xiang Zhu, P. Milanfar
  • Mathematics
    IEEE Transactions on Pattern Analysis and Machine Intelligence
  • 2013
TLDR
An approach capable of restoring a single high-quality image from a given image sequence distorted by atmospheric turbulence is proposed, which reduces the space and time-varying deblurring problem to a shift invariant one.
Atmospheric turbulence mitigation based on turbulence extraction
TLDR
This approach manages to extract information of distortion position from the sparse turbulent component to indicate the sharpest turbulence patches, and is capable of significantly alleviating atmospheric turbulence blur and geometric distortion.
Atmospheric Turbulence Mitigation Using Complex Wavelet-Based Fusion
TLDR
A simple and efficient frame selection method is proposed to select informative ROIs only from good-quality frames, and a learning-based metric specifically for image quality assessment in the presence of atmospheric distortion is proposed.
Geometric correction of atmospheric turbulence-degraded video containing moving objects.
TLDR
Simulation experiments demonstrate that this new method for compensating image shifting in a video sequence while keeping real moving objects in the video unharmed significantly improves the accuracy of image restoration while preserving moving objects.
Simulating Anisoplanatic Turbulence by Sampling Correlated Zernike Coefficients
TLDR
A propagation-free method for simulating imaging through anisoplanatic atmospheric turbulence by establishing the equivalence between the angle-of-arrival correlation by Basu, McCrae and Fiorino (2015) and the multi-aperture correlation by Chanan (1992), and proposing fast and scalable sampling strategies to draw spatially correlated tilts and high-order abberations in the Zernike space.
Removing Turbulence Effect via Hybrid Total Variation and Deformation-Guided Kernel Regression
TLDR
A new scheme to recover a latent image from observed frames by integrating a new hybrid total variation model and deformation-guided spatial-temporal kernel regression can effectively alleviate distortion, and blur and recover details of the original scene compared to the state-of-the-art methods.
Simultaneous Video Stabilization and Moving Object Detection in Turbulence
TLDR
A novel three-term low-rank matrix decomposition approach in which a Gaussian-based turbulence model can be employed to enforce an additional constraint on the search space of the minimization.
...
1
2
3
4
5
...