Restoring atmospheric-turbulence-degraded images

Abstract

Image data experiences geometric distortions and spatial-temporal varying blur due to the strong effects of random spatial and temporal variations in the optical refractive index of the communication path. Simultaneously removing these effects from an image is a challenging task. An efficient approach is proposed in this paper to address this problem. The approach consists of four steps. First, a frame selection strategy is employed by proposing an unsupervised k-means clustering technique. Second, a B-spline-based nonrigid image registration is carried out to suppress geometric distortions. Third, a spatiotemporal kernel regression is proposed by introducing the local sharp patch concept to fuse the registered frame sequences into an image. Finally, a blind deconvolution technique is employed to deblur the fused image. Experiments are carried out with synthetic and real-world turbulence-degraded data by implementing the proposed method and two recently reported methods. The proposed method demonstrates significant improvement over the two reported methods in terms of alleviating blur and distortions, as well as improving visual quality.

An optimal transport model for imaging in atmospheric turbulence

“…We believe the physical model described above could inform a new category of computational imaging methods for overcoming the barrier imposed by turbulence in open air imaging and communications. With regards to image enhancement, current algorithms for removing the effects of turbulence use an image registration-based procedure for spatially aligning (warping) sequential frames in a video segment [26], [37] . Our theory suggests that rather than being aligned, consecutive frames should be morphed instead via transport-based modeling.…”

“…Our theory suggests that rather than being aligned, consecutive frames should be morphed instead via transport-based modeling. Moreover, the model linking clean and corrupted images should not be linear (e.g., "deconvoution" methods, see again [26], [37]), but should instead involve the inversion of optimal transport.…”

Learning to Restore Images Degraded by Atmospheric Turbulence Using Uncertainty

“…Under the assumption that the scene and the imaging sensor are both static and that observed motions are due to the air turbulence alone, the image degradation due to atmospheric turbulence can be mathematically formulated as follows [2], [3], [4], [5]…”

Restoration of atmospheric turbulence-distorted images via RPCA and quasiconformal maps

“…Since we are working on a sequence of distorted images or turbulence-degraded video, we assume the original image is static and the image sensor is also fixed. In order to model this problem, the mathematical model in this paper is based on [10], [11], I t (x) = [D t H t (I) ](x) + n t (x), t = 1, · · · , N (1) where I t , I, and n t are the captured frame at time t, the true image, and the sensor noise respectively. The vector x lies in the two-dimensional Euclidean space.…”