Detecting Motion through Dynamic Refraction

Abstract

Abstract-Refraction causes random dynamic distortions in atmospheric turbulence and in views across a water interface. The latter scenario is experienced by submerged animals seeking to detect prey or avoid predators, which may be airborne or on land. Man encounters this when surveying a scene by a submarine or divers while wishing to avoid the use of an attention-drawing periscope. The problem of inverting random refracted dynamic distortions is difficult, particularly when some of the objects in the field of view (FOV) are moving. On the other hand, in many cases, just those moving objects are of interest, as they reveal animal, human, or machine activity. Furthermore, detecting and tracking these objects does not necessitate handling the difficult task of complete recovery of the scene. We show that moving objects can be detected very simply, with low false-positive rates, even when the distortions are very strong and dominate the object motion. Moreover, the moving object can be detected even if it has zero mean motion. While the object and distortion motions are random and unknown, they are mutually independent. This is expressed by a simple motion feature which enables discrimination of moving object points versus the background.

STELLA MARIS: Stellar marine refractive imaging sensor

“…Distortions can be reduced significantly using this physics-based method, but uncertainties pointed theoretically leave residual distortions. Nevertheless, images corrected by our process can later be handled by statistical un-distortion post-processes [26,29,47,48,50], such as lucky imaging [8,9,44], stochastic triangulation [2], and motion detection [1]. Rather than facing the full-blown distortions in raw images, such video post-processing methods may handle more easily images whose distortions are residual.…”

Triangulation in Random Refractive Distortions

“…Ref. [1] seeks motion detection through a wavy WAI. There, a motion classifier is used, based on statistical properties of water waves, without needing image rectification.…”

The aperture problem for refractive motion

“…Xue et al [31] estimate the fluid motion by tracking the distortion of the observed sequence over time. Alterman [7] proposed to recover turbulence strength field using linear tomography. However, we are not aware of any theoretical analysis of local motion ambiguity for refractive objects, and our paper tries to fill this gap.…”

“…It is easy to show that for any motion vector u, we can design r, J, and n such that 6 Notice that the observed motion of image boundary on the image plane (marked by the red arrow) does not necessarily have the same direction as the motion of the refractive object (marked by the blue arrow). 7 Second order term in r can be ignored because otherwise a curved pattern would have been observed instead of a straight edge. they satisfy this line equation and hence generate the same observed sequence for the boundary (see supplementary material for detailed proof).…”