Detection of moving objects through turbulent media. Decomposition of Oscillatory vs Non-Oscillatory spatio-temporal vector fields

TL;DR

This paper presents a geometric spatio-temporal point of view to the problem and shows that it is possible to distinguish movement due to the turbulence vs. moving objects.

Abstract

In this paper, we investigate how moving objects can be detected when images are impacted by atmospheric turbulence. We present a geometric spatio-temporal point of view to the problem and show that it is possible to distinguish movement due to the turbulence vs. moving objects. To perform this task, we propose an extension of 2D cartoon+texture decomposition algorithms to 3D vector fields. Our algorithm is based on curvelet spaces which permit to better characterize the movement flow geometry. We present experiments on real data which illustrate the efficiency of the proposed method.

Figures (20)

• Figure 1: Several frames from the OTIS Car1 sequence (frame index in first column and actual frame in second column). The third column shows the velocity vector field corresponding to each frame.
• Figure 2: Several frames from the OTIS Car2 sequence (frame index in first column and actual frame in second column). The third column shows the velocity vector field corresponding to each frame.
• Figure 3: Several frames from the OTIS Car4 sequence (frame index in first column and actual frame in second column). The third column shows the velocity vector field corresponding to each frame.
• Figure 4: 3D spatio-temporal patterns visualization in the velocity vector fields of the Car1 and Car2 sequences.
• Figure 5: Decomposition results on the OTIS Car1 sequence. The first column gives the frame index, the second one corresponds to the geometric component and the third one to the oscillating component.