CVIVJul 27, 2022

Inverse Airborne Optical Sectioning

arXiv:2207.13344v111 citationsh-index: 35
Originality Incremental advance
AI Analysis

This addresses the challenge of surveillance and tracking in occluded environments like forests, which is incremental as it adapts existing radar principles to optical imaging.

The paper tackles the problem of detecting and tracking moving targets occluded by vegetation using a stationary optical sensor, by introducing Inverse Airborne Optical Sectioning (IAOS) as an optical analogy to ISAR, and shows that it enables efficient tracking in integral images where conventional methods fail.

We present Inverse Airborne Optical Sectioning (IAOS) an optical analogy to Inverse Synthetic Aperture Radar (ISAR). Moving targets, such as walking people, that are heavily occluded by vegetation can be made visible and tracked with a stationary optical sensor (e.g., a hovering camera drone above forest). We introduce the principles of IAOS (i.e., inverse synthetic aperture imaging), explain how the signal of occluders can be further suppressed by filtering the Radon transform of the image integral, and present how targets motion parameters can be estimated manually and automatically. Finally, we show that while tracking occluded targets in conventional aerial images is infeasible, it becomes efficiently possible in integral images that result from IAOS.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes