Adaptive Spiral Scanning for Confocal Non-Line-of-Sight Imaging
IEEE Open Journal of Signal Processing · 2026 · 7 · 482-491
DOI: 10.1109/ojsp.2026.3688052 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Open Journal of Signal Processing, 2026. Verfasst von Tomoya Oyama und 2 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: JST Presto, JSPS KAKENHI.
Abstract (Originalsprache)
Non-line-of-sight (NLOS) imaging aims to reconstruct objects that are hidden from direct view by utilizing light reflected from a relay surface. Conventional NLOS imaging methods typically rely on uniform raster scanning of the relay surface, which poses a challenge due to the long acquisition times. To address this challenge, researchers have proposed various scanning trajectories, such as circular and spiral patterns. However, these methods rely on predefined scanning trajectories. As a result, scanning points cannot be adjusted according to the intensity distribution on the relay surface. When hidden objects are spatially localized, these methods do not allocate sufficient sampling to regions with strong return intensity, leading to reduced reconstruction quality. In this paper, we propose Dynamic Archimedean Spiral Confocal NLOS (DAS-CNLOS), which dynamically selects scanning trajectories based on the observed return intensity. Our method efficiently captures the return intensity from hidden objects using a limited number of scanning points. This is achieved by adaptively shifting the spiral center toward high-intensity regions based on a sequentially estimated intensity distribution. Furthermore, for 2D reconstruction, we introduce Voronoi Density Compensation (VDC) to correct sampling density non-uniformity induced by dynamic spiral scanning, enabling robust and high quality reconstructions. Experimental results demonstrate that our DAS-CNLOS achieves superior 2D reconstruction quality compared to conventional circular scanning and spiral scanning methods.
Förderung
- JST Presto
- JSPS KAKENHI
Zitieren
APA
Oyama, T., Yang, D., & Isogawa, M. (2026). Adaptive Spiral Scanning for Confocal Non-Line-of-Sight Imaging. IEEE Open Journal of Signal Processing, 7, 482-491. https://doi.org/10.1109/ojsp.2026.3688052
Vancouver
Oyama T, Yang D, Isogawa M. Adaptive Spiral Scanning for Confocal Non-Line-of-Sight Imaging. IEEE Open Journal of Signal Processing. 2026;7:482-491. doi:10.1109/ojsp.2026.3688052
BibTeX
@article{Oyama2026,
author = {Oyama, Tomoya and Yang, Dixin and Isogawa, Mariko},
title = {Adaptive Spiral Scanning for Confocal Non-Line-of-Sight Imaging},
journal = {IEEE Open Journal of Signal Processing},
year = {2026},
volume = {7},
pages = {482--491},
doi = {10.1109/ojsp.2026.3688052}
}
EUVIMED