Multi-target polarization imaging in turbid water via Laplacian pyramid segmentation and local polarization optimization
Physica Scripta · 14.08.2026 · 101 (32) · 325510
DOI: 10.1088/1402-4896/ae9414 ↗Kurz gefasst
Fachartikel in der Zeitschrift Physica Scripta, online erschienen am 11.08.2026. Verfasst von Junwei Li und 4 weiteren Autor:innen. Die Arbeit nennt 27 Literaturangaben. Gefördert von: Natural Science Foundation of Hebei Province, Science Research Project of Hebei Education Department, National Natural Science Foundation of China.
Abstract (Originalsprache)
The underwater polarization imaging technique effectively removes backscattered light resulting from water scattering and absorption, thereby recovering a clear target. Previous studies have typically involved manually selecting background regions in order to estimate the global backscattered light. This approach overlooks the polarization characteristics of targets with diverse attributes and introduces errors in image recovery, thus limiting enhancements in multi-target imaging quality. In this study, we propose a new method for multi-target polarization imaging in turbid waters that does not rely on prior background knowledge. Edge feature images are generated from the I ∥ image using a Laplace pyramid, and different target regions are precisely segmented through the employment of adaptive thresholding. Subsequently, the parameters are adjusted according to different connectivity domains to further optimize target polarization estimation. Finally, the spatial distribution of backscattered light intensity is accurately estimated using low-pass filtering to reduce dependence on the background region, thereby recovering a clear multi-target scene. The experimental results demonstrate that the current method effectively eliminates backscattered light, even in background-free scenes. It optimally accounts for the polarization characteristics of each target within the imaging range, significantly improving the quality of recovery. As the process requires no manual intervention, it is well-suited for underwater scenes with multiple targets and complex structures. This approach offers an innovative solution for high-quality imaging in intricate underwater environments.
Förderung
- Natural Science Foundation of Hebei Province
- Science Research Project of Hebei Education Department
- National Natural Science Foundation of China
Zitieren
APA
Li, J., Lu, B., Yu, N., Wang, X., & Xi, S. (2026). Multi-target polarization imaging in turbid water via Laplacian pyramid segmentation and local polarization optimization. Physica Scripta, 101(32), 325510. https://doi.org/10.1088/1402-4896/ae9414
Vancouver
Li J, Lu B, Yu N, Wang X, Xi S. Multi-target polarization imaging in turbid water via Laplacian pyramid segmentation and local polarization optimization. Physica Scripta. 2026;101(32):325510. doi:10.1088/1402-4896/ae9414
BibTeX
@article{Li2026,
author = {Li, Junwei and Lu, Bocheng and Yu, Nana and Wang, Xiaolei and Xi, Sixing},
title = {Multi-target polarization imaging in turbid water via Laplacian pyramid segmentation and local polarization optimization},
journal = {Physica Scripta},
year = {2026},
volume = {101},
number = {32},
pages = {325510},
doi = {10.1088/1402-4896/ae9414}
}
EUVIMED