多光谱图像
加权
计算机科学
贝叶斯概率
激光雷达
光学(聚焦)
人工智能
贝叶斯推理
遥感
计算机视觉
算法
模式识别(心理学)
数据挖掘
地理
光学
物理
声学
作者
Abderrahim Halimi,Jakeoung Koo,Robert A. Lamb,Gerald S. Buller,Stephen McLaughlin
标识
DOI:10.1109/icassp43922.2022.9746166
摘要
This paper presents a new Bayesian algorithm for the robust reconstruction of multispectral single-photon Lidar data acquired in extreme conditions. We focus on imaging through obscurants (i.e., fog, water) leading to high and possibly non-uniform background noise. The proposed hierarchical Bayesian method accounts for multiscale information to provide distribution estimates for the target’s depth and reflectivity, i.e., point and uncertainty measures of the estimates to improve decision making. The correlations between variables are enforced using a weighting scheme that allows the incorporation of guide information available from other sensors or state-of-the-art algorithms. Results on synthetic and real data show improved reconstruction of the scene in extreme conditions when compared to the state-of-the-art algorithms.
科研通智能强力驱动
Strongly Powered by AbleSci AI