Robust Bayesian Reconstruction of Multispectral Single-Photon 3D Lidar Data with Non-Uniform Background

Abstract

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.