Aim

The aim of the project is to advance 3D underwater imaging by enhancing its resilience and data resolution, thereby enabling high-level tasks such as object detection and tracking.

Objectives

1. Fusion of data from passive and active cameras to generate high-resolution 3D underwater images
2. Perform high level vision tasks in underwater environments (detection of pipes defects, tracking of moving objects, etc)
3. Interfacing statistical modelling with deep learning for enhanced performance and AI interpretability.

Description

3D LiDAR imaging captures comprehensive details about a scene, providing both reflectivity and depth information. At Heriot-Watt University  (HWU) and the University of Edinburgh, we have developed strong expertise in single-photon imaging techniques and data processing [1,2]. This approach provides exceptional sensitivity and depth resolution compared to other techniques, but it currently faces challenges with lower spatial
resolution and reduced performance in highly scattering environments. 

This project aims to enhance 3D imaging underwater by combining single-photon sensors with passive or active optical sensors that offer complementary strengths, such as improved resolution and resilience to scattering. The goal is to achieve high-resolution 3D imaging that enables advanced vision tasks such as detecting pipe defects or tracking moving objects. A focus will be on the development of computational algorithms
at the interface of statistical modelling and deep learning. This approach enhances imaging performance and AI interpretability, ensuring robustness and reliability for underwater applications.

References 

[1] S. Plosz, A. Maccarone, S. McLaughlin, G. S. Buller, A. Halimi, "Real-Time Reconstruction of 3D Videos from Single-Photon
LiDaR Data in the Presence of Obscurants", IEEE-TCI, vol. 9, p106-119, Feb. 2023.

[2] A. Maccarone, K. Drummond, A. McCarthy, UK Steinlehner, J. Tachella, DA Garcia et al, 'Submerged single-photon LiDAR imaging sensor used for real-time 3D scene reconstruction in scattering underwater environments', Optics Express, vol. 31, no. 10, 2023.