Developing a photo-acoustic imaging system for real-time crude oil characterization on a mobile underwater platform

Abstract

Oil spills pose severe threats to marine ecosystems and coastal economies. Detecting and characterizing oil beneath the water surface or under ice remains challenging, as existing methods such as satellite imaging, radar, sonar, and infrared sensing face limitations in resolution, reliability, or applicability under complex marine conditions. Underwater photoacoustic (PA) imaging (PAI), which leverages the strong optical absorption of oil and the deep penetration of ultrasound, has recently emerged as a promising alternative. This thesis presents the development of a compact underwater PA imager (UPAI) for real-time oil spill detection and characterization. The system integrates a nine-element hydrophone array, analog front-end electronics, and field-programmable gate array (FPGA)-based processing into a portable unit designed for deployment on autonomous platforms. By implementing time-of-arrival (TOA) detection, oil thickness estimation, and imaging directly on the embedded processor, the UPAI reduces the high-throughput data transmission requirements and power demands while enabling real-time, in-situ operation in mobile and remote environments. The contributions of this research are fourfold. First, a theoretical transfer function model for PA propagation in layered media is developed to describe signal generation and transmission. Second, a real-time imaging framework is introduced that combines robust detection of weak and saturated pulses with model-based multi-sensor imaging; using TOA measurements across hydrophones, the UPAI reconstructs oil layers without prior knowledge of laser angle or water sound speed. Third, key hardware, firmware, and software modules of an FPGA-based architecture are developed and integrated toward a complete underwater-deployable UPAI, enabling efficient real-time signal processing. Fourth, the system is experimentally evaluated in controlled aquatic conditions using an emulated PA source, demonstrating the feasibility of estimating oil layer thickness and depth. Collectively, these advances establish the foundation for a compact and deployable UPAI for underwater environmental monitoring and oil spill detection.