A Real-time Edge-AI System for Reef Surveys
This addresses the need for efficient COTS identification to assist marine experts in surveillance and control programs, though it is incremental as it applies existing deep learning techniques to a specific domain.
The paper tackles the problem of Crown-of-Thorn Starfish (COTS) outbreaks causing coral loss on the Great Barrier Reef by developing a real-time edge-AI system for underwater monitoring, resulting in a resource-efficient detector that runs on edge hardware with low resource consumption and low information loss.
Crown-of-Thorn Starfish (COTS) outbreaks are a major cause of coral loss on the Great Barrier Reef (GBR) and substantial surveillance and control programs are ongoing to manage COTS populations to ecologically sustainable levels. In this paper, we present a comprehensive real-time machine learning-based underwater data collection and curation system on edge devices for COTS monitoring. In particular, we leverage the power of deep learning-based object detection techniques, and propose a resource-efficient COTS detector that performs detection inferences on the edge device to assist marine experts with COTS identification during the data collection phase. The preliminary results show that several strategies for improving computational efficiency (e.g., batch-wise processing, frame skipping, model input size) can be combined to run the proposed detection model on edge hardware with low resource consumption and low information loss.