Automated visual event detection (AVED)
Project Manager/Lead Engineer: Duane Edgington
Lead Scientist: Bruce Robison

One of the most successful of the feasibility studies supported in 2002 was the application of artificial intelligence to train a computer to detect events of interest in video frames. In a first test using ROV video data, the computer agreed with the professional video analyst 94% of the time. This project will continue in 2003 as a development project.

The project team proposes the development of technology to  visually process images for event detection and for recognition of target biological species. The applications for these technologies include 1) systems to be deployed on underwater instruments and on autonomous underwater robotic vehicles (AUVs), 2) lab analysis of underwater video, 3) real time analysis of video from cabled observatory cameras, and 4) analysis of video gathered by ROVs in real time. 

The ultimate goal is to identify an organism autonomously in real time. The project will apply the powerful neuromorphic saliency model of visual attention that has been implemented in software by Laurent Itti (University of Southern California) as a biological vision system-inspired processing approach to detecting visual events.  State of the art feature and motion detection chips, modeled after biological vision systems as silicon implementations of the selected algorithms, will be used. These neuromorphic systems feature ultra-low power, large dynamic range and intrinsic real-time image processing and feature detection with greatly reduced data storage requirements. The  science application focus includes both mid-water and deep ocean animals as well as animals that use bioluminescence in the deep sea.