HR: 1340h
AN: IN13B-1216    [Abstracts]
TI: An Automated Visual Event Detection System for Cabled Observatory Video
AU: * Edgington, D R
EM: duane@mbari.org
AF: Monterey Bay Aquarium Research Institute (MBARI), 7700 Sandholdt Road, Moss Landing, CA 95039, United States
AU: Cline, D E
EM: dcline@mbari.org
AF: Monterey Bay Aquarium Research Institute (MBARI), 7700 Sandholdt Road, Moss Landing, CA 95039, United States
AU: Mariette, J
EM: jerome@mbari.org
AF: Monterey Bay Aquarium Research Institute (MBARI), 7700 Sandholdt Road, Moss Landing, CA 95039, United States
AB: The permanent presence of underwater cameras on oceanic cabled observatories, such as the Victoria Experimental Network Under the Sea (VENUS) and Eye-In-The-Sea (EITS) on Monterey Accelerated Research System (MARS), will generate valuable data that can move forward the boundaries of understanding the underwater world. However, sightings of underwater animal activities are rare, resulting in the recording of many hours of video with relatively few events of interest. The burden of video management and analysis often requires reducing the amount of video recorded and later analyzed. Sometimes enough human resources do not exist to analyze the video; the strains on human attention needed to analyze video demand an automated way to assist in video analysis. Towards this end, an Automated Visual Event Detection System (AVED) is in development at the Monterey Bay Aquarium Research Institute (MBARI) to address the problem of analyzing cabled observatory video. Here we describe the overall design of the system to process video data and enable science users to analyze the results. We present our results analyzing video from the VENUS observatory and test data from EITS deployments. This automated system for detecting visual events includes a collection of custom and open source software that can be run three ways: through a Web Service, through a Condor managed pool of AVED enabled compute servers, or locally on a single computer. The collection of software also includes a graphical user interface to preview or edit detected results and to setup processing options. To optimize the compute-intensive AVED algorithms, a parallel program has been implemented for high-data rate applications like the EITS instrument on MARS.
UR: http://www.mbari.org/AVED/
DE: 0540 Image processing
DE: 0594 Instruments and techniques
DE: 3050 Ocean observatories and experiments
DE: 4264 Ocean optics (0649)
DE: 4294 Instruments and techniques
SC: Earth and Space Science Informatics [IN]
MN: 2007 Fall Meeting