Title :
Segmentation methods for visual tracking of deep-ocean jellyfish using a conventional camera
Author :
Rife, Jason ; Rock, Stephen M.
Author_Institution :
Stanford Univ., CA, USA
Abstract :
This paper presents a vision algorithm that enables automated jellyfish tracking using remotely operated vehicles (ROVs) or autonomous underwater vehicles (AUVs). The discussion focuses on algorithm design. The introduction provides a novel performance-assessment tool, called segmentation efficiency, which aids in matching potential vision algorithms to the jelly-tracking task. This general-purpose tool evaluates the inherent applicability of various algorithms to particular tracking applications. This tool is applied to the problem of tracking transparent jellyfish under uneven time-varying illumination in particle-filled scenes. The result is the selection of a fixed-gradient threshold-based vision algorithm. This approach, implemented as part of a pilot aid for the Monterey Bay Aquarium Research Institute´s ROV Ventana, has demonstrated automated jelly tracking for as long as 89 min.
Keywords :
biological techniques; image segmentation; natural scenes; optical tracking; remotely operated vehicles; underwater vehicles; 89 min; AUV; ROV; automated jellyfish tracking; autonomous underwater vehicles; conventional camera deep-ocean visual tracking; deep-ocean jellyfish visual tracking; fixed-gradient threshold-based vision algorithm; image segmentation; natural images; natural scene visual tracking; particle-filled scenes; remotely operated vehicles; segmentation efficiency; transparent jellyfish; uneven time-varying illumination; Algorithm design and analysis; Cameras; Clustering algorithms; Image segmentation; Layout; Remotely operated vehicles; Robots; Underwater tracking; Underwater vehicles; Visual servoing;
Journal_Title :
Oceanic Engineering, IEEE Journal of
DOI :
10.1109/JOE.2003.819315