Title :
A robust visual attention system for detecting manufactured objects in underwater video
Author :
Barat, Christian ; Rendas, Maria-João
Author_Institution :
Nice Univ., Sophia Antipolis
Abstract :
The paper presents an approach to detect objects in underwater images. The method is composed of two main steps. The first step detects the motion of contrasting neighboring regions in consecutive frames, based on the Minimum Description Length test. The second step adapts a statistical snake to the boundary between the adjacent regions. The method is evaluated on real images and compared to classical snakes
Keywords :
object detection; object recognition; oceanographic techniques; underwater optics; video signal processing; Minimum Description Length test; consecutive frames; neighbouring regions motion; objects detection; robust visual attention system; statistical snake; underwater images; underwater video; Image edge detection; Lighting; Manufacturing; Motion detection; Object detection; Robots; Robustness; Testing; Underwater tracking; Working environment noise;
Conference_Titel :
OCEANS 2006
Conference_Location :
Boston, MA
Print_ISBN :
1-4244-0114-3
Electronic_ISBN :
1-4244-0115-1
DOI :
10.1109/OCEANS.2006.306810
