JORDAN Mario Alberto
congresos y reuniones científicas
Optical-Flow Algorithms based on Correlation and Differential Methods for Self-Identification of Motion in Underwater Vehicles
JORDAN, MARIO ALBERTO; ARLENGHI, LEONARDO
Conferencia; 6th Argentine Symposium on Computing Technology AST 2006; 2005
In this paper two classes of methods for computing optical flow from a sequence of images oriented to the self-identification of motion in subaquatic vehicles is presented. Correlation and differential methods are analyzed and compared under the criteria of precision, computational cost and robustness against continuity in the frame sequence. For motion tracking a forgetting factor is applied. Two case studies are analyzed corresponding to the external motion of a rigid body and the ego-motion of a moving vehicle in a synthetic sea bottom.