1. Comparing optical flow algorithms using 6-DOF motion of real-world rigid objects
- Author
-
Giampiero Campa, Marco Mammarella, Marcello R. Napolitano, and Mario Luca Fravolini
- Subjects
Camera fields ,Computer science ,Guidance and navigation ,ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION ,Optical flow ,Mathematics::General Topology ,Autonomous aircraft ,Comparative studies ,Onboard camera ,Optical flow algorithm ,Rigid body ,Rigid objects ,Cameras ,Computer vision ,Image processing ,Optical flows ,Angular velocity ,Motion (physics) ,Motion estimation ,Electrical and Electronic Engineering ,business.industry ,Computer Science Applications ,Human-Computer Interaction ,Motion field ,Control and Systems Engineering ,Artificial intelligence ,business ,Algorithm ,Software ,Information Systems - Abstract
The application of optical flow algorithms to guidance and navigation problems has gained considerable interest in recent years. This paper summarizes the results of a comparative study on the accuracy of nine different optical flow (OF) algorithms using videos that are captured from an on-board camera during the flight of an autonomous aircraft model. The comparison among the algorithms relies on two formulas that are used both to calculate the ideal OF generated by the motion of a rigid body in the camera field of view and to estimate the linear and angular velocity from the OF.
- Published
- 2012