Title :
A new method to calculate looming for autonomous obstacle avoidance
Author :
Joarder, Kunal ; Raviv, Daniel
Author_Institution :
Dept. of Electr. Eng., Florida Atlantic Univ., Boca Raton, FL, USA
Abstract :
The concept of visual looming can be used as a powerful visual cue for autonomous obstacle avoidance. In this paper a method that measures looming quantitatively by fixating a camera at a point on the surface of an object is presented. It is based on studying the texture and its temporal change near the fixation point. The surface may be tilted relative to the optical axis. Looming can be calculated from the relative change in texture density and from the local orientation of the surface. The orientation is obtained from a set of one-dimensional directional densities of the texture primitives. This visual cue is used as a feedback signal of a closed loop system to implement obstacle avoidance using a six-degree-of-freedom simulator
Keywords :
computer vision; image texture; mobile robots; path planning; autonomous obstacle avoidance; closed loop system; obstacle avoidance; orientation; temporal change; texture; texture density; visual looming; Image orientation analysis; Image texture analysis; Mobile robots, motion planning; Object recognition; Robots, vision systems;
Conference_Titel :
Computer Vision and Pattern Recognition, 1994. Proceedings CVPR '94., 1994 IEEE Computer Society Conference on
Conference_Location :
Seattle, WA
Print_ISBN :
0-8186-5825-8
DOI :
10.1109/CVPR.1994.323897