Title :
Visual navigation using pixel intensity information
Author :
Lee, Changmin ; Kim, DaeEun
Author_Institution :
Biol. Cybern. Lab., Yonsei Univ., Seoul, South Korea
Abstract :
Homing navigation is an important survival skill in animals. They have fast and accurate homing capability. In this paper, we propose a new technique that encodes landmark vectors using gray values of pixels. It has been improved by using a sorting method which removes unnecessary part in the image. Based on the snapshot model, the reference map is created by using the landmark vector set of a snapshot image. Then the landmark arrangement method is applied between landmark vectors from the current image and landmarks on the reference map to do the localization. After localization, the homing navigation can be done by estimating the homing direction using the estimated position on the map. We show the result of the suggested method and effectiveness in aspect of homing navigation.
Keywords :
image coding; image colour analysis; mobile robots; robot vision; animal; gray value; homing navigation; landmark arrangement method; landmark vector encoding; mobile robot; pixel intensity information; reference map; snapshot model; sorting method; survival skill; visual navigation; Biology; Cybernetics; Image color analysis; Navigation; Robots; Vectors; Visualization; intensity; landmark arrangement; landmark pixel; localization; reference map; snapshot model; visual Navigation;
Conference_Titel :
Control, Automation and Systems (ICCAS), 2012 12th International Conference on
Conference_Location :
JeJu Island
Print_ISBN :
978-1-4673-2247-8