A stereo system for vehicles, Unmanned Aerial Vehicles (UAVs), and hand eye systems must tolerate vibration of its cameras and be robust against disturbances. We propose a stereo self-adjustment methodology for resuming autonomous stereo system operations. The methodology aims to achieve autonomous recovery of camera directions, even if the cameras rotate after a weak impact. In autonomous recovery, the self-adjustment procedure calculates the rotation values of rotated cameras. The camera directions of the stereo system then recover their initial directions using the rotation parameters. Moreover, we have verified the validity of our approach through experiments using an active stereo camera in which three cameras can rotate independently.
|ジャーナル||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|
|出版物ステータス||Published - 2009 2 19|
|イベント||3rd Pacific Rim Symposium on Image and Video Technology, PSIVT 2009 - Tokyo, Japan|
継続期間: 2009 1 13 → 2009 1 16
ASJC Scopus subject areas
- Theoretical Computer Science
- Computer Science(all)