DocumentCode :
505191
Title :
A study on detecting elevator entrance door using stereo vision in multi floor environment
Author :
Baek, JunYoung ; Lee, Mincheol
Author_Institution :
Sch. of Mech. Eng., Univ. of Pusan Nat., Busan, South Korea
fYear :
2009
fDate :
18-21 Aug. 2009
Firstpage :
1370
Lastpage :
1373
Abstract :
This paper describes algorithm to recognize an elevator entrance door in the multi floor environment in the house and office. The algorithm is useful in developing a service mobile robot which can be the moving on the environment. First, the process to detect the entrance is started with a reliable camera calibration and then through image process to acquire a clear image. Second, a candidate of elevator door is extracted by method to detect the corner of an elevator door using stereo vision instead of conventional method such as clustering process using a laser scanner. Third, the size of an elevator door is compared the uncertain door image acquired from stereo vision with the information of the standard size to find an elevator door. This paper confirmed that the proposed algorithm is easily to detect the elevator door on the multi floor.
Keywords :
mobile robots; robot vision; service robots; stereo image processing; camera calibration; elevator entrance detection; image process; multi floor environment; service mobile robot; stereo vision; Calibration; Cameras; Data mining; Elevators; Mechanical engineering; Mobile robots; Navigation; Robot kinematics; Robot vision systems; Stereo vision;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
ICCAS-SICE, 2009
Conference_Location :
Fukuoka
Print_ISBN :
978-4-907764-34-0
Electronic_ISBN :
978-4-907764-33-3
Type :
conf
Filename :
5335409
Link To Document :
بازگشت