Abstract | ||
---|---|---|
We propose a method for extracting low-dimensional features from omnidirectional Images to be used for robot localization and navigation. Edge detection is combined with thresholding to locate sharp edge pixels, the coordinates of which are led into a Parzen density estimator to compute the edge spatial density. The use of the Fast Fourier Transform makes this density estimate feasible in real-time,,while Principal Component Analysis further drops the dimensionality of the resulting feature vector to a manageable number. We show experimental results from a Nomad XR4000 robot in an office environment. |
Year | DOI | Venue |
---|---|---|
2001 | 10.1109/ROBOT.2001.932836 | 2001 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION, VOLS I-IV, PROCEEDINGS |
Keywords | Field | DocType |
robot kinematics,fft,real time,navigation,density estimation,fast fourier transform,edge detection,principal component analysis,feature vector,pca,mobile robots,fast fourier transforms,feature extraction,interpolation | Computer vision,Feature vector,Edge detection,Robot kinematics,Feature extraction,Fast Fourier transform,Artificial intelligence,Pixel,Thresholding,Mobile robot,Mathematics | Conference |
Volume | Issue | ISSN |
2 | 1 | 1050-4729 |
Citations | PageRank | References |
8 | 1.01 | 6 |
Authors | ||
5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Nikos A. Vlassis | 1 | 2050 | 158.24 |
Yoichi Motomura | 2 | 312 | 40.26 |
isao hara | 3 | 105 | 9.79 |
Hideki Asoh | 4 | 705 | 89.85 |
Toshihiro Matsui | 5 | 380 | 62.51 |