Title :
Visual data mining using omni-directional sensor
Author :
Kato, Koji ; Shibata, Tomoyuki ; Ishiguro, Hiroshi
Author_Institution :
Dept. of Media Archit., Future Univ. Hakodate, Hokkaido, Japan
fDate :
30 July-1 Aug. 2003
Abstract :
In this paper, we propose a method that analyses images recorded throughout long term. Conventional methods used feature extraction and geometric technique, when these images were analyzed. However, the problem of the lack of information occurs in such methods. Then, we propose the method that uses pattern conversion of the images and statistical technique. Moreover, we experimented, and the effectiveness of our method was verified.
Keywords :
data mining; image classification; image sensors; object detection; pattern clustering; statistical analysis; image acquisition; image classification; images recording; information extraction; object detection; omni-directional sensor; pattern conversion; spatio-temporal image; statistical technique; visual data mining; Cameras; Data mining; Image analysis; Image motion analysis; Image sequence analysis; Image sequences; Information analysis; Layout; Motion analysis; Object detection;
Conference_Titel :
Multisensor Fusion and Integration for Intelligent Systems, MFI2003. Proceedings of IEEE International Conference on
Print_ISBN :
0-7803-7987-X
DOI :
10.1109/MFI-2003.2003.1232671