Higher Order Autocorrelations For Pattern Classification (2001)
| Venue: | In Proceedings of the International Conference on Image Processing (ICIP). IEEE |
BibTeX
@INPROCEEDINGS{And01higherorder,
author = {Vlad Popovici And and Vlad Popovici},
title = {Higher Order Autocorrelations For Pattern Classification},
booktitle = {In Proceedings of the International Conference on Image Processing (ICIP). IEEE},
year = {2001},
pages = {724--727}
}
Bookmark
 |
 |
 |
 |
 |
 |
OpenURL
Abstract
The use of higher-order local autocerrelations as features for pattern recognition has been acknowledged since many years, but their applicability was restricted to relatively low orde (2 or 3) and small local neighborhoods, due to com binatorial increase in computational costs. In this paper a new method for using these features is presented, which allows the use of autocorrelations of any order and of larger neighborhoods. The method is closely related to the classifier used, a Support Vector Machine (SVM), and exploits the special form of the inner products of autocorrelations and the properties of some kernel lhnctions used by SVMs. Using SVM, linear and non-linear classification functions can be learned, extending the previous works on higher-order autocorrelations which were based on linear classifiers.
Citations
| 6694 | Statistical Learning Theory - Vapnik - 1998 |
| 233 | XM2VTSDB: The Extended M2VTS Database - Messer, Matas, et al. - 1999 |
| 12 | Nth–order autocorrelations in pattern recognition - McLaughlin, Raviv - 1968 |
| 12 | Scale and rotation invariant recognition method using higher-order local autocorrelation features of log-polar image - Kurita, Hotta, et al. - 1998 |
| 11 | Scale invariant image recognition based on higher order autocorrelation features - Kreutz, Volpel, et al. - 1996 |
| 6 | Higher-order statistics in visual object recognition - Breuel - 1993 |
