Journal of University of Chinese Academy of Sciences >
An automatic object detection method based on covariance matrix
Received date: 2009-10-10
Revised date: 2010-01-18
Online published: 2010-05-15
In order to apply the covariance matrix algorithm to automatic target detection we present feature similarity and covariance matrix similarity. Feature similarity is the similarity of the target feature. Covariance matrix similarity integrates all the feature similarities. In addition, because features are different in validity and importance, we raise minimized feature similarity. Minimized feature similarity can be used to get rid of basically ineffective features. Experiments show that with this method one can effectively apply the covariance matrix algorithm to automatic target detection with high detection rate and low false alarm rate.
Key words： covariance matrix; automatic target detection; feature fusion
NING Zhong-Lei , WANG Hong-Qi , ZHANG Zheng . An automatic object detection method based on covariance matrix[J]. Journal of University of Chinese Academy of Sciences, 2010 , 27(3) : 370 -375 . DOI: 10.7523/j.issn.2095-6134.2010.3.010
[1] Tuzel O, Porikli F, Meer P. Region covariance: A fast despcriptor for detection and classification //ECCV. 2006.
[2] Tom Fawceet. An introduction to ROC analysis
[J]. Pattern Recognition Letters, 2005.
[3] Forstner W, Moonean B. A metric for covariance matrices . Technical Report. Dept of Geodesy and Geoinformatics, Stuttgart University, 1999.
[4] Rowley H, Baluja S, Kanade T. Neural network-based face detection
[J]. IEEE Trans Pattern Recognition and Machine Learning, 1998, 20(1):23-38.
[5] Schneiderman H, Kanade T. Probabilistic modeling of local appearance and spatial relationships for object recognition //Proc IEEE Conference on Computer Vision and Pattern Recognition. Santa Barbara, California, 1998:45-51.
[6] Feraud R, Olivier J Bernier, Viallet J, et al. A fast and accurate face detector based on neural networks
[J]. IEEE Trans on Pattern Analysis and Machine Intelligence, 2001,23(1):42-53.
[7] Viola P. Rapid object detection using a Boosted cascade of simple features //Proc IEEE Conference on Computer Vision and Pattern Recognition. 2001:511-518.
[8] Lienhart R, Maydt J. An extended set of haar-like features for rapid object detection
[J]. IEEE ICIP 2002. 2002,1:900-903.
[9] Li S Z, Zhu L, Zhang Z Q, et al. Statistical learning of multi-view face detection //Proceedings of the 7th European Conference on Computer Vision.Copenhagen, Denmark, 2002.
[10] Frba B, Ernst A. Fast frontal-view face detection using a multi-path decision tree //Proc Audio- and Video-based Biometric Person Authentication (AVBPA 2003), 2003:921-928.
/
| 〈 |
|
〉 |