Title :
Resilient Subclass Discriminant Analysis
Author :
Wu, Dijia ; Boyer, Kim L.
Author_Institution :
Rensselaer Polytech. Inst., Troy, NY, USA
fDate :
Sept. 29 2009-Oct. 2 2009
Abstract :
We propose a dimension reduction technique named Resilient Subclass Discriminant Analysis (RSDA) for high dimensional classification problems. The technique iteratively estimates the subclass division by embedding the Fisher Discriminant Analysis (FDA) with Expectation-Maximization (EM) in Gaussian Mixture Models (GMM). The new method maintains the adaptability of SDA to a wide range of data distributions by approximating the distribution of each class as a mixture of Gaussians, and provides superior feature selection performance to SDA with modified EM clustering that estimates a posteriori probability of latent variables in lower-dimensional Fisher´s discriminant space, which also improves the robustness in problems of small training datasets compared with conventional EM algorithm. Extensive experiments and comparison results against other well-known Discriminant Analysis (DA) methods are presented using synthetic data, benchmark datasets as well as a real computational vision problem.
Keywords :
Gaussian processes; expectation-maximisation algorithm; feature extraction; iterative methods; pattern classification; Fisher discriminant analysis; Gaussian mixture model; computational vision problem; data distribution; dimension reduction technique; expectation maximization method; feature selection; high dimensional classification problems; iterative estimation; resilient subclass discriminant analysis; variables posteriori probability; Clustering algorithms; Clustering methods; Computer vision; Covariance matrix; Feature extraction; Kernel; Linear discriminant analysis; Neural networks; Pattern recognition; Robustness;
Conference_Titel :
Computer Vision, 2009 IEEE 12th International Conference on
Conference_Location :
Kyoto
Print_ISBN :
978-1-4244-4420-5
Electronic_ISBN :
1550-5499
DOI :
10.1109/ICCV.2009.5459212