A numeral character recognition using the PCA mixture model

Pattern Recognition Letters

Volumn 23, Issue 1-3, 2002, Pages 103-111

  Kim, Hyun Chul ^a   Kim, Daijin ^a   Yang Bang, Sung ^a  

^a Pohang University of Science and Technology (POSTECH)   (South Korea)

Author keywords

EM algorithm; Mixture model; Numeral character recognition; PCA mixture model; Principal Component Analysis

Indexed keywords

ALGORITHMS; COMPUTER SIMULATION; INDEPENDENT COMPONENT ANALYSIS; ITERATIVE METHODS; PARAMETER ESTIMATION; PRINCIPAL COMPONENT ANALYSIS;

EXPECTATION MAXIMIZATION (EM) ALGORITHMS;

CHARACTER RECOGNITION;

IMAGE CLASSIFICATION; PRINCIPAL COMPONENT ANALYSIS;

EID: 0036132369     PISSN: 01678655     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0167-8655(01)00093-9     Document Type: Article

References (13)

1. Maximum likelihood from incomplete data via the EM algorithm
2. Modeling the manifolds of images of handwritten digits
3. Analysis of a complex statistical variables into principal components
4. A database for handwritten text recognition research
5. Adaptive mixtures of local experts
6. Hierarchical mixtures of experts and the EM algorithm
7. Unsupervised classification with non-Gaussian mixture models using ICA
8. Mixtures of probabilistic principal component analyzers
9. A subclass model for non-linear pattern classification