Pattern recognition using boundary data of component distributions |
| |
| Authors: | Masako Omachi Shinichiro Omachi Hirotomo Aso Tsuneo Saito |
| |
| Affiliation: | 1. Advanced Course of Production System and Design Engineering, Sendai National College of Technology, 48 Nodayama, Medeshima-Shiote, Natori-shi 981-1239, Japan;2. Graduate School of Engineering, Tohoku University, Aoba 6-6-05, Aramaki, Aoba-ku, Sendai-shi 980-8579, Japan;3. College of Engineering, Nihon University, 1 Naka-gawara, Tokusada, Tamura-machi, Koriyama-shi 963-8642, Japan;4. Faculty of Science and Technology, Tohoku Bunka Gakuen University, 6-45-1, Kunimi, Aoba-ku, Sendai-shi 981-8551, Japan |
| |
| Abstract: | In statistical pattern recognition, a Gaussian mixture model is sometimes used for representing the distribution of vectors. The parameters of the Gaussian mixture model are usually estimated from given sample data by the expectation maximization algorithm. However, when the number of data attributes is large, the parameters cannot be estimated correctly. In this paper, we propose a novel approach for estimating the parameters of the Gaussian mixture model by using sample data located on the boundary of regions defined by the component density functions. Experiments are carried out to show the characteristics of the proposed method. |
| |
| Keywords: | Pattern recognition Gaussian mixture model Parameter estimation High-dimensional vector Probabilistic model |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
