In general cases of pattern recognition, a pattern to be recognized is first represented by a set of features and the measured values of the features are then classified. Finding features relevant to recognition is thus an important issue in recognizer design. As a fundamental design framework taht systematically enables one to realize such useful features, the Subspace Method (SM) has been extensively used in various recognition tasks. However, this promising methodological framework is still inadequate. The discriminative power of early versions was not very high. The training behavior of a recent discriminative version called the Learning Subspace Method has not been fully clarified due to its empirical definition, though its discriminative power has been improved. To alleviate this insufficiency, we propose in this paper a new discriminative SM algorithm based on the Minimum Classification Error/Generalized Probabilistic Descent method and show that the proposed algorithm achieves an optimal accurate recognition result, i.e., the (at least locally) minimum recognition error situation, in the probabilistic descent sense.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Hideyuki WATANABE, Shigeru KATAGIRI, "Subspace Method for Minimum Error Pattern Recognition" in IEICE TRANSACTIONS on Information,
vol. E80-D, no. 12, pp. 1195-1204, December 1997, doi: .
Abstract: In general cases of pattern recognition, a pattern to be recognized is first represented by a set of features and the measured values of the features are then classified. Finding features relevant to recognition is thus an important issue in recognizer design. As a fundamental design framework taht systematically enables one to realize such useful features, the Subspace Method (SM) has been extensively used in various recognition tasks. However, this promising methodological framework is still inadequate. The discriminative power of early versions was not very high. The training behavior of a recent discriminative version called the Learning Subspace Method has not been fully clarified due to its empirical definition, though its discriminative power has been improved. To alleviate this insufficiency, we propose in this paper a new discriminative SM algorithm based on the Minimum Classification Error/Generalized Probabilistic Descent method and show that the proposed algorithm achieves an optimal accurate recognition result, i.e., the (at least locally) minimum recognition error situation, in the probabilistic descent sense.
URL: https://global.ieice.org/en_transactions/information/10.1587/e80-d_12_1195/_p
Copy
@ARTICLE{e80-d_12_1195,
author={Hideyuki WATANABE, Shigeru KATAGIRI, },
journal={IEICE TRANSACTIONS on Information},
title={Subspace Method for Minimum Error Pattern Recognition},
year={1997},
volume={E80-D},
number={12},
pages={1195-1204},
abstract={In general cases of pattern recognition, a pattern to be recognized is first represented by a set of features and the measured values of the features are then classified. Finding features relevant to recognition is thus an important issue in recognizer design. As a fundamental design framework taht systematically enables one to realize such useful features, the Subspace Method (SM) has been extensively used in various recognition tasks. However, this promising methodological framework is still inadequate. The discriminative power of early versions was not very high. The training behavior of a recent discriminative version called the Learning Subspace Method has not been fully clarified due to its empirical definition, though its discriminative power has been improved. To alleviate this insufficiency, we propose in this paper a new discriminative SM algorithm based on the Minimum Classification Error/Generalized Probabilistic Descent method and show that the proposed algorithm achieves an optimal accurate recognition result, i.e., the (at least locally) minimum recognition error situation, in the probabilistic descent sense.},
keywords={},
doi={},
ISSN={},
month={December},}
Copy
TY - JOUR
TI - Subspace Method for Minimum Error Pattern Recognition
T2 - IEICE TRANSACTIONS on Information
SP - 1195
EP - 1204
AU - Hideyuki WATANABE
AU - Shigeru KATAGIRI
PY - 1997
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E80-D
IS - 12
JA - IEICE TRANSACTIONS on Information
Y1 - December 1997
AB - In general cases of pattern recognition, a pattern to be recognized is first represented by a set of features and the measured values of the features are then classified. Finding features relevant to recognition is thus an important issue in recognizer design. As a fundamental design framework taht systematically enables one to realize such useful features, the Subspace Method (SM) has been extensively used in various recognition tasks. However, this promising methodological framework is still inadequate. The discriminative power of early versions was not very high. The training behavior of a recent discriminative version called the Learning Subspace Method has not been fully clarified due to its empirical definition, though its discriminative power has been improved. To alleviate this insufficiency, we propose in this paper a new discriminative SM algorithm based on the Minimum Classification Error/Generalized Probabilistic Descent method and show that the proposed algorithm achieves an optimal accurate recognition result, i.e., the (at least locally) minimum recognition error situation, in the probabilistic descent sense.
ER -