SUBSPACE METHOD FOR MINIMUM ERROR PATTERN-RECOGNITION

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 recognitio n is thus an important issue in recognizer design. As a fundamental de sign framework that systematically enables one to realize such useful features, the Subspace Method (SM) has been extensively used in variou s recognition tasks. However. this promising methodological framework is still inadequate. The discriminative power of early versions was no t very high. The training behavior of a recent discriminative version called the Learning Subspace Method has not been fully clarified due t o its empirical definition, though its discriminative power has been i mproved. To alleviate this insufficiency, we propose in this paper a n ew discriminative SM algorithm based on the Minimum Classification Err or/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 probabi listic descent sense.