论文信息 - Exemplar-based Learning in Adaptive Optical Music Recognition System

Exemplar-based Learning in Adaptive Optical Music Recognition System

The learning process of an adaptive optical music recognition system (AOMR) is described here. By combining k-nearest neighbor classifier and genetic algorithm, the system can learn to recognize new music symbols and handwritten music notations, and it also continually improves the accuracy in recognizing these objects. Given the wide range of music notation styles, these are essential characteristics of a music recognizer.

Ichiro Fujinaga

[1] A. Scott,et al. Ann Arbor , 1980 .

[2] Douglas L. Medin,et al. Context theory of classification learning. , 1978 .

[3] Douglas L. Hintzman,et al. "Schema Abstraction" in a Multiple-Trace Memory Model , 1986 .

[4] Stephen K. Reed,et al. Pattern recognition and categorization , 1972 .

[5] Peter E. Hart,et al. Nearest neighbor pattern classification , 1967, IEEE Trans. Inf. Theory.

[6] John H. Holland,et al. Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .