An intelligent air-noise recognition system is described that uses pattern recognition techniques to distinguish noise signatures of five different types of acoustic sources, including jet planes, propeller planes, a helicopter, train, and wind turbine. Information for classification is calculated using the power spectral density and autocorrelation taken from the output of a single microphone. Using this system, as many as 90 percent of test recordings were correctly identified, indicating that the linear discriminant functions developed can be used for aerospace source identification.
A smart pattern recognition system for the automatic identification of aerospace acoustic sources
01.04.1989
Aufsatz (Konferenz)
Keine Angabe
Englisch
Pattern recognition system for automatic identification of acoustic sources
Tema Archiv | 1991
|