An intelligent recognition system was designed using pattern recognition techniques to distinguish the noise signatures of five different types of acoustic sources. Information for classification was calculated from the power spectral density and autocorrelation taken from the output of a single microphone. The system included a training step where it learned to distinguish the sources and automatically selected descriptive quantities for optimal classification performance. Information learned in training was stored and used to identify recordings of the five types of sources presented during testing. The results of testing indicate the current optimal design could correctly identify 90 % of the recordings. The ability of the system to identify noise corrupted signatures is studied as well as its ability to identify recordings not used in training.
A smart pattern recognition system for the automatic identification of aerospace acoustic sources
Leistungsfaehiges Mustererkennungssystem zur automatischen Identifizierung von Schallquellen des Luftverkehrs
AIAA-Papers ; Apr ; 1-9
1988
9 Seiten, 10 Bilder, 12 Quellen
Conference paper
English
Pattern recognition system for automatic identification of acoustic sources
Tema Archive | 1991
|