AUTOMATIC FEATURE EXTRACTION AND CLASSIFICATION OF CROSSBILL (LOXIA SPP.) FLIGHT CALLS

ABSTRACT In this paper a new method for the automatic classification of bird sounds is presented. Our method is based on acoustic parameters (features) taken from the first harmonic component computed from the sound spectrogram. The features are based on a line segment approximation of the first harmonic component. The final feature vectors, consisting of 16 real numbers, are then classified using a self-organizing map (SOM) neural network. Flight calls of four crossbill species (Loxia spp.) are used as a test example. In the first phase, an unsupervised network was trained and tested using common crossbill L. curvirostra flight calls recorded mainly in the Netherlands. The network was tested using two-barred L. leucoptera, Scottish L. scotica and parrot L. pytyopsittacus crossbill flight calls in the second phase. Finally, the results were validated applying the same network to flight calls of common crossbills and parrot crossbills recorded in Finland. The method automatically separated common crossbill flight calls from those of parrot crossbills. The classification accuracy of the Dutch recordings was 58% in the first phase and 54% in the second phase. The Finnish recordings were classified with 54% accuracy.

[1]  J. G. Groth.  Call matching and positive assortative mating in red crossbills , 1993 .

[2]  J. G. Groth,et al.  Evolutionary Differentiation in Morphology, Vocalizations, and Allozymes Among Nomadic Sibling Species in the North American Red Crossbill (Loxia curvirostra) Complex , 1993 .

[3]  VINCENT M. JANIK,et al.  Pitfalls in the categorization of behaviour: a comparison of dolphin whistle classification methods , 1999, Animal Behaviour.

[4]  Peter K. McGregor,et al.  Census and monitoring based on individually identifiable vocalizations: the role of neural networks , 2002 .

[5]  H. C. Card,et al.  Birdsong recognition using backpropagation and multivariate statistics , 1997, IEEE Trans. Signal Process..

[6]  A. Knox,et al.  The sympatric breeding of Common and Scottish Crossbills Loxia curvirostra and L. scotica and the evolution of crossbills , 2008 .

[7]  B. Ripley,et al.  Pattern Recognition , 1968, Nature.

[8]  Pim Edelaar,et al.  The ecology and evolution of crossbills Loxia spp: the need for a fresh look and an international research programme , 2004 .

[9]  M. Ryan,et al.  Neural networks predict response biases of female túngara frogs , 1998, Proceedings of the Royal Society of London. Series B: Biological Sciences.

[10]  J Placer,et al.  A fuzzy-neural system for identification of species-specific alarm calls of Gunnison's prairie dogs , 2000, Behavioural Processes.

[11]  M. C. Baker,et al.  Population Differentiation in a Complex Bird Sound: A Comparison of Three Bioacoustical Analysis Procedures , 2003 .

[12]  V B Deecke,et al.  Quantifying complex patterns of bioacoustic variation: use of a neural network to compare killer whale (Orcinus orca) dialects. , 1999, The Journal of the Acoustical Society of America.

[13]  Jack P. Hailman,et al.  ANALYSIS OF COMPLEX VARIATION: DICHOTOMOUS SORTING OF PREDATOR-ELICITED CALLS OF THE FLORIDA SCRUB JAY , 1991 .

[14]  Peter M. Clarkson,et al.  Optimal and Adaptive Signal Processing , 1993 .