Recognition of Bumblebee Species by their Buzzing Sound
Mukhiddin Yusupov, Mitja Luštrek, Janez Grad, Matjaz Gams Czech Technical University in Prague – Computer Science Department, Czech Republic & Jozef Stefan Institute – Intelligent Systems Department, Slovenia
Abstract: The goal of our work is to help people to automatically identify the species and worker/queen type of bumblebee based on their recorded buzzing. Many recent studies of insect and bird classification based on their sound have been published, but there is no thorough study that deals with the complex nature of buzzing sound characteristic of bumblebees. In this paper, a database of recorded buzzings of eleven species were preprocessed and segmented into a series of sound samples. Then we applied J48, MLP and SVM supervised classification algorithms on some predetermined sets of feature vectors. For five species the recognition rate was above 80% and for other six species it was above 60%. At the end we consider how to further improve the results.
Keywords: Buzzing Sound, Supervised Classification Recognition of Bumblebee Species by their Buzzing Sound
References:[1] Fagerlund, Seppo. (2007). Bird Species Recognition Using Support Vector Machines. EURASIP Journal on Advances in Signal Processing, vol 2007, Article ID 38637. doi:10.1155/2007/38637
[2] Leqing, Zhu., Zhen, Zhang. (2010). Insect sound recognition based on SBC and HMM. 2010 International Conference on Intelligent Computation Technology and Automation.
[3] Briggs, Forrest., Raich, Raviv., Xiaoli, Z., Fern. Audio Classification of Bird Species: a Statistical Manifold Approach.
[4] Diego, F., Silva, VinÃÂcius, M. A., de Souza, Eamonn Keogh, Daniel, P. W., Ellis. Applying Machine Learning and Audio Analysis Techniques to Insect Recognition in Intelligent Traps.
[5] Ruben Gonzalez. Better than MFCC Audio Classification Features.
[6] Bueno, J. F., Francoy, T. M., Imperatriz-Fonseca, V. L., Saraiva, A. M. Modeling an automated system to identify and classify stingless bees using the wing morphometry – A pattern recognition approach.
[7] Fujisaki, H., Ohno, S. (1996). Analysis and Modeling of Fundemental Frequency Contour of English Utterances. Proc. EUROSPEECH’95. 2, 634-637. Philadelphia.
[8] Erling Wold, Thom Blum., Douglas Keislar, James Wheaton. Content Based Classification, Search, and Retrieval of Audio.
[9] McEnnis, Daniel., McKay, Cory., Fujinaga, Ichiro, Depalle, Philippe. jAudio: A feature extraction library.