Kızılgerdan Kuş Popülasyonu Biyoakustik Kayıtlarının Takibi İçin Kuş Sesi Tanıma Yöntemi Geliştirilmesi

Year 2020, , 1720 - 1731, 30.04.2020

Selim Aras , Seda Üstün Ercan

https://doi.org/10.29130/dubited.569642

Abstract

Bu çalışmada, Kızılgerdan kuş popülasyonuna ait dört alt türün biyoakustik özelliklerinden tespiti için uygun öznitelik ve sınıflandırma yöntemi araştırılmıştır. Özniteliklerin belirlenmesi için Mel Frekansı Kepstrum Katsayıları temel alınmış ve bu katsayılardan istatistiksel parametreler yardımıyla hesaplanabilecek uygun öznitelik araştırması yapılmıştır. Sınıflandırma aşamasında Doğrusal Ayırma Ayıracı, Destek Vektör Makineleri ve k-En Yakın Komşuluk ve Ardışıl İleri Yönlü Öznitelik yöntemleri kullanılmıştır. Sınıflandırıcı parametreleri 10-kat çapraz doğrulama yöntemi ile eğitim setinde belirlenmiştir. Daha sonra, eğitilmiş sınıflandırıcı parametreleri test veri setine uygulanarak sınıflandırma doğruluğu elde edilmiştir. Sonuç olarak, çalışmamızda Mel Frekansı Kepstrum katsayıları temel alınarak hesaplanan ortalama, etkinlik ve karmaşıklık parametreleri k-En Yakın Komşuluk Yöntemi ile sınıflandırıldığında en iyi başarım elde edilmiştir. Önerdiğimiz yöntemin sınıflandırma başarımı eğitim kümesinde %97, test kümesinde ise %94 olarak elde edilmiştir.

Keywords

Kuş sesi sınıflandırma, Mel frekansı kepstrum katsayısı, İstatistiksel parametreler

Developing Bird Song Recognition Method for Monitoring Robin Birds Population Bioacoustics Records

Abstract

In this study, suitable features and classification methods were investigated to determine the four subspecies of Robin birds population from their bioacoustic characteristics. Mel Frequency Cepstrum Coefficients were taken as basis for the determination of the features and a suitable feature search was performed by using statistical parameters from these coefficients. In the classification stage Linear Discriminant Analysis, Support Vector Machines, k-Nearest Neighborhood, and Sequential Forward Feature Selection methods were used. Classifier parameters were determined by 10-fold cross validation method. Then, the classification accuracy was obtained by applying the trained classifier parameters to the test data set. As a result, in our study, the best performance was obtained when the mean, efficiency and complexity parameters, which were calculated based on Mel Frequency Kepstrum coefficients, were classified by k-Nearest Neighborhood Method. The classification performance of the proposed method was obtained 97% in the training set and 94% in the test set.

Keywords

Bird song classification, Mel frequency cepstrum coefficients, Statistical parameters

References

• [1] A. Thakur, V. Abrol, P. Sharma, and P. Rajan, “Local Compressed Convex Spectral Embedding for Bird Species Identification,” The Journal of the Acoustical Society of America, c. 143(6), ss. 3819-3828, 2018.
• [2] Xeno-Canto Veri Seti, May. 24, 2019. [Online]. Erişim: https://www.xeno-canto.org/.
• [3] W. Chu, and D.T. Blumstein, “Noise Robust Bird Song Detection Using Syllable Pattern-Based Hidden Markov Models,” IEEE International Conference on Acoustics, Speech and Signal Processing, 2011, ss. 345-348.
• [4] J. A. Kogan and D. Margoliash, “Automated Recognition of Bird Song Elements from Continuous Recordings Using Dynamic Time Warping and Hidden Markov Models: A Comparative Study,” The Journal of the Acoustical Society of America, 1998, c. 103, s. 4, ss. 2185–2196.
• [5] A. Marini, A. J. Turatti, A. S. Britto, and A. L. Koerich, “Visual Andacoustic Identification of Bird Species,” 2015 IEEE International Conference on Acoustics, Speech and Signal Processing 2015, ss. 2309–2313.
• [6] A. L. McIlraith and H. C. Card, “Birdsong Recognition with DSP and Neural Networks,” IEEE WESCANEX 95. Communications, Power, and Computing. Conference Proceedings,” 1995, c. 2, ss. 409–414.
• [7] A. L. McIlraith and H. C. Card, “Birdsong Recognition Using Backpropagation and Multivariate Statistics,” IEEE Transactions on Signal Processing, c. 199745(11), ss. 2740–2748.
• [8] D. Chakraborty, P. Mukker, P. Rajan, and A. Dileep, “Bird Call Identification Using Dynamic Kernel Based Support Vector Machines and Deep Neural Networks,” in Proceedings of Int. Conf. Mach. Learn. App. 2016, ss. 280–285.
• [9] E. M. Albornoz, L. D. Vignolo, J. A. Sarquis, and E. Leon,“Automatic Classification of Furnariidae Species from the Paranaense Littoral Region Using Speech-Related Features and Machine Learning,” Ecological Informatics, 2017, c. 38, ss. 39–49.
• [10] Priyadarshani, N., Marsland, S., Juodakis, J., Castro, I., and Listanti, V. “Wavelet Filters for Automated Recognition of Birdsong in Long‐Time Field Recordings,” Methods in Ecology and Evolution, 2020, c. 11(3), ss. 403-417.
• [11] D. E. Kroodsma, E. H. Miller, and H. Ouellet, “Acoustic Communication in Birds: Song Learning and Its Consequences,“ Academic, New York, 1982, c. 2.
• [12] Á. Incze, H. B. Jancsó, Z. Szilágyi, A. Farkas, and C. Sulyok, “Bird Sound Recognition Using a Convolutional Neural Network,” 2018 IEEE 16th International Symposium on Intelligent Systems and Informatics, 2018, ss. 295-300.
• [13] J. P. Campbell, “Speaker Recognition: A Tutorial,” Proceedings of the IEEE, 1997, c. 85(9), ss. 1437-1462.
• [14] J. R. Deller, J. H. L Hansen, & J. G. Proakis, “Discrete-Time Processing of Speech Signals,” IEEE Press, Piscataway, N.J, 2000.
• [15] K. Hechenbichler and K. Schliep, “Weighted K-Nearest-Neighbor Techniques and Ordinal Classification Technical Report,” Ludwig-Maximilians-Universit¨at M¨unchen, Institut f¨ur Statistik, 2004.
• [16] J. Koronacki and J. C´wik, “Statistical Learning Systems (in Polish),” Wydawnictwa Naukowo-Techniczne, Warsaw, Poland, 2005.
• [17] Hyeran Byun and Seong-Whan Lee, “Applications of Support Vector Machines for Pattern Recognition: A Survey,” In Proceedings of the First International Workshop on Pattern Recognition with Support Vector Machines, SVM ’02, London, UK, 2002, ss. 213–236.
• [18] C.J.C. Burges, “A Tutorial on Support Vector Machines for Pattern Recognition,” Data Min. Knowl. Discov., 1998, c. 2(2), ss. 121–167.
• [19] Ö. Aydemir, “Ardışıl İleri Yönlü Öznitelik Seçim Algoritmasında Etkin Özniteliklerin Belirlenmesi,” Dicle Üniversitesi Mühendislik Fakültesi Mühendislik Dergisi, c. 8(3), ss. 495-501, 2017.