Divergence Based Weighting for Information Channels in Deep Convolutional Neural Networks for Bird Audio Detection

In this paper, we address the problem of bird audio detection and propose a new convolutional neural network architecture together with a divergence based information channel weighing strategy in order to achieve improved state-of-the-art performance and faster convergence. The effectiveness of the methodology is shown on the Bird Audio Detection Challenge 2018 (Detection and Classification of Acoustic Scenes and Events Challenge, Task 3) development data set.

[1]  Mark B. Sandler,et al.  Automatic Tagging Using Deep Convolutional Neural Networks , 2016, ISMIR.

[2]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.

[3]  Klaus Riede,et al.  Automatic bird sound detection in long real-field recordings: Applications and tools , 2014 .

[4]  R. A. Leibler,et al.  On Information and Sufficiency , 1951 .

[5]  Hervé Glotin,et al.  Bird detection in audio: A survey and a challenge , 2016, 2016 IEEE 26th International Workshop on Machine Learning for Signal Processing (MLSP).

[6]  Thomas Grill,et al.  Two convolutional neural networks for bird detection in audio signals , 2017, 2017 25th European Signal Processing Conference (EUSIPCO).

[7]  Jian Sun,et al.  Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[8]  Miroslaw Bober,et al.  REMAP: Multi-Layer Entropy-Guided Pooling of Dense CNN Features for Image Retrieval , 2019, IEEE Transactions on Image Processing.

[9]  Josef Kittler,et al.  Delta Divergence: A Novel Decision Cognizant Measure of Classifier Incongruence , 2016, IEEE Transactions on Cybernetics.

[10]  Peter Jancovic,et al.  Automatic Detection and Recognition of Tonal Bird Sounds in Noisy Environments , 2011, EURASIP J. Adv. Signal Process..

[11]  Josef Kittler,et al.  Error sensitivity analysis of Delta divergence - a novel measure for classifier incongruence detection , 2018, Pattern Recognit..

[12]  Dan Stowell,et al.  Automatic large-scale classification of bird sounds is strongly improved by unsupervised feature learning , 2014, PeerJ.

[13]  Vincent M. Stanford,et al.  Bird classification algorithms: theory and experimental results , 2004, 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[14]  Hervé Glotin,et al.  Automatic acoustic detection of birds through deep learning: The first Bird Audio Detection challenge , 2018, Methods in Ecology and Evolution.