Masters Thesis Defense “Multiple Drone Detection and Acoustic Scene Classification with Deep Learning” By Hari Charan Vemula

Committee:  Drs. John Gallagher, Advisor, Mateen Rizki, and Thomas Wischgoll

ABSTRACT:

Classification of environmental scenes and detection of events in one's environment from audio signals enables to create, better-planning agents, intelligent navigation systems, pattern recognition systems, and audio surveillance systems. This thesis will explore the use of Convolutional Neural Networks(CNN'S) with Spectrograms and raw audio waveforms as inputs, and Deep Neural Networks with hand engineered features extracted from large-scale feature extraction schemes to identify the acoustic scenes and events. The first part focuses on building an audio pattern recognition system capable of detecting the presence zero, one, or two DJI phantoms in the scene within the range of a stereo microphone. The ability to distinguish the presence multiple UAV's could be used to augment information from other sensors less making of cleanly making such a determination. The second part of the thesis focuses on building an acoustic scene detector to Task1a in the DCASE2018 challenge. In both cases, this document will explain the preprocessing techniques, CNN and DNN architectures, data augmentation methods including the use of Generative Adversarial Network’s(GAN's), and performance results compared to existing benchmarks when available. This thesis will conclude with a discussion of how one might expand the techniques in the construction of commercial off the shelf audio scene classifier for multiple UAV detections.