Empirical Study of Drone Sound Detection in Real-Life Environment with Deep Neural Networks, EUSIPCO17
This project contains Python experimental implementation used for research paper published in EUSIPCO17.
Original paper: https://arxiv.org/abs/1701.05779
Please use the following citation:
@inproceedings{jeon2017empirical,
author = {Jeon, Sungho and Shin, Jong-Woo and Lee, Young-Jun and Kim, Woong-Hee and Kwon, YoungHyoun and Yang, Hae-Yong},
title = {Empirical Study of Drone Sound Detection in Real-Life Environment with Deep Neural Networks},
year = 2017,
booktitle = {Signal Processing Conference (EUSIPCO), 2017 25th European},
organization={IEEE}
}
Contact person: Sungho Jeon, sungho.jeon@h-its.org
dd_rnn.py-- Experiments using RNN (Tensorflow 0.12)dd_cnn.py-- Experiments using CNN (Tensorflow 0.12)dd_gmm.py-- Experiments using GMM (Scikit-learn 0.18)
- Software dependencies
- Python 2.7
- Tensorflow 0.12
- Scikit-learn 0.18
- Librosa 0.4.3
Please note that our experimental audio data cannot be shared due to security contact of research project. However, I described how to augment audio dataset in the paper.