Cooperative abnormal sound event detection in end-edge-cloud orchestrated systems
- PDF / 2,320,119 Bytes
- 13 Pages / 595.276 x 790.866 pts Page_size
- 85 Downloads / 176 Views
REGULAR PAPER
Cooperative abnormal sound event detection in end‑edge‑cloud orchestrated systems Jingrong Wang1,2 · Kaiyang Liu1,3 · George Tzanetakis1 · Jianping Pan1 Received: 1 July 2020 / Accepted: 17 October 2020 © China Computer Federation (CCF) 2020
Abstract In this paper, we propose a novel cooperative abnormal sound event detection framework for city surveillance in end-edgecloud orchestrated systems. A novel offloading decision-making scheme that leverages hierarchical computational capabilities is proposed to speed up the detection process. The audio pre-processing (feature extraction) and post-processing (sound source localization and sound event classification) can be locally executed or offloaded to the edge or cloud based on the calculation of the so-called communication-to-computation ratio. Furthermore, considering the biased audio information due to source-sensor geometries, a cooperative decision-making algorithm is proposed to aggregate the sound event classification results with adaptive control and ensemble learning. In the audio pre-processing, the log-mel spectrogram and time of arrival information are first extracted from the audio waveform captured by the distributed acoustic sensors and then sent to the computation entity assigned by the offloading scheme. In the audio post-processing, the sound source is localized through least-square minimization. Guided by the localized sound source, the spectrograms are fed into the pre-trained neural networks and then the result aggregation algorithm for further classification. Extensive experiments regarding latency and detection accuracy show the superiority and robustness of the proposed scheme, avoiding the cumulative latency caused by the increased number of sensors while maintaining high detection accuracy. Keywords Sound event detection · Cooperative processing · Audio classification · End-edge-cloud orchestrated systems
1 Introduction In the Internet-of-Things (IoT) era, abnormal sound event detection has played an essential role in city surveillance, especially against gunshot violence. It has been reported that deploying acoustic sensors followed by the audio analysis * Jingrong Wang [email protected] Kaiyang Liu [email protected] George Tzanetakis [email protected] Jianping Pan [email protected] 1
Department of Computer Science, University of Victoria, Victoria, BC V8W 2Y2, Canada
2
Department of Electrical and Computer Engineering, University of Toronto, Toronto, ON M5S 3G4, Canada
3
School of Computer Science and Engineering, Central South University, Changsha 410075, China
of the data they collect for city security can identify unreported gunfire incidents to the police (Shotspotter 2019). This reduces the police response time when compared with the traditional alarm process through 911 calls. Typically, abnormal sound event detection consists of audio feature extraction, sound source localization, and sound event classification (Wang et al. 2018). Advanced technologies such as machine learning algorithms are used for i
Data Loading...
