论文信息 - Detection of Abnormal Sound Using Multi-stage GMM for Surveillance Microphone

Detection of Abnormal Sound Using Multi-stage GMM for Surveillance Microphone

We developed a system that detects abnormal sound from sound signal observed by a surveillance microphone. Our system learns the “normal sound” from observation of the microphone, and then detects sounds never observed before as “abnormal sounds.” To this end, we developed a technique that uses multiple GMMs for modeling different levels of sound events efficiently. We also consider how to determine thresholds of GMM switching and event detection.As a result, we obtained almost same detection performance using the percentile method to the manually optimized GMMs. Besides, we exploited the segment-based feature, which gave the best result among all methods.

[1] Shuming Tang,et al. A survey of vision-based automatic incident detection technology , 2005, IEEE International Conference on Vehicular Electronics and Safety, 2005..

[2] Kurumatani Koichi,et al. A Security Monitoring System of Detecting Unusual Sounds and Hazardous Situations by Sound Environment Measurement Using Microphone Arrays , 2008 .

[3] Sergio A. Velastin,et al. Detection of Potentially Dangerous Situations involving Crowds using Image Processing , 1999, IIA/SOCO.

[4] Zeno Geradts,et al. Forensic audio and visual evidence 2004-2007 : A Review , 2007 .

[5] Yongjun Kim,et al. Sound Detection as an Aid to Increase Detectability of CCTV in Surveillance System , 2007, HCI.

[6] Nikos Fakotakis,et al. On acoustic surveillance of hazardous situations , 2009, 2009 IEEE International Conference on Acoustics, Speech and Signal Processing.

[7] Mark Hasegawa-Johnson,et al. Acoustic fall detection using Gaussian mixture models and GMM supervectors , 2009, 2009 IEEE International Conference on Acoustics, Speech and Signal Processing.