用于安全监控系统的音频事件检测研究

Audio Events Detection for Surveillance System

针对视频监控过于依赖图像信息,在黑暗条件下或视线盲区无法进行及时、有效监控的问题,本文提出一种可以用于辅助安全监控的音频事件检测系统,以枪击声和尖叫声为关键事件,把过零率、短时能量、子带能量比和Mel倒谱系数作为音频特征,基于SVM分类器设计实现了一种多级分类系统。结果表明,该系统可以有效地检测出两种关键事件,最优识别率达90%。

Public surveillance system relies on image information to a large extent,therefore,places in dark environ?ments or blind areas of surveillance cameras would not get effective real-time surveillance.To solve this problem,this paper proposed an audio events detection system to assist video-based public safety surveillance.The systemtook the shooting and screaming as the key events,and defined Zero Crossing Rate(ZCR),Short Time Energy(STE),Sub Band Energy Ratio(SBER)and MFCC as audio features.Then a hierarchical SVM-based classification system isimplemented to classify different types of audio events.According to experiment results,the proposed audio events de?tection system can effectively classify gunshots and screaming sound from environmental noise with an optimal classi?fication accuracy of90%.