Sound classification based on spectrogram for surveillance applications

This paper presents an audio event classification algorithm which automatically classifies an audio event as footstep, glass breaking, gunshot or scream mainly for surveillance applications. First, the Gabor feature of the audio spectrogram is extracted, there are two kinds of Gabor features, namely global Gabor feature and local Gabor feature. Then we use Principal Components Analysis (PCA) and Linear Discriminant Analysis (LDA) to compress the feature dimension, finally the K nearest neighbor classifier (KNN) is used to recognize audio events. We carried out extensive experiments on the clean and noisy audio sets. Our results demonstrate that the algorithm is able to guarantee a recall of 96.1% on clean sets and is proved to be more effective than traditional methods.