噪音情境下生态环境声音的分类

Eco-environmental Sounds Classification Under Noise Conditions

提出一种对含有噪音的生态环境声音进行分类的方法.首先,匹配追踪(Matching Pursuit,简称MP)算法可以提取有效信号的时-频特征,减弱噪音的干扰.支持向量机(support Vector Machine,简称SVM)分类器的鲁棒性比较好,所以提出使用SVM基于MP时-频特征建立模型(简称MP-SVM)对含有噪音的生态环境声音进行分类.实验得出MP-SVM可取得较好的分类效果,证明了MP时-频特征和SVM分类器具有较好的抗噪性.

A classification approach for eco-environmental sounds under noise conditions is presented in this paper.Matching pursuit（MP） algorithm is proposed to extract time-frequency features of effective signals,so that it can reduce the interference of noise.In addition,the classification model using support vector machine（SVM） is more robust,so a classification model using MP-based features and SVM（MP-SVM） is proposed.Experimentally,MP-SVM is able to achieve a higher accuracy rate for discriminating eco-environmental sounds under noise conditions.The result shows that MP-based features and SVM classifier have better noise immunity.