生猪咳嗽声识别与定位方法的研究

Research on recognition and localization of porcine cough sounds

为了及时发现并定位咳嗽生猪,从而预防生猪呼吸道传染疾病的传播,试验设计了一种可以对生猪咳嗽声进行识别并定位的方法。首先通过麦克风阵列采集声音,并对声音信号进行预处理;然后提取声音信号的梅尔频率倒谱系数及其一阶差分,采用支持向量数据描述模型识别生猪咳嗽声;最后利用相位变换算法计算麦克风信号之间的到达时间差值,并结合麦克风的几何位置得到声源位置估计值。结果表明:生猪咳嗽声识别方法的总识别率可达87.3%,定位方法得到的平均距离误差为0.52 m。说明所采用的生猪咳嗽声识别与定位方法可有效地识别并定位生猪咳嗽声,为生猪现代福利化养殖提供技术指导。

To prevent the spread of respiratory infections among pigs as well as detect and locate coughing pigs in time,this article proposes a method to identify and locate coughing pigs.First step is to collect the sound through the microphone array,then the sound is preprocessed.Second step is extracting the Mel-frequency cepstrum coefficient of the sound and its first order difference,thenusing the Support Vector Data Description model to identify the porcine cough.The final step is using the Phase Transformation algorithm to calculate the Time Difference of Arrival between the microphone signals.Furthermore,the geometric position of the microphones iscombinedto obtain the estimated source position.The results show that the total recognition rate of the porcine cough can reach 87.3%,and the average distance error obtained by the locating method is 0.52 m.It shows that the method used in this paper can effectively identify and locate the porcine coughand provide technical guidance for the modern welfare farming of pigs.