基于MFCC参数和HMM的低空目标声识别方法研究

A Novel Low Altitude Passive Acoustic Target Identify Approach Research Based on MFCC and HMM

提出了一种战场声目标识别方法,模拟人耳的听觉机理提取更能反应出声音信号动态特征的Mel倒谱系数(MFCC)作为识别战场低空目标的参数;利用隐马尔可夫过程具有很强地表征时变信号的能力来表现声信号随时间变化呈现出的模式演变现象,建立隐马尔可夫模型(HMM);由K-均值聚类得出HMM模型的训练和识别特征向量,识别时设定阈值判定输入的未知声信号。实际数据的分析结果表明了该识别方法的准确性与有效性。

An approach is proposed to identify low altitude passive acoustic target in battlefield. According to the hearing mechanism. MFCC which responses the characteristic of the sound more aggressively is extracted in the system. For the hidden Markov models work bctter in representing the time-variant signal, the HMM models are employed to simulate model change of the sound signals as the rime going. K-means algorithm is used as cluster MFCC to produce training and identifying eigenvector. Simulation results indicate this approach is effective in target recognition.