一种优化的舰船基频特征提取算法  

An Optimized Algorithm of Feature Extraction for Vessel Fundamental Frequency

作　　者：刘义海[1] 张效民[1] 赵延安[1] 郭思明[1] 

机构地区：[1]西北工业大学航海学院,陕西西安710072

出　　处：《鱼雷技术》2011年第4期255-259,共5页Torpedo Technology

摘　　要：舰船目标的轴频是表征其目标类别的一种本质特征,有效地提取这种特征可以提高对水中目标的检测能力。本文基于多重自相关的目标推进器噪声的基频分析,利用1D模糊处理的方法求取了基频分布频段中各频点基频出现的置信度,通过对目标噪声基频分析中的线谱分量及线谱分量对应频点的置信度阈值门限实现对舰船目标基频特征的提取,完成对目标的检测识别。利用海上多种型号和多种工况目标的大量实录数据检验结果表明,文中基于多重相关功率谱和信息融合置信度阈值的基频提取方法对水中目标的检测具有良好效果,检测概率达到92.5%。Shaft frequency, which is an essential feature of different vessel targets, possesses a crucial application value to the underwater targets detection. Based on the fundamental frequency analysis of target propeller noise with multiple autocorrelation, we propose a one-dimensional fuzzy judgment method to get the credible degree of the fundamental frequency in each fundamental frequency band. Adopting the component of the linear spectrum and its corresponding credible degree threshold of fundamental frequency, the purpose of extracting target＇s characteristic fundamental frequency is realized to identify the target. The sea trial results show that the proposed fundamental frequency extraction method based on multiple autocorrelation power spectrum and credible degree threshold filtering of information fusion is effective and feasible in underwater target detection with a detection probability of 92.5%.

关 键 词：多重自相关 基频线谱提取 基频置信度阈值 目标检测 

分 类 号：TJ630.2[兵器科学与技术—武器系统与运用工程] TN911.7[电子电信—通信与信息系统]