针对声呐在搜潜过程中目标识别正确率提升的问题，分析声呐目标特征及其获取手段，研究声呐目标特征提取、分析和识别方法，提出一种多信息源融合的声呐目标综合识别方法。综合运用声呐探测回波特征信息、频谱信息、音频信息、目标运动要素信息、雷达信息、AIS信息等信息进行联合识别，深层次挖掘目标特征信息，通过图谱特性、频域特征、听音识别、运动要素等多维识别，将孤立的、碎片的数据转化成信息优势，形成标准的声呐目标综合识别使用流程，从而提高目标识别正确率。
Aiming at the problem that the accuracy of target recognition is improved during sonar search, the characteristics of sonar target and its acquisition means are analyzed. The method of feature extraction, analysis and recognition of sonar target is studied. A sonar target synthesis with multi-information source fusion is proposed. The identification method comprehensively uses the sonar detection echo characteristic information, spectrum information, audio information, target motion element information, radar information, AIS information and other information for joint recognition, deep mining target feature information, through map characteristics, frequency domain characteristics, Multi-dimensional recognition such as listening recognition and motion elements converts the isolated and fragmented data into information advantages, forming a standard sonar target comprehensive recognition and use process, thereby improving the target recognition accuracy rate.
2019,41(8): 127-130 收稿日期：2019-02-14
DOI：10.3404/j.issn.1672-7649.2019.08.025
分类号：TB556
作者简介：梁民赞(1981-),男,工程师,主要研究方向为水声探测
参考文献：
[1] 闫福旺. 水声对抗技术[M]. 北京:海洋出版社, 2003.
[2] A.D.WAITE. 实用声纳工程(第三版)[M]北京:电子工业出版社, 2004.
[3] 孙军平, 林建恒, 江鹏飞, 等. 舰艇水下辐射噪声谱特征传播仿真分析[J]. 声学技术, 2017, 36(5)
[4] 杨露菁, 余华. 多源信息融合理论及与用[M]. 北京:北京邮电大学出版社, 2006.
[5] 李启虎. 数字式声纳设计原理[M]. 合肥:安徽教育出版社, 2002.
