Ship-Radiated Noise Analysis Based on the Gammatone Frequency Cepstrum Coefficient
-
摘要: 舰船辐射噪声的声学特征提取对目标训练和识别有着重要影响。文中提出一种基于Gammatone频率倒谱系数(GFCC)的特征分析方法: 以目标特征提取方法——Mel频率倒谱系数(MFCC)算法作为比照组, 针对小型低速船只、小型高速船只及大型船只三大类目标, 在2种不同水声环境中提取的5 122个样本进行了分类识别比对试验。试验结果表明, 2种算法的目标识别率均大于80%, 且GFCC在海洋复杂声环境中的识别率显著高于MFCC, 并对高频目标更敏感。说明GFCC算法与标准的MFCC算法相比, 在海洋等强干扰环境下具有更好的抗噪性和更高的快速目标识别率。
-
关键词:
- 水下目标识别 /
- 舰船辐射噪声 /
- 特征提取 /
- Gammatone频率倒谱系数 /
- Mel频率倒谱系数
Abstract: Acoustic feature extraction of ship-radiated noise has a major effect on target training and recognition. This research proposes a feature analysis method based on the gammatone frequency cepstrum coefficient(GFCC). The method uses the typical target feature extraction method——Mel frequency cepstrum coefficient(MFCC) algorithm for comparison and uses 5 122 samples with three types of targets, namely, small low-speed, small high-speed, and large vessels. The samples are extracted in two different underwater acoustic environments to generate a classification identification comparison test. Results show that the recognition rate of the two algorithms is greater than 80%. However, the recognition rate of GFCC in an ocean complex acoustic environment is found to be significantly higher than that of MFCC and is more sensitive to high-frequency targets. These results show that the GFCC algorithm has better noise resistance and a higher recognition rate for fast targets in oceanic and other strong interference environments as compared with the standard MFCC algorithm. -
[1] 徐及, 黄兆琼, 李琛, 等.深度学习在水下目标被动识别中的应用进展[J].信号处理, 2019, 35(9): 1460-1475.Xu Ji, Huang Zhao-qiong, Li Chen, et al. Advances in Underwater Target Passive Recognition Using Deep Learning[J]. Journal of Signal Processing, 2019, 35(9): 1460-1475. [2] 方世良, 杜栓平, 罗昕炜, 等. 水声目标特征分析与识别技术[J]. 中国科学院院刊, 2019, 34(3): 297-305.Fang Shi-liang, Du Shuan-ping, Luo Xin-wei, et al. Development of Underwater Acoustic Target Feature Analysis and Recognition Technology[J]. Bulletin of the Chinese Academy of Sciences, 2019, 34(3): 297-305. [3] 杜雪, 廖泓舟, 张勋. 基于深度卷积特征的水下目标智能识别方法[J]. 水下无人系统学报, 2019, 27(3): 260-265.Du Xue, Liao Hong-zhou, Zhang Xun. Underwater Target Recognition Method Based on Deep Convolution Feature[J]. Journal of Unmanned Undersea Systems, 2019, 27(3): 260- 265. [4] 孔晓鹏, 姚直象, 胡金华, 等. 舰船辐射噪声MFCC特征分析与分类识别[J]. 声学技术, 2019, 38(5): 155-156. [5] 程锦盛, 杜选民, 周胜增, 等. 基于目标MFCC特征的监督学习方法在被动声呐目标识别中的应用研究[J]. 舰船科学技术, 2018, 40(9): 116-121.Cheng Jin-sheng, Du Xuan-min, Zhou Sheng-zeng, et al. Application of the MFCC Feature Based Supervised Learning Method in Passive Sonar Target Recognition[J]. Ship Science and Technology, 2018, 40(9): 116-121. [6] 吴姚振, 杨益新, 田丰, 等.基于Gammatone频率离散小波系数的水下目标鲁棒识别[J].西北工业大学学报, 2014, 32(6): 906-911.Wu Yao-zhen, Yang Yi-xin, Tian Feng, et al. Robust Underwater Target Recognition Based on Gammatone Frequency Discrete Wavelet Coefficients(GFDWC)[J]. Journal of Northwestern Polytechnical University, 2014, 32(6): 906-911. [7] 周萍, 沈昊, 郑凯鹏.基于MFCC与GFCC混合特征参数的说话人识别[J]. 应用科学学报, 2019, 37(1): 24-32.Zhou Ping, Shen Hao, Zheng Kai-peng. Speaker Recognition Based on Combination of MFCC and GFCC Feature Parameters[J]. Journal of Applied Sciences, 2019, 37(1): 24-32. [8] 刘雨柔, 张雪英, 陈桂军, 等. VMD改进GFCC的情感语音特征提取[J]. 计算机工程与设计, 2020, 41(8): 2265-2270.Liu Yu-rou, Zhang Xue-ying, Chen Gui-jun, et al. Feature Extraction of Emotional Speech Based on Improved GFCC with VMD[J]. Computer Engineering and Design, 2020, 41(8): 2265-2270. [9] Jiang J, Shi T, Huang M, et al. Multi-Scale Spectral Feature Extraction for Underwater Acoustic Target Recognition[J]. Measurement, 2020, 166: 108227. [10] 张少康, 田德艳. 水下声目标的梅尔倒谱系数智能分类方法[J]. 应用声学, 2019, 38(2): 267-272.Zhang Shao-kang, Tian De-yan. Intelligent Classification Method of Mel Frequency Cepstrum Coefficient for Un-derwater Acoustic Targets[J]. Applied Acoustics, 2019, 38(2): 267-272. [11] 曾向阳. 水下智能目标识别[M]. 北京: 国防工业出版社, 2016: 8-15.
点击查看大图
计量
- 文章访问数: 195
- HTML全文浏览量: 3
- PDF下载量: 135
- 被引次数: 0