Page 109 - 《应用声学》2021年第5期

第 40 卷 第 5 期 卢佳敏等： 基于 DEMON 谱和 LSTM 网络的水下运动目标噪声基频检测 753

Cheng Yusheng, Qiu Jiaxing, Liu Zhen. Challenges and Shock, 2018, 37(16): 57–61.
prospects of underwater acoustic passive target recogni- [14] 吴国清, 王美刚, 陈耀明. 水声波导中包络线谱强度数值预
tion technology[J]. Journal of Applied Acoustics, 2019, 报 [J]. 声学学报, 2012, 37(3): 432–439.
38(3): 653–659. Wu Guoqing, Wang Meigang, Chen Yaoming. Numerical
[2] Nielsen R O. Cramer-rao lower bounds for sonar broad- prediction of envelope line specturm intensity in under-
band modulation parameters[J]. IEEE Journal of Oceanic water acoustic waveguide[J]. Acta Acustica, 2012, 37(3):
Engineering, 1999, 24(2): 285–290. 432–439.
[3] Lourens J G, du Prcez J A. Passive sonar ml estimator [15] 蒋国健, 林建恒, 孙军平, 等. 海洋信道随机性引起的船舶噪
for ship propeller speed[J]. IEEE Journal of Oceanic En- 声线谱衰减 [J]. 声学学报, 2015, 40(1): 170–177.
gineering, 1998, 23(3): 448–453. Jiang Guojian, Lin Jianheng, Sun Junping, et al. At-
[4] Song L, Yang D, Zhang H. Experimental study of under- tenuation of ship noise line-spectrum caused by the ran-
water propeller low-frequency noise flied[C]. IEEE/OES domness of ocean channel[J]. Acta Acustica, 2015, 40(1):
China Ocean Acoustics, 2016. 170–177.
[5] 杨宏, 李亚安, 李国辉. 基于集合经验模态分解的舰船辐射噪 [16] 刘振, 邱家兴, 程玉胜. 深度神经网络在螺旋桨叶片数识别中
声能量分析 [J]. 振动与冲击, 2015, 34(16): 55–59. 的应用 [J]. 声学技术, 2019, 38(3): 459–463.
Yang Hong, Li Ya’an, Li Guohui. Energy analysis of ship Liu Zhen, Qiu Jiaxing, Cheng Yusheng. Application of
radiated noise based on ensemble empirical mode decom- deep neural network in blade-number recognition of ship
position[J]. Journal of Vibration and Shock, 2015, 34(16): propeller[J]. Technical Acoustics, 2019, 38(3): 459–463.
55–59. [17] 朱可卿, 田杰, 黄海宁, 等. 基于深度学习的船舶辐射噪声识
[6] Pollara A, Sutin A, Salloum H. Improvement of the detec- 别研究 [J]. 应用声学, 2018, 37(1): 238–245.
tion of envelope modulation on noise (DEMON) and its Zhu Keqing, Tian Jie, Huang Haining, et al. Ship-radiated
application to small boats[C]. OCEANS 2016 MTS/IEEE noise recognition research based deep learning[J]. Journal
Monterey. IEEE, 2016. of Applied Acoustics, 2018, 37(1): 238–245.
[7] 殷敬伟, 惠俊英, 姚直象, 等. 基于 DEMON 线谱的轴频提取 [18] 王鹏. 基于深度神经网络的水中目标识别研究 [D]. 哈尔滨:
方法研究 [J]. 应用声学, 2005, 24(5): 369–374. 哈尔滨工程大学, 2018.
Yin Jingwei, Hui Junying, Yao Zhixiang, et al. Extraction [19] Artusi E, Chaillanb F. Automatic recognition of underwa-
of shaft frequency based on the DEMON line spectrum[J]. ter acoustic signature for naval applications[C]. Maritime
Journal of Applied Acoustics, 2005, 24(5): 369–374. Situational Awareness Workshop, 2019.
[8] 童峰, 陆佶人, 方世良. 水声目标识别中一种轴频提取方法 [J]. [20] 陈越超, 徐晓男. 基于降噪自编码器的水中目标识别方法 [J].
声学学报, 2004, 29(4): 398–402. 声学与电子工程, 2018(1): 30–33.
Tong Feng, Lu Jiren, Fang Shiliang. A method to estimate [21] 张少康, 王超, 田德艳, 等. 长短时记忆网络水下目标噪声智
the propeller shaft rate in underwater acoustic vessel clas- 能识别方法 [J]. 舰船科学技术, 2019, 41(23): 181–185.
sification[J]. Acta Acustica, 2004, 29(4): 398–402. Zhang Shaokang, Wang Chao, Tian Deyan, et al. Intelli-
[9] 骆国强, 尚金涛, 杨柳. 一种时频综合的 DEMON 谱融合方 gent recognition of underwater target noise based on long
法 [C]. 2016’ 中国西部声学学术交流会论文集, 2016. short-term memory networks[J]. Ship Science and tech-
[10] 高鑫, 程玉胜. 舰船螺旋桨轴频估计中线谱要素提取算法 [J]. nology, 2019, 41(23): 181–185.
应用声学, 2010, 29(5): 443–448. [22] 张少康, 王超, 孙芹东. 基于多类别特征融合的水声目标噪声
Gao Xin, Cheng Yusheng. A line spectrum element ex- 识别分类技术 [J]. 西北工业大学学报, 2020, 38(1): 366–376.
traction algorithm in ship-prpeller-shaft frequency esti- Zhang Shaokang, Wang Chao, Sun Qindong. Underwa-
mations[J]. Journal of Applied Acoustics, 2010, 29(5): ter target noise recognition and classification technology
443–448. based on multi-classes feature fusion[J]. Journal of North-
[11] Hanson D, Antoni J, Brown G, et al. Cyclostationarity for western Polytechnical University, 2020, 38(1): 366–376.
passive underwater detection of propeller craft: a develop- [23] Tang Z, Kanu J, Manocha D, et al. Regression and clas-
ment of DEMON processing[J]. Proceedings of Acoustics, sification for direction-of-arrival estimation with convolu-
2008: 1–6. tional recurrent neural networks[C]. Interspeech, 2019.
[12] 白敬贤, 高天德, 夏润鹏. 基于 DEMON 谱信息提取算法的目 [24] van den Oord A, Kalchbrenner N, et al. Pixel recurrent
标识别方法研究 [J]. 声学技术, 2017, 36(1): 88–92. neural networks[C]. International Conference on Machine
Bai Jingxian, Gao Tiande, Xia Runpeng. Target recog- Learning, 2016.
nition based on the information extraction algorithm of [25] Gonzalez S, Brookes M. PEFAC—A pitch estimation algo-
DEMON spectrum[J]. Technical Acoustics, 2017, 36(1): rithm robust to high levels of noise[J]. IEEE ACM Trans-
88–92. actions on Audio Speech and Language Processing, 2014,
[13] 杨日杰, 郑晓庆, 韩建辉, 等. 基于序列匹配的螺旋桨轴频自 22(1): 518–530.
动提取方法 [J]. 振动与冲击, 2018, 37(16): 57–61. [26] Lu J, Song S, Hu Z, et al. Fundamental frequency detec-
Yang Rijie, Zheng Xiaoqing, Han Jianhui, et al. An au- tion of underwater acoustic target using DEMON spec-
tomatic extraction method of propeller shaft frequency trum and CNN network[C]. IEEE International Confer-
based on sequence matching[J]. Journal of Vibration and ence on Unmanned Systems, 2020.