JU Donghao1,2,3,LI Yu1,3,WANG Yujie1,2,3,ZHANG Chunhua1,3
1. Institute of Acoustics, Chinses Academy of Sciences, Beijing 100190, China;
2. University of Chinese Academy of Sciences, Beijing 100039, China;
3. Key Laboratory of Science and Technology on Advanced Underwater Acoustic Signal Processing, Chinese Academy of Science, Beijing 100190, China
Abstract:The traditional feature extraction algorithm relies on the prior knowledge of the algorithm designer. Because it does not have the advantage of highlighting big data, the classification accuracy in practical application is poor and the generalization ability for different application scenarios is also obviously insufficient. In this paper, the deep learning algorithm is used for feature extraction of ship radiated noise, and a large number of classless data are fully utilized. The stack sparse self-encoder algorithm is used to train the feature extraction neural network, and the Softmax classifier algorithm is used to fine-tune the parameters of the neural network by using class-based data. By comparing with principal component analysis algorithm, linear discriminant analysis algorithm and local linear embedding algorithm, it can be seen that the ssdae-softmax algorithm proposed in this paper can extract more discriminative features from ship radiated noise and improve the classification and recognition accuracy to some extent.
[1] 吴国清,李靖,陈耀明,袁毅,陈岳.舰船噪声识别(Ⅰ)总体框架、线谱分析和提取[J].声学学报,1998(05):394-400.
WU Guoqing, LI Jing, CHEN Yaoming, et al. Ship radiated- noise recognition (Ⅰ) the overall framework, analysis and extraction of line-spectrum[J].Acta Acustica, 1998(05):394-400.
[2] 章新华,王骥程,林良骥.基于小波变换的舰船辐射噪声特征提取[J].声学学报,1997(02):139-144.
ZHANG Xinhua, WANG Jicheng, LIN Liangji. Feature extraction of ship radited noises based on wavelet transform[J]. Acta Acustica, 1997(02):139-144.
[3] Tucker S, Brown G J. Modelling the auditory perception of size, shape and material: Applications to the classification of transient sonar sounds[C]//Audio Engineering Society Convention 114. Audio Engineering Society, 2003.
[4] 张岩,尹力.主成分分析在舰船辐射噪声分类识别中的应用[J].应用声学,2009,28(01):20-26.
ZHANG Yan, YIN Li. Application of principal component analysis to ship-radiated noise classification and recognition[J]. Journal of applied Acoustics, 2009,28(01):20-26.
[5] 杨宏晖,申昇,姚晓辉,韩振.用于水声目标特征学习与识别的混合正则化深度置信网络[J].西北工业大学学报,2017,35(02):220-225.
YANG Honghui, SHEN Sheng, YAO Xiaohui, et al. Underwater Acoustic Target Feature Learning and Recognition using
Hybrid Regularization Deep Belief Network[J]. Journal of Northwestern Polytechnical University, 2017,35(02):220-225.
[6] Williams D P. Underwater target classification in synthetic aperture sonar imagery using deep convolutional neural networks[C]//2016 23rd international conference on pattern recognition (ICPR). Cancún :IEEE, 2016: 2497-2502.
[7] Ogasawara E, Martinez L C, De Oliveira D, et al. Adaptive normalization: A novel data normalization approach for non-stationary time series[C]//The 2010 International Joint Conference on Neural Networks (IJCNN). Barcelona :IEEE, 2010: 1-8.
[8] Vincent P, Larochelle H, Bengio Y, et al. Extracting and composing robust features with denoising autoencoders[C]//Proceedings of the 25th international conference on Machine learning. Helsinki :ICML,2008: 1096-1103.
[9] Xing C, Ma L, Yang X. Stacked denoise autoencoder based feature extraction and classification for hyperspectral images[J]. Journal of Sensors, 2016, 2016.
[10] Masci J, Meier U, Cireşan D, et al. Stacked convolutional auto-encoders for hierarchical feature extraction[C]//International conference on artificial neural networks. Springer, Berlin, Heidelberg, 2011: 52-59.
[11] Cao X, Zhang X, Yu Y, et al. Deep learning-based recognition of underwater target[C]//2016 IEEE International Conference on Digital Signal Processing (DSP). IEEE, 2016: 89-93.
[12] Amaral T, Silva L M, Alexandre L A, et al. Using different cost functions to train stacked auto-encoders[C]//2013 12th Mexican international conference on artificial intelligence. Mexico City: IEEE, 2013: 114-120.
[13] Marchi E, Vesperini F, Squartini S, et al. Deep recurrent neural network-based autoencoders for acoustic novelty detection[J]. Computational intelligence and neuroscience, 2017, 2017.
[14] Bianco M J, Gerstoft P, Traer J, et al. Machine learning in acoustics: Theory and applications[J]. The Journal of the Acoustical Society of America, 2019, 146(5): 3590-3628.
[15] Goodfellow I, Bengio Y, Courville A. Deep learning[M]. MIT press, 2016.
[16] 李宏坤,郝佰田,代月帮,杨蕊.基于压缩感知和加噪堆栈稀疏自编码器的铣刀磨损程度识别方法研究[J].机械工程学报,2019,55(14):1-10.
LI Hongkun, HAO Baitian, DAI Yuebang, et al. Wear Status Recognition for Milling Cutter Based on Compressed Sensing and Noise Stacking Sparse Auto-encoder[J]. Journal of Mechanical Engineering,2019,55(14):1-10.
[17] 吴国清,李靖,李训诰,陈耀明,袁毅.舰船噪声识别(Ⅳ)──模糊神经网络识别[J].声学学报,1999(03):275-280.
WU Guoqing, LI Jing, LI Xunhao, et al. Ship radiated-noise recognition using fuzzy neural network[J]. Acta Acustica. 1999(03):275-280.
[18] Kamal S, Mohammed S K, Pillai P R S, et al. Deep learning architectures for underwater target recognition[C]//2013 Ocean Electronics (SYMPOL). Kochi :IEEE, 2013: 48-54.
[19] Premus V E, Evans M E, Abbot P A. Machine learning-based classification of recreational fishing vessel kinematics from broadband striation patterns[J]. The Journal of the Acoustical Society of America, 2020, 147(2): EL184-EL188.
[20] 李琛,黄兆琼,徐及,郭新毅,宫在晓,颜永红.使用深度学习的多通道水下目标识别[J].声学学报,2020,45(04):506-514.
LI Chen, HUANG Zhaoqiong, XU Ji, et al. Multi-channel underwater target recognition using deep learning[J]. The Journal of the Acoustical Society of America, 2020, 45(04):506-514.
[21] 李余兴,李亚安,陈晓,蔚婧.基于VMD和中心频率的舰船辐射噪声特征提取方法研究[J].振动与冲击,2018,37(23):213-218+285.
LI Yuxing, LI Yaan, CHEN Xiao,et al. Ships’ radiated noise feature extraction based on EEMD[J]. Journal of vibration and shock, 2018,37(23):213-218+285.