Related references
Note: Only part of the references are listed.
Article
Engineering, Mechanical
Yuxing Li et al.
Summary: Recently, a new method called coded dispersion entropy (CDE) is proposed to improve the noise immunity by coding and partitioning the dispersion patterns in dispersion entropy (DE). The simplified CDE (SCDE) is also proposed to reduce computational consumption while maintaining effectiveness. Simulation experiments demonstrate the advantages of SCDE and CDE in detecting nonlinear dynamic changes, and real-world experiments show that SCDE has better performance in medical diagnosis, fault diagnosis, and signal classification.
NONLINEAR DYNAMICS
(2023)
Article
Engineering, Marine
Yuxing Li et al.
Summary: This paper uses the snake optimizer (SO) to optimize SloEn and proposes a new entropy indicator, named SO SloEn (SO-SloEn). Then, a novel adaptive S-RNS feature extraction method is put forward in combination with the successive variational mode decomposition (SVMD) and SO-SloEn, which solves the parameter selection problem of variational mode decomposition (VMD). The effectiveness of the proposed method is verified by two S-RNS cases, and the results indicate that the recognition rate of the proposed method is always the highest compared with other decomposition algorithms and other entropy indicators under the condition of extracting different numbers of IMFs; moreover, the highest recognition rate can reach more than 92% in both cases.
Article
Acoustics
Honghui Yang et al.
Summary: This paper proposes a new recognition method based on deep learning and template matching for underwater acoustic target open set recognition. The method includes feature extraction, template establishment, and template matching. Experimental results show that the proposed method achieves excellent recognition accuracy.
Article
Automation & Control Systems
Lichao Meng et al.
Summary: This paper proposes a new perspective for handling unsupervised domain adaptation, which maximizes the knowledge of the target domain by observing the source domain through a deep adversarial network. The method, named Cross-domain Mutual Information Adversarial Maximization (CMIAM), outperforms previous state-of-the-arts on various unsupervised domain adaptation tasks on multiple datasets.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2022)
Article
Acoustics
Qinggang Sun et al.
Summary: This paper proposes a deep learning-based single-channel multitarget underwater acoustic signal recognition method that can recognize an unknown number of targets in a specified category. The experimental results show that the method can effectively recognize synthetic multitarget ship signals when the magnitude STFT spectrum, complex-valued STFT spectrum, and log-mel spectrum are used as network inputs.
JOURNAL OF THE ACOUSTICAL SOCIETY OF AMERICA
(2022)
Article
Chemistry, Analytical
Xiaorui Shao et al.
Summary: This article proposes a domain adaptive and lightweight framework for fault diagnosis based on 1D-CNN, which can extract features with robustness and domain invariance through CORAL processing to minimize domain shifts, effectively improving FD performance.
Article
Engineering, Multidisciplinary
Zhuodong Yang et al.
Summary: This paper proposes a denoising method using the least mean square (LMS) algorithm for percussion acoustic signals collected in the marine environment. The method is validated by experiments and shows excellent performance in denoising the raw signals with a small error in the estimated primary peak frequency. This study demonstrates the broad potential for the method to be applied towards damage detection for underwater structures.
Article
Acoustics
Guoli Song et al.
Summary: We proposed a machine learning-based underwater noise classification method that extracted five underwater noise features from three domains and classified underwater noise using SVM and CNN methods. The results showed that NL and PSD features achieved better classification performance under different SNR conditions, and the CNN method outperformed the SVM method for underwater noise classification.
Article
Acoustics
Junhao Li et al.
Summary: This paper proposes a deep learning model for perceiving the timbre of underwater acoustic target radiated noise, which can extract stable line spectrum features and achieve a relatively high recognition accuracy.
Article
Acoustics
Qi Zhang et al.
Summary: This paper proposes an integrated neural network method for underwater acoustic target recognition, aiming to enhance recognition accuracy and noise robustness through feature fusion learning. The method extracts multiple features of underwater acoustic signals and uses them as input for training the network, optimizing recognition performance through adjustment of the neural network weight coefficients.
Article
Computer Science, Artificial Intelligence
Muhammad Irfan et al.
Summary: Underwater acoustic classification is a challenging problem due to high background noise and complex sound propagation patterns in the sea environment. Researchers have constructed and presented an underwater acoustic dataset named DeepShip to evaluate existing algorithms and benefit future research. They also conducted a comprehensive study on various machine learning and deep learning algorithms using the dataset.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Chemistry, Analytical
Gang Hu et al.
Summary: A new deep neural network model for underwater target recognition is proposed in this paper, which can automatically extract features from the raw data of ship-radiated noise. Evaluation using measured data shows that the features extracted from the model have good characteristics of intra-class aggregation and inter-class separation. Verification with cross-folding model enhances the generalization ability of the model.
Article
Computer Science, Artificial Intelligence
Jianfei Yang et al.
Summary: This paper proposes a domain adaptation framework called AD3, which combines adversarial and statistical domain alignment to reduce training instability and manual hyper-parameter tuning. Experimental results demonstrate that the method outperforms existing state-of-the-art methods on domain adaptation benchmarks.
Article
Acoustics
Xu Xiao et al.
Summary: In this study, an attention-based neural network (ABNN) is proposed for underwater acoustic target recognition. The ABNN showed high accuracy in target detection and recognition by focusing on the frequency-domain feature of the target ship and suppressing environmental noises and interference.
JASA EXPRESS LETTERS
(2021)
Article
Engineering, Multidisciplinary
Jaskaran Singh et al.
MEASUREMENT SCIENCE AND TECHNOLOGY
(2020)
Article
Engineering, Civil
Yi Zhang et al.
IEEE JOURNAL OF OCEANIC ENGINEERING
(2020)
Article
Computer Science, Artificial Intelligence
Jinyang Jiao et al.
KNOWLEDGE-BASED SYSTEMS
(2020)
Article
Engineering, Electrical & Electronic
Xu Cao et al.
IEEE SENSORS JOURNAL
(2019)
Article
Acoustics
David Santos-Dominguez et al.
Article
Acoustics
Shuguang Wang et al.
Article
Computer Science, Artificial Intelligence
Shai Ben-David et al.