Research on Signal Extraction and Classification for Ship Sound Signal Recognition

Abstract

The movements and intentions of other ships can be determined by gathering and examining ship sound signals. The extraction and analysis of ship sound signals fundamentally support the autonomous navigation of intelligent ships. Mel scale frequency cepstral coefficient (MFCC) feature parameters are improved and optimized to form NewMFCC by introducing second-order difference and wavelet packet decomposition transformation methods in this paper. Transforming sound signals into a feature vector that fully describes the dynamic characteristics of ship sound signals and the high- and low-frequency information solves the problem of the inability to transport ordinary sound signals directly as signals for training in machine learning models. Radial basis function kernels are used to conduct support vector machine classifier simulation experiments. Five types of sound signals, namely, one type of ship sound signals and four types of interference sound signals, are categorized and identified as classification targets to verify the feasibility of the classification of ship sound signals and interference signals. The proposed method improves classification accuracy by approximately 15%.

Keywords

Ship signal identification / Signal extraction / Automatic classification / Intelligent ships / Support vector machine

References

[1] Amelia F, Gunawan D (2019) DWT-MFCC method for speaker recognition system with noise. 7th International Conference on Smart Computing & Communications (ICSCC), Miri, Malaysia, 310–314. DOI: https://doi.org/10.1109/ICSCC.2019.8843660

[2] Bhattacharjee U, Kshirod S (2013) Language identification system using MFCC and prosodic features. International Conference on Intelligent Systems and Signal Processing (ISSP), 194–197. DOI: https://doi.org/10.1109/Issp.2013.6526901

[3] Bin Hamzah HI, Bin Abdullah A, Candrawati R (2009) Biologically-inspired abstraction model to analyze sound signal. IEEE Student Conference on Research and Development, Serdang, Malaysia, 180–183. DOI: https://doi.org/10.1109/Scored.2009.5443168

[4] Bryant ML, Garber FD (1999) SVM classifier applied to the mstar public data set. Proc. SPIE 3721, Algorithms for Synthetic Aperture Radar Imagery VI. https://doi.org/10.1117/12.357652

[5] Chomorlig, Zhang Z, Xiang YL (2012) Research of feature extraction in mongolian speech based on an improved algorithm of MFCC parameter. 2nd International Conference on Advanced Engineering Materials and Technology (AEMT), Zhuhai, 833–837. DOI: https://doi.org/10.4028/www.scientific.net/AMR.542-543.833

[6] Chu XT, Wang HP, Yang HT, Lin NH (2022) Speaker recognition based on convolutional neural network. Police Technology (1): 46–50

[7] Cortes C, Vapnik V. Support-vector networks. Machine Learning, 1995, 20(3): 273-297

[8] Deng MQ, Meng TT, Cao JW, Wang SM, Zhang J, Fan HJ. Heart sound classification based on improved MFCC features and convolutional recurrent neural networks. Neural Networks, 2020, 130: 22-32

[9] Du C, Shao JH, Yang W, Et A. Indoor visible light positioning based on support vector machine optimized by grid search method. Laser Journal, 2021, 42(3): 104-109

[10] Fan P (2017) A complex sound recognition method in noisy environments. Hefei University of Technology (10): 18–21. DOI: https://doi.org/10.7666/D.Y3235272

[11] Geoffrey L, Collie A. Comparison of novices and experts in the identification of sonar signals. Speech Communication, 2004, 43(4): 297-310

[12] Hu L, Hu XJ, Huang ZH, Xu L, Hu K, Zhang JM. Mura defect detection based on effective background reconstruction and contrast enhancement. Journal of Liquid Crystals and Displays, 2021, 36(10): 1395-1402

[13] Lin W, Yang L, Xu B. Speaker recognition based on modified MFCC parameters of chinese mandarin whispered speech. Journal of Nanjing University (Natural Sciences), 2006, 42(1): 54-62

[14] Lu MJ, Dong SJ, Tang M, Wang CR, Cao L, Yan XP (2023) Research on the development of china’s marine transportation equipment industry. Chinese Engineering Science (3): 53–61. DOI: https://doi.org/10.15302/J-Sscae-2023.03.006

[15] Lv DJ, Zhang Y, Fu QJ, Xu HF, Liu J, Zi JL, Huang X (2020) Birdsong recognition based on MFCC combined with vocal tract properties. 5th International Conference on Mechanical, Control and Computer Engineering (ICMCCE), Harbin, 1523–1526. DOI: https://doi.org/10.1109/Icmcce51767.2020.00334

[16] Ma YJ, Zheng XY. Research and application based on improved support vector machine. Industrial Control Computer, 2015, 28(12): 25-26

[17] Mahesha P, Vinod DS (2017) LP-Hillbert transform based MFCC for effective discrimination of stuttering dysfluencies. 2nd IEEE International Conference on Wireless Communications, Signal Processing and Networking (Wispnet), Chennai, 2561–2565. DOI: https://doi.org/10.1109/Wispnet.2017.8300225

[18] Maurya A, Kumar D, Agarwal RK (2017) Speaker recognition for hindi speech signal using MFCC-GMM approach. 6th International Conference on Smart Computing and Communications (ICSCC), Kurukshetra, 880–887. DOI: https://doi.org/10.1016/J.Procs.2017.12.112

[19] Nassim A, Abderrahmane A. Boosting scores fusion approach using front-end diversity and adaboost algorithm, for speaker verification. Computers and Electrical Engineering, 2017, 62: 648-662

[20] Jiang N, Qiu M, Dai W. SROC: A speaker recognition with data decision level fusion method in cloud environment. J Sign Process Syst, 2017, 86: 123-133

[21] Shi YB, Wang L (2011) Improved MFCC algorithm in speaker recognition system. International Conference on Graphic and Image Processing (ICGIP 2011), Cairo, Egypt, 828567. https://doi.org/10.1117/12.913462

[22] Srivastava S, Bhardwaj S, Bhandari A, Gupta K, Bahl H, Gupta JRP (2012) Wavelet packet based mel frequency cepstral features for text independent speaker identification. 1st International Symposium on Intelligent Informatics (ISI12), Chennai, India, 237–247. https://doi.org/10.1007/978-3-642-32063-7_26

[23] Tang L, Tian YJ, Pardalos Panos M. A novel perspective on multiclass classification: Regular simplex support vector machine. Information Sciences, 2018, 480: 324-338

[24] Tian J, Xue SH, Huang HN, Zhang CH. Classification of underwater still objects based on multi-field features and SVM. J Mar. Sc. Appl., 2007, 6: 36-40

[25] Tomchuk KK (2018) Spectral masking in MFCC calculation for noisy speech. Wave Electronics and Its Application In Information and Telecommunication Systems (WECONF), St Petersburg, Russia, 1–4. DOI: https://doi.org/10.1109/WECONF.2018.8604460

[26] Tuncer T, Aydemir E. An automated local binary pattern ship identification method by using sound. Acta Infologica, 2020, 4(1): 57-63

[27] Wang LH, Zhao ZH. Marine mechanical noise monitoring system based on MFCC-SVM. Automation and Instrumentation, 2020, 35(12): 54-58

[28] Wang S, Ding N, Li N, Zhang JJ, Zong CQ. Language cognition and language computation: language understanding by human and machine. Chinese Science: Information Science, 2023, 52(10): 1748-1774

[29] Wang Y, Hu WP. Speech emotion recognition based on improved MFCC. Proceedings of the 2nd International Conference on Computer Science and Application Engineering (CSAE’18), New York, 2018, 88: 1-7 Article

[30] Wróbel K, Montewka J, Kujala P. Towards the assessment of potential impact of unmanned vessels on maritime transportation safety. Reliability Engineering & System Safety, 2017, 165: 155-169

[31] Xie GD, Yang JC, Yi Q, Han ZF. Research on battlefield target sound detection technology. Control Engineering, 2010, 17(S1): 41-44

[32] Xie SS, Xu HF, Liu J, Zhang Y, Lv DJ (2021) Research on bird songs recognition based on MFCC-HMM. International Conference on Computer, Control and Robotics (ICCCR), Shanghai, 262–266. DOI: https://doi.org/10.1109/Icccr49711.2021.9349284

[33] Xu H, Liu YL. Unmanned aerial vehicle sound recognition algorithm based on deep learning. Computer Science, 2021, 48(7): 225-232

[34] Yan CF, Chun HZ, Yuan LD, Zhi YS. Coarse frequency offsetestimation of TD-LTE based on spectrum centroid. Advanced Materials Research, 2013, 756–759: 3602-3606

[35] Yan Q, Zhou ZJ, Li S (2011) Chinese accents identification with modified MFCC. International Conference on Instrumentation, Measurement, Circuits and Systems (ICIMCS 2011), Hong Kong, 659–666. https://doi.org/10.1007/978-3-642-27334-6_77

[36] Yang Y, Chen X. Experimental design of speaker recognition based on neural networks. Laboratory Research and Exploration, 2020, 39(9): 38-41

[37] Yu Y, Xu JX, Zhang MC, Shi H. Comparative analysis of wavelet analysis and wavelet packet analysis in bearing fault diagnosis. Coal Mine Machinery, 2019, 40(12): 170-173

[38] Zhang C, Ma Y (2012) Ensemble machine learning: methods and applications. Springer Science & Business Media, 10–329. DOI: https://doi.org/10.1007/9781441993267

[39] Zhang XY, Bai J, Liang WZ (2006) The speech recognition system based on bark wavelet MFCC. 8th International Conference on Signal Processing, Guilin, 833–835. DOI: https://doi.org/10.1109/ICOSP.2006.345539

[40] Zheng F, Zhang GL, Song ZJ. Comparison of different implementations of MFCC. Journal of Computer Science and Technology, 2001, 16(6): 582-589

[41] Zhong M, Cai W. Feature fusion-based recognition of marine mammal sounds. Electronic Science and Technology, 2019, 32(5): 32-37