臺灣博碩士論文加值系統

透過水面或水下載具裝載水下聲紋辨識系統，進行水下特徵辨識的技術已被應用於反潛作戰和港灣防禦中。然而，水聲通道會受到海底地形、海況條件的時空變化、水文環境的時變性干擾，造成多路徑效應及噪音遮蔽等影響。因此，如何管控所收錄之聲學資料品質具有挑戰性。除此之外，水下聲紋辨識系統亦須解決的挑戰為如何從不同時空環境所錄製的聲學檔案中識別出相同的目標特徵-海洋研究船三號(Ocean Researcher III, OR3)。
本研究選定以海洋研究船三號之船舶聲紋為識別目標，參考挪威船級協會
DNV GL-Silent之量測規範，使用水下聲音錄製系統量測其水下輻射噪音並於2016年和2017年間共量測四次。水下聲紋辨識系統乃透過特徵萃取、特徵篩選與資料探勘等步驟，並用統計分法與機器學習兩種方法進行聲紋辨識並比較其優劣。由於基頻及其功率譜密度為船舶頻譜上之重要分類特徵。在特徵萃取過程中，先萃取出此特徵後，再透過特徵篩選及資料探勘等程序，分析出最重要之特徵。
第一種方法，利用多項式回歸(Polynomial Regression, PR)和田口方法(Taguchi method)及方差分析(analysis of variance,ANOVA)，求得不同因子在不同水準組合下之渴望函數(Desirability function)，篩選出最大渴望函數值為最佳組合。
第二種方法，採用自適應性神經模糊推論系統(Adaptive Network-Based FuzzyInference System, ANFIS)，同樣透過計算渴望函數值，利用遺傳演算法(geneticalgorithm, GA)找出最佳的因子組合。
由結果可知，特徵萃取步驟，所萃取之特徵可成功辨識不同時空下之同一目標船舶之水下輻射噪音，以倍頻作為辨識特徵，可解決噪音遮蔽的問題。利用統計分析及機器學習所建立之演算法皆可達到90%以上之準確度。由本研究所開發之「海用載具聲紋智能辨識系統(Artificial Intelligence Application in a Marine Vehicle Identification System Using Acoustic Signatures ) 」，可應於於港灣防禦作為船船監測之用， 若應用於海上即時監測，建議仍以智能學習演算法所建立之辨識器為主。

Underwater acoustic signature identification has been employed as a technique for detecting underwater vehicles, such as in anti-submarine warfare or harbor security systems. The underwater sound channel, however, has interference due to spatial variations in topography or sea state conditions and temporal variations in water column properties, which causes multipath and scattering in acoustic propagation. Thus, acoustic data quality control can be very challenging. One of challenges for an identification system is how to recognize the same target signature from measurements following the rule of DNV GL-Silent under different temporal and spatial settings.
This paper deals with the above challenges by establishing an identification system composed of feature extraction, feature selection, and data mining with statistical analysis and machine learning approaches to recognize the target signatures of underwater radiated noise from a research vessel, Ocean Researcher III (OR3), with a bottom mounted hydrophone in four cruises in 2016 and 2017. The fundamental frequency and its power spectral density are significant features for classification. In feature extraction, we extract the features before deciding which of the two aforementioned features is more significant. The first approach utilizes Polynomial Regression (PR) classifiers and feature selection by Taguchi method and analysis of variance (ANOVA) for finding the maximum desirability function under a different combination of factors and levels. The second approach utilizes Adaptive Network-Based Fuzzy Inference System (ANFIS) and selecting the optimized desirability value via genetic algorithm (GA).
The result proves that the feature extraction process successfully detects the OR3 targets under different topography and temporal variations and solves the noise masking problem by utilizing harmonic frequency features extracted from unmasking the frequency bandwidth for ship noises. Nevertheless, the accuracy of PR and ANFIS classifiers all approach 90%. The “Artificial Intelligence Application in a Marine Vehicle Identification System Using Acoustic Signatures” developed here can be carried out in harbor security for monitoring acoustic signatures of a specific ship. In addition, it utilizes the artificial intelligence classifier, which has faster detecting performance than the regression classifier on real-time monitoring.

博士學位論文口試委員會審定書 i
謝誌 ii
中文摘要 iv
ABSTRACT v
符號說明 vii
縮寫說明 xi
目錄 xiv
圖目錄 xvi
表目錄 xx
第1章 緒論 1
1.1 研究動機 1
1.2 船舶水下輻射噪音介紹 2
1.3 量測船舶水下輻射噪音之文獻回顧 5
1.4 水下聲紋資料分析之文獻回顧 9
1.5 研究目的 12
1.6 論文架構 15
第2章 船舶輻射噪音量測實驗 17
2.1 船舶水下噪音量測程序介紹 17
2.2 實驗規劃 27
2.3 實驗過程 34
2.4 實驗討論 36
2.5 實驗小結 39
2.6 聲紋樣本分類 45
第3章 特徵萃取 49
3.1 萃取基頻(FF) 50
3.2 計算頻率出現次數(OT)與倍頻出現次數(HT) 53
3.3 萃取主頻(MF) 56
3.4 特徵萃取之結果 57
第4章 特徵建模 59
4.1 以多項式回歸(PR)進行 63
4.2 以自適應性神經模糊推論系統(ANFIS)進行 65
第5章 特徵篩選之結果 70
5.1 以多項式回歸(PR)進行 72
5.2 以自適應性神經模糊推論系統(ANFIS)進行 82
第6章 討論與結果 85
6.1 比較不同特徵建模方法之差異 85
6.2 比較不同頻段下之偵測模型差異 92
6.3 最佳化偵測模組應用於OR3實測資料(2017年9月航次) 93
6.4 結果 102
第7章 結論 104
附錄 A 主成分分析法(PCA)之可行性分析 106
A.1 主成分分析法介紹 106
A.2 分類器演算法：支持向量機（SVM） 110
A.3 PCA可行性分析之結果與討論 111
參考文獻 114

1.Jae-Byung, J., et al. Broadband Active Sonar Swimmer Detection and Identification. in Neural Networks, 2006. IJCNN ''06. International Joint Conference on. 2006.
2.Merchant, N.D., et al., Measuring acoustic habitats. Methods in Ecology and Evolution, 2015. 6(3): p. 257-265.
3.Merchant, N.D., et al., Averaging underwater noise levels for environmental assessment of shipping. The Journal of the Acoustical Society of America, 2012. 132(4): p. EL343-EL349.
4.Hagerty, M., et al., Analysis of seismic and acoustic observations at Arenal Volcano, Costa Rica, 1995–1997. Journal of Volcanology and Geothermal Research, 2000. 101(1): p. 27-65.
5.Iyama, J. and H. Kuwamura, Application of wavelets to analysis and simulation of earthquake motions. Earthquake engineering & structural dynamics, 1999. 28(3): p. 255-272.
6.Ibs-von Seht, M., Detection and identification of seismic signals recorded at Krakatau volcano (Indonesia) using artificial neural networks. Journal of Volcanology and Geothermal Research, 2008. 176(4): p. 448-456.
7.Hsu, S.K., et al. Marine cable hosted observatory (MACHO) project in Taiwan. in Underwater Technology and Workshop on Scientific Use of Submarine Cables and Related Technologies, 2007. Symposium on. 2007. IEEE.
8.許樹坤等人., 台灣東部海域海底電纜觀測系統建置規劃. 中央氣象局2007年地震技術報告, 2007: p. p361-382.
9.許樹坤等人., 台灣東部海域電纜觀測系統建置諮詢研究. 中央氣象局2008年地震技術報告, 2008: p. pp.1-10.
10.Chen, C.F., et al. Data demonstrations on physical oceanography and underwater acoustics from the MArine Cable Hosted Observatory (MACHO). in OCEANS, 2012-Yeosu. 2012. IEEE.
11.Fang, Y.Y., K.C. Wu, and C. Chen, Underwater acoustic detection and classification for cetaceans'' vocalizations of Marine Observatory in the Northeastern Taiwan (MONET). The Journal of the Acoustical Society of America, 2012. 131(4): p. 3494-3494.
12.Yashaswini, S.S.P., et al., A literature survey on ambient noise analysis for underwater acoustic signals. Internation Journal of Computer Engineering and Science, 2015. 1(7): p. 1-9.
13.Chiu, L.Y. and H.-R. Chen. Estimation and reduction of effects of sea surface reflection on underwater vertical channel. in Underwater Technology Symposium (UT), 2013 IEEE International. 2013. IEEE.
14.Chang, A., et al. Modeling of shallow water ambient noise based on adiabatic mode theory. in Underwater Technology Symposium (UT), 2013 IEEE International. 2013. IEEE.
15.Wenz, G.M., Acoustic ambient noise in the ocean: spectra and sources, in The Journal of the Acoustical Society of America. 1962. p. 1936-1956.
16.Ross, D., Mechanics of Underwater Noise Pergamon Press. New York, USA, 1976.
17.Ross, D. and W. Kuperman, Mechanics of underwater noise. 1989, The Journal of the Acoustical Society of America.
18.Ross, D., Mechanics of underwater noise. 2013: Elsevier.
19.魏宏源, 船體振動與其水下輻射噪音 Ship Hull Vibration And Its Underwater Radiated Noise, in 造船工程學系. 1993, 國立臺灣大學: 國立臺灣大學.
20.Urick, R.J., Principles of underwater sound. Vol. 3. 1983: McGraw-Hill New York.
21.劉倬騰, 陳琪芳, 船舶振動雨水中訊號之現場實驗. 1995, 國科會國防科技學術合作協調小組委託計畫成果報告: 臺灣大學應用力學研究所.
22.Wright, A.J., Internatioanl Workshop on Shipping Noise and Marine Mammals. 2008. Hamburg, Germany.
23.Au, W.W., Characteristics of dolphin sonar signals, in The Sonar of Dolphins. 1993, Springer. p. 115-139.
24.Clark, C.W., Acoustic behavior of mysticete whales, in Sensory abilities of cetaceans. 1990, Springer. p. 571-583.
25.Edds-Walton, P.L., Acoustic communication signals of mysticete whales. Bioacoustics, 1997. 8(1-2): p. 47-60.
26.Evans, G.W. and K. English, The environment of poverty: Multiple stressor exposure, psychophysiological stress, and socioemotional adjustment. Child development, 2002. 73(4): p. 1238-1248.
27.White, P.R. Underwater noise limits and measurement of underwater radiated noise from merchant vessels. in 24th International Congress on Sound and Vibration. 2017. London.
28.IMO MEPC.1/Circ.833: Guidelines for the Reduction of Underwater Noise from Commercial Shipping to Address Adverse Impacts on Marine Life, I.M.O. (IMO), Editor. 2014.
29.Williams, R., et al., Marine mammals and ocean noise: Future directions and information needs with respect to science, policy and law in Canada. Marine pollution bulletin, 2014. 86(1-2): p. 29-38.
30.Tavener, S. and T. Cooper, Automatic Identification System: AIS-A Reception of AIS-B, 2008 Study. 2008.
31.Bittencourt, L., et al., Underwater noise pollution in a coastal tropical environment. Marine pollution bulletin, 2014. 83(1): p. 331-336.
32.Codarin, A. and M. Picciulin, Underwater noise assessment in the Gulf of Trieste (Northern Adriatic Sea, Italy) using an MSFD approach. Marine pollution bulletin, 2015. 101(2): p. 694-700.
33.Merchant, N.D., et al., Assessing sound exposure from shipping in coastal waters using a single hydrophone and Automatic Identification System (AIS) data. Marine pollution bulletin, 2012. 64(7): p. 1320-1329.
34.Kaplan, M.B. and T.A. Mooney, Ambient noise and temporal patterns of boat activity in the US Virgin Islands National Park. Marine pollution bulletin, 2015. 98(1): p. 221-228.
35.陳俊廷, 船舶輻射聲紋量測. 第27 屆海洋工程研討會論文集.國立中興大. 2005: p. p2-4.
36.黃明志, 水下聲響感測器實作應用整合型研究-子計畫四:船舶幅射聲紋量測與分析 (I). 行政院國家科學委員會專題研究計畫期中進度報告. 2004.
37.黃明志, 水下聲響感測器實作應用整合型研究─子計畫四:船舶幅射聲紋量測與分析 (2/2). 行政院國家科學委員會專題研究計畫期中進度報告.2005.
38.陳琪芳, 海洋研究船五號船舶噪音量測研究. 2012, 國立臺灣大學工程科學及海洋工程學系.
39.Hung, J.-w., Several New Approaches for Robust Speech Recognition under Noisy Environment. Ph.D. Dissertation Graduate Institute of Communication Engineering National Taiwan University, 2001.
40.Gonzalez-Sanchez, J., et al., Affect measurement: A roadmap through approaches, technologies, and data analysis, in Emotions and Affect in Human Factors and Human-Computer Interaction. 2017, Elsevier. p. 255-288.
41.Kayaalp, F. and M.S. Başarslan, Open source data mining programs: a case study on R. 2018.
42.Harding, J., M. Shahbaz, and A. Kusiak, Data mining in manufacturing: a review. Journal of Manufacturing Science and Engineering, 2006. 128(4): p. 969-976.
43.Cabric, D., A. Tkachenko, and R.W. Brodersen, Experimental Study of Spectrum Sensing based on Energy Detection and Network Cooperation. The 2nd Annual International Wireless Internet Conference (WICON), 2006(TAPAS Workshop).
44.Kongrattanaprasert, W., et al. Application of neural network analysis to automatic detection of road surface conditions utilizing tire noise from vehicles. in ICCAS-SICE2009. 2009.
45.方銀營等人. 針對MACHO計畫水下麥克風之聲學事件偵測研究, 第15屆水下技術研討會國科會成果發表會. 2013. 國立臺灣海洋大學.
46.陳松琳, 以類神經網路為架構之語音辨識系統The Speech Recognition System using Neural Networks. 國立中山大學電機工程學系碩士論文, 2002.
47.周嘉文, 船艦結構應力與應變狀況診斷用之多層光纖感測器設計研究. 國防大學中正理工學院造船工程研究所碩士論文, 2008.
48.羅華強, 類神經網路－Matlab的應用. 2005, 臺北：高立圖書有限公司.
49.焦李成, 神經網路系統理論. 1991, 臺北：儒林圖書出版有線公司.
50.Fang, Y.-Y., et al., Event Detection of Underwater Acoustic Data from MACHO hydrophone. 2011.
51.Wang, S. and X. Zeng, Robust underwater noise targets classification using auditory inspired time–frequency analysis. Applied acoustics, 2014. 78: p. 68-76.
52.Weiss, L. and T. Dixon, Wavelet-based denoising of underwater acoustic signals. The Journal of the Acoustical Society of America, 1997. 101(1): p. 377-383.
53.Bodisco, T.A., et al., Characterising an ECG signal using statistical modelling: a feasibility study. ANZIAM Journal, 2014. 55: p. 32-46.
54.Ribeiro-Fonseca, J. and L. Correia. Identification of underwater acoustic noise. in OCEANS''94.''Oceans Engineering for Today''s Technology and Tomorrow''s Preservation.''Proceedings. 1994. IEEE.
55.Li, C.-T., S.-J. Wu, and W.-L. Yu, Parameter design on the multi-objectives of PEM fuel cell stack using an adaptive neuro-fuzzy inference system and genetic algorithms. International Journal of Hydrogen Energy, 2014. 39(9): p. 4502-4515.
56.Jang, J.-S. Input selection for ANFIS learning. in Proceedings of IEEE 5th International Fuzzy Systems. 1996. IEEE.
57.王文俊, 認識 Fuzzy. 2005, 全華科技圖書股份有限公司.
58.Gaffney, J., C. Pearce, and D. Green, Binary versus real coding for genetic algorithms: A false dichotomy? Anziam journal, 2010. 51: p. 347-359.
59.Drummond, C. and R.C. Holte. Exploiting the cost (in) sensitivity of decision tree splitting criteria. in ICML. 2000.
60.Elkan, C. The foundations of cost-sensitive learning. in International joint conference on artificial intelligence. 2001. Lawrence Erlbaum Associates Ltd.
61.Gillard, C., et al., Automatic classification of active sonar echoes for improved target identification. Maritime Operations Division, Defence Science and Technology Organisation. 2011.
62.Montgomery, D.C., Design and analysis of experiments. 2017: John wiley & sons.
63.Akteke Ozturk, B., G. Koksal, and G.W. Weber. Optimization of desirability functions as a DNLP model by GAMS/BARON. in AIP Conference Proceedings. 2010. AIP.
64.Wu, S.-J., S.-W. Shiah, and W.-L. Yu, Parametric analysis of proton exchange membrane fuel cell performance by using the Taguchi method and a neural network. Renewable Energy, 2009. 34(1): p. 135-144.
65.Taguchi, G., Taguchi on robust technology development methods, 1993. ASME Press, New York). Google Scholar.
66.Karna, S.K. and R. Sahai, An overview on Taguchi method. International Journal of Engineering and Mathematical Sciences, 2012. 1(1): p. 1-7.
67.Taguchi, G. Off-line and on-line quality control systems. in Proceedings of International Conference on Quality Control. 1978.
68.Taguchi, G., Introduction to quality engineering: designing quality into products and processes. Asian Productivity Organization. 1986.
69.Horvath, R., G. Matyasi, and A. Dregelyi-Kiss, Optimization of machining parameters for fine turning operations based on the response surface method. ANZIAM Journal, 2014. 55: p. 250-265.
70.Guisan, A., T.C. Edwards Jr, and T. Hastie, Generalized linear and generalized additive models in studies of species distributions: setting the scene. Ecological modelling, 2002. 157(2-3): p. 89-100.
71.Mitson, R., ICES Cooperative Research Report 209: Five Years On. ICES Fisheries Acoustic Science & Technology Working Group, Haarlem, Netherlands, 2000. 7.
72.USA, ANSI/ASA S12.64-2009 part1: Quantities and Procedures for Description and Measurement of Underwater Sound from Ships –Part 1 General Requirements. 2009.
73.VERITAS, D.N., DNV rules for classification of ships new buildings- special equipment and systems additional class- Part 6 Chapter 24: Silent class notation. 2010.
74.Veritas, B., Underwater Radiated Noise (URN), Rule Note NR 614 DT R00 E, in Bureau Veritas. 2014.
75.China, GJB 4057-2000: Measurement method for noise of ship. 2000.
76.国家海洋局第三海洋研究所, 一種船舶輻射噪音的測量方法. 2017, 国家海洋局第三海洋研究所.
77.ISO 60565, I., Underwater Acoustics—Hydrophones—Calibration in the Frequency Range From 0.01 Hz to 1 MHz. 2006, International Electrotechnical Commission Geneva, Switzerland.
78.ISO 61260, I., Electroacoustics–Octave‐Band and Fractional‐Octave‐Band Filters. International Electrotechnical Commission,(January 1996), 1995.
79.ISO 17208-1:2012(E), I.P., ISO/PAS 17208-1:2012(E) Acoustics — Quantities and procedures for description and measurement of underwater sound from ships. Part 1: General requirements for measurements in deep water. 2012, ISO: ISO.
80.方銀營等人. 海洋研究船之船舶水下輻射噪音實海域量測分析研究, 第二十九屆中國造船暨輪機工程研討會暨科技部成果發表會 2017. 國立成功大學系統及船舶機電工程學系.
81.Fang, Y.-Y., C.-F. Chen, and S.-J. Wu. Feature identification using acoustic signature of Ocean Reasearcher III (ORIII) in The 13th Engineering Mathmetics and Applications Conference 2017. Auckland, New Zealand: University of Auckland.
82.Fang, Y.-Y., C.-F. Chen, and S.-J. Wu. Automatic Identification System for Vehicles using Acoustic Signature. in 2nd Oceanoise Asia. 2018. Hakodate, Hokkaido, Japan.
83.Fang, Y.-Y., C.-F. Chen, and S.-J. Wu. Analysis of vibration and underwater radiated noise of Ocean Researcher III (OR III). in The 30th Taiwan Society of Naval Architects and Marine Engineers Conference. 2018. National Taiwan Ocean University.
84.Fang, Y.-Y., C.-F. Chen, and S.-J. Wu. Optimization of acoustic signature identification system for Ocean Researcher III (OR3) in The 6th Pacific Rim Underwater Acoustic Conference 2018. Taipei/ Hualien, Taiwan
85.Burgstahler, L. and M. Neubauer. New modifications of the exponential moving average algorithm for bandwidth estimation. in Proc. of the 15th ITC Specialist Seminar. 2002.
86.Lamichhane, B.P., Removing a mixture of Gaussian and impulsive noise using the total variation functional and split Bregman iterative method. ANZIAM Journal, 2015. 56: p. 52-67.
87.Feng, C.-X.J., Z.-G. Yu, and A. Kusiak, Selection and validation of predictive regression and neural network models based on designed experiments. IIE Transactions, 2006. 38(1): p. 13-23.
88.Su, C.-T., Quality engineering: off-line methods and applications. 2016.
89.Antony, J. and M. Kaye, Experimental quality: a strategic approach to achieve and improve quality. 2012: Springer Science & Business Media.
90.Fawcett, T., An introduction to ROC analysis. Pattern recognition letters, 2006. 27(8): p. 861-874.
91.Jang, J.-S., ANFIS: adaptive-network-based fuzzy inference system. IEEE transactions on systems, man, and cybernetics, 1993. 23(3): p. 665-685.
92.Takagi, T. and M. Sugeno, Fuzzy identification of systems and its applications to modeling and control, in Readings in Fuzzy Sets for Intelligent Systems. 1993, Elsevier. p. 387-403.
93.Sugeno, M. and G. Kang, Structure identification of fuzzy model. Fuzzy sets and systems, 1988. 28(1): p. 15-33.
94.Cawley, G.C. and N.L. Talbot, Efficient leave-one-out cross-validation of kernel fisher discriminant classifiers. Pattern Recognition, 2003. 36(11): p. 2585-2592.
95.Chang, H.-H., A data mining approach to dynamic multiple responses in Taguchi experimental design. Expert Systems with Applications, 2008. 35(3): p. 1095-1103.
96.Demuth, H. and M. Beale, Neural Network Toolbox For Use with Matlab--User''S Guide Version 4.0. 1993.