Cobalt crust recognition based on kernel Fisher discriminant analysis and genetic algorithm in reverberation environment

Abstract

Recognition of substrates in cobalt crust mining areas can improve mining efficiency. Aiming at the problem of unsatisfactory performance of using single feature to recognize the seabed material of the cobalt crust mining area, a method based on multiple-feature sets is proposed. Features of the target echoes are extracted by linear prediction method and wavelet analysis methods, and the linear prediction coefficient and linear prediction cepstrum coefficient are also extracted. Meanwhile, the characteristic matrices of modulus maxima, sub-band energy and multi-resolution singular spectrum entropy are obtained, respectively. The resulting features are subsequently compressed by kernel Fisher discriminant analysis (KFDA), the output features are selected using genetic algorithm (GA) to obtain optimal feature subsets, and recognition results of classifier are chosen as genetic fitness function. The advantages of this method are that it can describe the signal features more comprehensively and select the favorable features and remove the redundant features to the greatest extent. The experimental results show the better performance of the proposed method in comparison with only using KFDA or GA.

Keywords

feature extraction / kernel Fisher discriminant analysis (KFDA) / genetic algorithm / multiple feature sets / cobalt crust recognition

References

[1] JavierR J, kimY. Application of linear predictive coding for human activity classification based on micro-doppler signatures [J]. IEEE Geoscience & Remote Sensing Letters, 2014, 11(10): 1831-1834

[2] YANG San-wei, CAO Jiu-wen, WANG Jian-zhong. Acoustics recognition of construction equipments based on LPCC features and SVM [C]//Proceedings of the Control Conference. IEEE, 2015: 3987–3991.

[3] MaY, LiZ-shun. On improving feature extraction and identification of underwater wideband echo with orthogonal wavelet packet [J]. Journal of Northwestern Polytechnical University, 2003, 21(1): 54-57(in Chinese)

[4] ZhaoJ-p, HuangJ-g, XieY-q, ZhangH-feng. Transient feature extraction and discrimination of wideband echos based on wavlet transform [J]. Acta Acustica, 1998, 23(1): 31-38(in Chinese)

[5] LiuJ-g, LiZ-s, LiuD. Features of underwater echo extraction based on SWT and SVD [J]. ACTA ACUSTICA-PEKING-, 2006, 31(2): 167-172(in Chinese)

[6] BuY-y, ZhangC, NieS-shuang. Tail-wave envelop feature extraction of an underwater echo based on discrete wavelet transform [J]. Journal of Zhengzhou University (Engineering Science), 2007, 28(4): 80-83(in Chinese)

[7] LiuH, HiroshiM. Feature transformation and subset selection [J]. IEEE Intelligent Systems, 1998, 13(2): 26-28

[8] ZHOU Mu-rong. Research on recognition of deep-sea cobalt crust based on acoustic fractal features [D]. Changsha: College of Mechanical and Electrical Engineering, Central South University, 2009: 41–54. (in Chinese)

[9] XuX-n, MuZ-c, LiY. Feature-level fusion method based on KFDA for multimodal recognition fusing ear and profile face [C]//Proceedings of the International Conference on Wavelet Analysis and Pattern Recognition. Piscataway, USA: IEEE, 2007, 3: 1306-1310

[10] ZhaoH-m, JiY-q, HongY-j, QiH, MaL-Y. A Volterra series-based method for extracting target echoes in the seafloor mining environment [J]. Ultrasonics, 2016, 71: 29-39

[11] QiH, ZhaoH-m, JiY-q, WangY-li. Experimental device for simulating reverberation environment in deep sea mining [J]. Chinese Journal of Engineering, 2017, 39(5): 655-662

[12] ZhaoH-m, WangY-l, HanF-l, JiY-q, LuoB-W. Acoustic pressure simulation and experiment design in seafloor mining environment [J]. Journal of Central South University, 2018, 25(6): 1409-1417

[13] ZhuW, JinP, TaoL-x, TangY-kai. The seismic signal based target recognition of moving vehicles [C]. National Safety Geophysics Series (VII)-Geophysical and Nuclear Exploration, 2011, China, Xi’an: Xi’an Map Publishing House, 394400(in Chinese)

[14] CaoDResearch on human pulse signals recognition algorithm based on cepstrum features [D], 2009, Chongqing, Chongqing University. College of Communication Engineering, 1935(in Chinese)

[15] GeX-m, FanZ-y, FanY-r, LiR-z, XuY-jun. Oil/water contact prediction of complex reservoir using kernel principal component analysis and wavelet power spectrum analysis [J]. Journal of Central South University, 2015, 46(5): 1747-1753(in Chinese)

[16] WangZ-l, ChangJ, ZhangS-s, LuoS, JiaC-w, SunB-n, JiangS, LiuY-n, LiuX-h, LvG-p, LiuZ. Application of wavelet transform modulus maxima in raman distributed temperature sensors [J]. Photonic Sensors, 2014, 4(2): 142-146

[17] EkiciS, YildirimS, PoyrazM. Energy and entropy-based feature extraction for locating fault on transmission lines by using neural network and wavelet packet decomposition [J]. Expert Systems with Applications, 2008, 34(4): 2937-2944

[18] NasonG P, silvermanB W. The stationary wavelet transform and some statistical applications [J]. Science, 1995, 346(6212): 918-919

[19] Galiana-merinoJ J, rosa-herranzJ L, parolaiS. Seismic phase picking using a kurtosis-based criterion in the stationary wavelet domain [J]. IEEE Transactions on Geoscience & Remote Sensing, 2008, 46(11): 3815-3826

[20] ZhuY-w, ShiX-c, LiP. Classification of islanding and grid disturbance based on multi-resolution singular spectrum entropy and SVM [J]. Proceedings of the Csee, 2011, 31(7): 64-70

[21] NasonG P, silvermanB WThe stationary wavelet transform and some statistical applications [M], 1995, New York, Springer, 281299

[22] KodipakaS, VemuriB C, rangarajanA, LeonardC M, schmallfussI, EisenschenkS. Kernel fisher discriminant for shape-based classification in epilepsy [J]. Medical Image Analysis, 2007, 11(1): 79-90

[23] MikaS, RatschG, WestonJ, ScholkopfB, MullerK R. Fisher discriminant analysis with kernels [C]. Proceedings of the Neural networks for signal processing IX: Proceedings of the 1999 IEEE Signal Processing Society Workshop (cat. no. 98th8468), 1999, Piscataway, USA, IEEE, 4148

[24] TanF, FuX-z, ZhangY-q, AnuG B. A genetic algorithm-based method for feature subset selection [J]. Soft Computing, 2008, 12(2): 111-120

[25] ZhouY, LiY-w, XiaS-xiong. An improved knn text classification algorithm based on clustering [J]. Journal of Computers, 2009, 4(3): 230-237

[26] WangS-l, LiX-l, ZhangS-w, GuiJ. Tumor classification by combining PNN classifier ensemble with neighborhood rough set based gene reduction [J]. Computers in Biology and Medicine, 2010, 40(2): 179-189

[27] FengG-y, XiaoH-t, FuQ, HuangM-jun. Method of radar target classification based on adaptive SVDD [J]. Systems Engineering and Electronics, 2011, 33(2): 253-258(in Chinese)

[28] Azimi-sadjadiM R, yaoD, HuangQ, DobeckG J. Underwater target classification using wavelet packets and neural networks [J]. IEEE Transactions on Neural Networks, 2000, 11(3): 177-182