基于声信号多重分形和支持向量机的目标识别研究
|
摘要
为提高智能地雷对地面装甲目标的识别率,针对地面装甲目标辐射的噪声信号具有非线性的特性,建立了一种基于多重分形和支持向量机(SVM)相结合的分类识别模型。通过野外场地实验,采集到两种装甲目标在不同工况(运行速度)下的各40组样本信号;利用多重分形分析计算得到两种目标信号的广义分形维数谱(GFDS),分析了两种目标信号在不同工况下多重分形谱的特征;将GFDS值作为目标特征向量输入SVM分类模型,经训练得到最优分类结果,并与小波包能量(WPE)法提取样本特征后输入SVM的识别效果进行了对比,结果表明前者的识别率达到92.5%,高于后者的85%的识别率。
In order to improve the recognition rate of smart landmines for armored target,as the acoustic signals radiated from armored vehicles have been proved to be nonlinear,an identification model based on multifractal analysis and support vector machine(SVM) was established.40 sample signals for each armored target(a certain type of wheeled armored vehicle and a tank) running in different speeds(2 working conditions) were collected by outdoor experiment.The generalized fractal dimension spectrums(GFDS) for both target signals were calculated based on multifractal analysis,and the characters of GFDS under 2 working conditions were analyzed.The GFDS values were input into SVM classification model,and the optimal identification results were obtained by training the model.After an identification effect comparison between GFDS and wavelet packet energy(WPE) method,the results show that the model based on GFDS and SVM has a recognition rate of 92.5%,which is higher than the 85% by WPE method.
In order to improve the recognition rate of smart landmines for armored target,as the acoustic signals radiated from armored vehicles have been proved to be nonlinear,an identification model based on multifractal analysis and support vector machine(SVM) was established.40 sample signals for each armored target(a certain type of wheeled armored vehicle and a tank) running in different speeds(2 working conditions) were collected by outdoor experiment.The generalized fractal dimension spectrums(GFDS) for both target signals were calculated based on multifractal analysis,and the characters of GFDS under 2 working conditions were analyzed.The GFDS values were input into SVM classification model,and the optimal identification results were obtained by training the model.After an identification effect comparison between GFDS and wavelet packet energy(WPE) method,the results show that the model based on GFDS and SVM has a recognition rate of 92.5%,which is higher than the 85% by WPE method.
引文
[1]栗苹,崔占忠,施聚生.声目标识别技术及其在坦克目标识别中的应用[J].兵工学报,1996,17(1):51-54.LI Ping,CUI Zhan-zhong,SHI Ju-sheng.Technique of acoustic tar-get identification and its application in tank recognition[J].ActaArmamentarii,1996,17(1):51-54.(in Chinese)
[2]丁庆海,庄志洪,祝龙石,等.混沌、分形和小波理论在被动声信号特征提取中的应用[J].声学学报,1999,24(2):197-203.(in Chinese)DING Qing-hai,ZHUANG Zhi-hong,ZHU Long-shi,et al.Applica-tion of the chaos,fractal and wavelet theories to the feature extrac-tion of passive acoustic signal[J].Acta Acoustica,1999,24(2):197-203.(in Chinese)
[3]张胜,方向,张卫平.基于1阶Minkovski维数的多重分形算法[J].兵工学报,2012,33(7):822-825.ZHANG Sheng,FANG Xiang,ZHANG Wei-ping.Multi-fractal al-gorithms based on first order Minkovski dimension[J].Acta Arma-mentarii,2012,33(7):822-825.(in Chinese)
[4]Kiyashchenko D,Smirnova N,Troyan V,et al.Seismic hazardprecursory evolution:fractal and multifractal aspects[J].Physicsand Chemistry of the Earth,2004,29:367-378.
[5]杜恩祥,李科杰.基于多重分形和小波变换的声目标信号特征提取[J].自动化学报,2004,30(5):742-746.DU En-xiang,LI Ke-jie.Feature extraction of acoustic signal of tar-get based on multifractal and wavelet theories[J].Acta AutomaticaSinica,2004,30(5):742-746.(in Chinese)
[6]GUAN Jian,LIU Ning-bo,ZHANG Jian,et al.Multifractal correla-tion characteristic for radar detecting low-observable target in seaclutter[J].Signal Processing,2010,90:523-535.
[7]唐克,魏琪,杜涛.基于支持向量机的高空无人机侦察目标识别[J].火力指挥与控制,2010,35(3):82-85,99.TANG Ke,WEI Qi,DU Tao.The reconnaissance target recognitionof the high altitude[J].Fire Control&Command Control,2010,35(3):82-85,99.(in Chinese)
[8]陈时军,David H,王丽凤,等.广义地震应变能释放的多重分形特征[J].地震学报,2003,25(2):182-190.CHEN Shi-jun,David H,WANG Li-feng,et al.Multifractal charac-teristics of general stress release of earthquakes[J].Acta Seis-mologica Sinica,2003,25(2):182-190.(in Chinese)
[9]余英林,谢胜利,蔡汉添,等.信号处理新方法导论[M].北京:清华大学出版社,2004.YU Ying-lin,XIE Sheng-li,CAI Han-tian,et al.Introduction tonew methods of signal processing[M].Beijing:Tsinghua UniversityPress,2004.(in Chinese)
[10]方向,陆凡东,丁凯,等.基于动态粗糙集和支持向量机的振动参量预报[J].解放军理工大学学报:自然科学版,2012,13(1):97-100.FANG Xiang,LU Fan-dong,DING Kai,et al.Prediction of blas-ting vibration parameter based on rough set with dynamic discreti-zation and support vector machine[J].Journal of PLA Universityof Science and Technology:Natural Science Edition,2012,13(1):97-100.(in Chinese)
[11]俞玉莲,郭世杰.基于支持向量机的多类音频分类[J].计算机应用与软件,2010,27(4):98-101.YU Yu-lian,GUO Shi-jie.Multi-class audio classification basedon support vector machine[J].Computer Applications and Soft-ware,2010,27(4):98-101.(in Chinese)
[12]Alan V O,Alan S W.Signals&systems[M].2nd ed.NewJersey:Prentice-Hall Inc,1997.
[13]周德廉,邵国友.现代测试技术与信号分析[M].徐州:中国矿业大学出版社,2005.ZHOU De-lian,SHAO Guo-you.Modern measurement technologyand signal analysis[M].Xuzhou:China University of Mining&Technology Press,2005.(in Chinese)
[2]丁庆海,庄志洪,祝龙石,等.混沌、分形和小波理论在被动声信号特征提取中的应用[J].声学学报,1999,24(2):197-203.(in Chinese)DING Qing-hai,ZHUANG Zhi-hong,ZHU Long-shi,et al.Applica-tion of the chaos,fractal and wavelet theories to the feature extrac-tion of passive acoustic signal[J].Acta Acoustica,1999,24(2):197-203.(in Chinese)
[3]张胜,方向,张卫平.基于1阶Minkovski维数的多重分形算法[J].兵工学报,2012,33(7):822-825.ZHANG Sheng,FANG Xiang,ZHANG Wei-ping.Multi-fractal al-gorithms based on first order Minkovski dimension[J].Acta Arma-mentarii,2012,33(7):822-825.(in Chinese)
[4]Kiyashchenko D,Smirnova N,Troyan V,et al.Seismic hazardprecursory evolution:fractal and multifractal aspects[J].Physicsand Chemistry of the Earth,2004,29:367-378.
[5]杜恩祥,李科杰.基于多重分形和小波变换的声目标信号特征提取[J].自动化学报,2004,30(5):742-746.DU En-xiang,LI Ke-jie.Feature extraction of acoustic signal of tar-get based on multifractal and wavelet theories[J].Acta AutomaticaSinica,2004,30(5):742-746.(in Chinese)
[6]GUAN Jian,LIU Ning-bo,ZHANG Jian,et al.Multifractal correla-tion characteristic for radar detecting low-observable target in seaclutter[J].Signal Processing,2010,90:523-535.
[7]唐克,魏琪,杜涛.基于支持向量机的高空无人机侦察目标识别[J].火力指挥与控制,2010,35(3):82-85,99.TANG Ke,WEI Qi,DU Tao.The reconnaissance target recognitionof the high altitude[J].Fire Control&Command Control,2010,35(3):82-85,99.(in Chinese)
[8]陈时军,David H,王丽凤,等.广义地震应变能释放的多重分形特征[J].地震学报,2003,25(2):182-190.CHEN Shi-jun,David H,WANG Li-feng,et al.Multifractal charac-teristics of general stress release of earthquakes[J].Acta Seis-mologica Sinica,2003,25(2):182-190.(in Chinese)
[9]余英林,谢胜利,蔡汉添,等.信号处理新方法导论[M].北京:清华大学出版社,2004.YU Ying-lin,XIE Sheng-li,CAI Han-tian,et al.Introduction tonew methods of signal processing[M].Beijing:Tsinghua UniversityPress,2004.(in Chinese)
[10]方向,陆凡东,丁凯,等.基于动态粗糙集和支持向量机的振动参量预报[J].解放军理工大学学报:自然科学版,2012,13(1):97-100.FANG Xiang,LU Fan-dong,DING Kai,et al.Prediction of blas-ting vibration parameter based on rough set with dynamic discreti-zation and support vector machine[J].Journal of PLA Universityof Science and Technology:Natural Science Edition,2012,13(1):97-100.(in Chinese)
[11]俞玉莲,郭世杰.基于支持向量机的多类音频分类[J].计算机应用与软件,2010,27(4):98-101.YU Yu-lian,GUO Shi-jie.Multi-class audio classification basedon support vector machine[J].Computer Applications and Soft-ware,2010,27(4):98-101.(in Chinese)
[12]Alan V O,Alan S W.Signals&systems[M].2nd ed.NewJersey:Prentice-Hall Inc,1997.
[13]周德廉,邵国友.现代测试技术与信号分析[M].徐州:中国矿业大学出版社,2005.ZHOU De-lian,SHAO Guo-you.Modern measurement technologyand signal analysis[M].Xuzhou:China University of Mining&Technology Press,2005.(in Chinese)
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心