用于水下目标识别的无监督谱特征选择算法
详细信息 查看官网全文
- 英文论文题名：Unsupervised spectral feature selection algorithm for underwater acoustic target recognition
- 论文作者：杨宏晖 ; 李江涛 ; 甘安琴 ; 姚晓辉
- 英文论文作者：YANG Hong-hui ; LI Jiang-tao ; GAN An-qin ; YAO Xiao-hui ; Northwestern Polytechnical University
- 年：2016
- 作者机构：西北工业大学航海学院;
- 论文关键词：水声目标识别 ; 谱特征选择 ; 无监督
- 英文论文关键词：underwater acoustic target recognition ; spectral feature selection ; unsupervised
- 会议召开时间：2016-10-25
- 会议录名称：2016年中国造船工程学会水中目标特性学组学术交流会论文集
- 语种：中文
- 分类号：TB56
- 学会代码：SXJS
- 会议名称：2016年中国造船工程学会水中目标特性学组学术交流会
- 会议地点：中国辽宁大连
- 主办单位：中国船舶重工集团公司第七六0研究所水中测控技术重点实验室
- 学会名称：《声学技术》编辑部
- 页数：6
- 文件大小：763k
- 原文格式：D
- 会议级别：全国
摘要
针对水声目标数据的特征冗余问题,提出了一种用于水声目标识别的无监督谱特征选择算法。首先根据给定的数据样本集利用特定的方法计算能够表示其几何结构的关系矩阵;然后选择该矩阵合适的特征向量,通过相应的低维嵌入代表原始数据集;最后通过稀疏化约束求得该转换矩阵。对该转换矩阵求2,1?范数可有效地选择特征。用实测水声数据集和sonar数据集进行特征选择和分类实验,该算法能以较少的特征得到比原来还高的分类识别正确率;实验结果证明该算法有能有效去除冗余特征和不相关特征。
The problem of feature redundancy in underwater target recognition has been studying by plenty of researchers. In this paper, we propose a new unsupervised spectral feature selection algorithm. We primarily utilize a low dimension graph embedding represents the original dataset, and then obtain the transformation matrix by sparse constraint. We select useful features by calculating the ?_(2,1)-norm of the transformation matrix. Results of classification experiments with actual measured underwater acoustic target dataset and sonar dataset after feature selection show that the proposed algorithm can obtain higher classification accuracy by less features. Finally, the unsupervised spectral feature selection algorithm can effectively remove redundant and irrelevant features.
The problem of feature redundancy in underwater target recognition has been studying by plenty of researchers. In this paper, we propose a new unsupervised spectral feature selection algorithm. We primarily utilize a low dimension graph embedding represents the original dataset, and then obtain the transformation matrix by sparse constraint. We select useful features by calculating the ?_(2,1)-norm of the transformation matrix. Results of classification experiments with actual measured underwater acoustic target dataset and sonar dataset after feature selection show that the proposed algorithm can obtain higher classification accuracy by less features. Finally, the unsupervised spectral feature selection algorithm can effectively remove redundant and irrelevant features.
引文
[1]杨宏晖,王芸,孙进才,等.融合样本选择与特征选择的Ada Boost支持向量机集成算法[J].西安交通大学学报,2014,48(12):63-68.YANG H H,WANG Y,SUN J C,et al.An Ada Boost Support Vector Machine Ensemble Method with Integration of Instance Selection and Feature Selection[J].Journal of Xi’an Jiaotong University,2014,48(12):63-68.
[2]申昇,杨宏晖,王芸,等.联合互信息水下目标特征选择算法[J].西北工业大学学报,2015,33(4):639-643.Shen S,Yang H,Wang Y,et al.Joint mutual information feature selection for underwater acoustic targets[J].Journal of Northwestern Polytechnical University,2015,33(4):639-643.
[3]He X.Locality preserving projections[J].Advances in Neural Information Processing Systems,2005,45(1):186-197.
[4]Hou C,Nie F,Yi D,et al.Feature Selection via Joint Embedding Learning and Sparse Regression.[C]//IJCAI 2011,Proceedings of the,International Joint Conference on Artificial Intelligence,Barcelona,Catalonia,Spain,July.2011:1324-1329.
[5]Nie F,Huang H,Cai X,et al.Efficient and Robust Feature Selection via Joint l2,1-Norms Minimization.[C]//Advances in Neural Information Processing Systems 23:,Conference on Neural Information Processing Systems 2010.Proceedings of A Meeting Held 6-9 December 2010,Vancouver,British Columbia,Canada.2010:1813-1821.
[6]Yang Y,Shen H T,Ma Z,et al.l2,1-norm regularized discriminative feature selection for unsupervised learning[C]//International Joint Conference on Artificial Intelligence.2011:1589-1594.
[7]Li Z,Yang Y,Liu J,et al.Unsupervised feature selection using nonnegative spectral analysis.AAAI,2012.
[8]Mukhopadhyay S.Statistical Foundation of Spectral Graph Theory[J].Technical Report of Ieice Hip,2016,107:73-78.
[9]Yan S,Xu D,Zhang B,et al.Graph embedding and extensions:a general framework for dimensionality reduction.[J].IEEE Trans.pattern Anal.mach.intell,2007,29(1):40--51.
[2]申昇,杨宏晖,王芸,等.联合互信息水下目标特征选择算法[J].西北工业大学学报,2015,33(4):639-643.Shen S,Yang H,Wang Y,et al.Joint mutual information feature selection for underwater acoustic targets[J].Journal of Northwestern Polytechnical University,2015,33(4):639-643.
[3]He X.Locality preserving projections[J].Advances in Neural Information Processing Systems,2005,45(1):186-197.
[4]Hou C,Nie F,Yi D,et al.Feature Selection via Joint Embedding Learning and Sparse Regression.[C]//IJCAI 2011,Proceedings of the,International Joint Conference on Artificial Intelligence,Barcelona,Catalonia,Spain,July.2011:1324-1329.
[5]Nie F,Huang H,Cai X,et al.Efficient and Robust Feature Selection via Joint l2,1-Norms Minimization.[C]//Advances in Neural Information Processing Systems 23:,Conference on Neural Information Processing Systems 2010.Proceedings of A Meeting Held 6-9 December 2010,Vancouver,British Columbia,Canada.2010:1813-1821.
[6]Yang Y,Shen H T,Ma Z,et al.l2,1-norm regularized discriminative feature selection for unsupervised learning[C]//International Joint Conference on Artificial Intelligence.2011:1589-1594.
[7]Li Z,Yang Y,Liu J,et al.Unsupervised feature selection using nonnegative spectral analysis.AAAI,2012.
[8]Mukhopadhyay S.Statistical Foundation of Spectral Graph Theory[J].Technical Report of Ieice Hip,2016,107:73-78.
[9]Yan S,Xu D,Zhang B,et al.Graph embedding and extensions:a general framework for dimensionality reduction.[J].IEEE Trans.pattern Anal.mach.intell,2007,29(1):40--51.