基于改进测地距离的等度规映射及其在声目标特征提取中的应用

doi:10.3969/j.issn.1000-1093.2012.10.005

兵工学报 ›› 2012, Vol. 33 ›› Issue (10): 1178-1184.doi: 10.3969/j.issn.1000-1093.2012.10.005

基于改进测地距离的等度规映射及其在声目标特征提取中的应用

刘辉，杨俊安，王一，蔡学良

(解放军电子工程学院安徽省电子制约技术重点实验室，安徽合肥 230037)

收稿日期:2010-06-25 修回日期:2010-06-25 上线日期:2014-03-04
作者简介:刘辉(1983—)，男，讲师，博士
基金资助:
国家自然科学基金项目（60872113）

An Improved Isometric Mapping Algorithm Based on New Geodesic Distance and Its Application in theFeature Extraction of Acoustic Targets

LIU Hui, YANG Jun-an, WANG Yi, CAI Xue-liang

(Key Laboratory of Electronic Restriction of Anhui Province, Electronic Engineering Institute of PLA,Hefei 230037, Anhui, China)

Received:2010-06-25 Revised:2010-06-25 Online:2014-03-04

摘要/Abstract

摘要： 经典等度规映射(ISOMAP)算法由于其固有的拓扑不稳定性，容易受噪声影响出现“短路边”，导致其在声目标特征提取的应用中性能不佳。提出了一种基于改进测地距离的ISOMAP算法，将近邻图的构造看作是建立一个电路模型，以电路中各节点电容从初始阶段到一定的状态所需要的时间为测地距离的量度，将保持局部结构的鲁棒性与保持全局几何结构的准确性结合在一起，克服了噪声短路点对算法的影响，提高了算法性能。在SensIT实验数据和外场实际采集数据上的实验结果表明，基于改进测地距离的ISOMAP算法的准确性和鲁棒性都有了较大提高。

关键词: 信息处理技术, 目标识别, 流形学习, 等度规映射算法, 拉普拉斯特征映射算法

Abstract: A drawback of the isometric mapping (ISOMAP) algorithm is topological instability, that is, the algorithm is sensitive to the noise, which makes ISOMAP do a bad performance when used in the feature extraction of acoustic targets in the battlefield. To solve the problem, this paper proposes a new method for more robust approximation of the geodesic distances of ISOMAP using a viewpoint of electric circuits. The charge times are used to construct new distance measures. In this way, we achieve both the stability of local methods and the global approximation property of global methods. Experimental results prove that the new distance measures can achieve significant advancement over the former method, which can improve the system performance effectively.

Key words: information processing, acoustic target recognition, manifold learning, isometric mapping arithmetic, Laplacian eigenmap arithmetic

中图分类号:

TB56

刘辉，杨俊安，王一，蔡学良. 基于改进测地距离的等度规映射及其在声目标特征提取中的应用[J]. 兵工学报, 2012, 33(10): 1178-1184.

LIU Hui, YANG Jun-an, WANG Yi, CAI Xue-liang. An Improved Isometric Mapping Algorithm Based on New Geodesic Distance and Its Application in theFeature Extraction of Acoustic Targets[J]. Acta Armamentarii, 2012, 33(10): 1178-1184.

参考文献

［1］ Liu H, Yang J A. A novel acoustic feature extraction algorithm for robust low altitude flying target recognition[C]∥2009 International conference on Artificial intelligence and Computational Intelligence. Jinan: IEEE, 2009,11(2):185-188.
［2］陈克安, 王娜, 伍莹，等. 基于音色属性辨识声目标的主观评价实验研究[J]. 科学通报, 2010,55(8):651-659.
CHEN Ke-an, WANG Na, WU Ying, et al. Subiective evaluation experiments of timbre attribute based acoustic tareet identification[J]. Chinese Science Bulletin, 2010,55(8): 651-659. (in Chinese)
［3］马元锋, 陈克安, 马苗. 一种新的可应用于声目标识别的倒谱系数[J]. 兵工学报, 2009,30(11):1477-1483.
MA Yuan-feng, CHEN Ke-an, MA Miao. A new cepstrum coefficients applied to acoustic target recognition[J]. Acta Armamentarii, 2009, 30(11):1477-1483. (in Chinese)
［4］黄琦. 智能传感器侦察网络中的地面目标识别算法研究[D]. 合肥: 中国科学技术大学，2006.
HUANG Qi. Research on the recognition of ground acoustic targets in smart sensor network[D]. Hefei: University of Science and Technology of China, 2006. (in Chinese)
［5］侯臣平, 吴诩, 易东云. 新的流形学习方法统一框架及改进的拉普拉斯特征映射方法[J]. 计算机研究与发展, 2009,46(4):676-682.
HOU Chen-ping, WU Yi, YI Dong-yun. A novel unified manifold learning framework and an improved laplacian eigenmap[J]. Journal of Computer Research and Development, 2009,46(4):676-682. (in Chinese)
［6］ Viren J, Lawrence K S. Exploratory analysis and visualization of speech and music by locally linear embedding [J]. ICASSP, 2004,3: 984-987.
［7］ Errity A, Mckenna J, Kirkpatrick B. Manifold learning-based feature transformation for phone classification[C]∥International Conference on Non-Linear Speech Processing. Dublin: IEEE, 2007, 4885:132-141.
［8］ Errity A. Exploring the dimensionality of speech using manifold learning and dimensionality reduction methods[D]. Dublin: Dublin City University, 2010.
［9］管鲁阳, 鲍明, 张鹏, 等. 基于流形学习的单类分类算法及其在不均衡声目标识别中的应用[J]. 声学学报, 2009, 34(1):67-72.
GUAN Lu-yang, BAO Ming, ZHANG Peng, et al. One-class classification algorithm based on manifold learning and its application to imbalanced acoustic target recognition[J]. Journal of Acoustic, 2009, 34(1):67-72. (in Chinese)
［10］ Tenenbaum J B, De Silva, Langford J C. A global geometric framework for nonlinear dimensionality reduction[J]. Science, 2000, 290(5500):2319-2323.
［11］ Balasubramanian M, Schwartz E L. The ISOMAP algorithm and topological stability[J]. Science, 2002, 295(5552):7a.
［12］王庆刚. 流形学习算法及若干应用研究[D]. 重庆:重庆大学, 2009.
WANG Qing-gang. Study on algorithms and application in manifold learning[D]. Chongqing: Chongqing University, 2009. (in Chinese)
［13］ Belkin M, Niyogl P. Laplacian eigenmaps for dimensionality reduction and data representation[J]. Neural Computation, 2003, 15 (6):1373-1396.
［14］ Ham J, Lee D D. A kernel view of the dimensionality reduction of manifolds[C]∥Proceedings of ICML. US: ICML, 2004: 47-55.
［15］ Kondor I, Lafferty J. Diffusion kernels on graphs and other discrete structures[C]∥Proceedings of International Conference on Machine Learning. US: ICML, 2002:315-322.
［16］ Doyle P G, Snell J L. Random walks and electric networks [M]. New York: Mathematical Association of America, 2000.

[1]	宋晓茹, 刘康, 高嵩, 陈超波, 阎坤. 复杂战场环境下改进YOLOv5军事目标识别算法研究[J]. 兵工学报, 2024, 45(3): 934-947.
[2]	赵晓冬, 张洵颖. 基于YOLOv5的无人车自主目标识别优化算法[J]. 兵工学报, 2023, 44(9): 2732-2744.
[3]	刘冰, 郝新红, 周文, 杨瑾. 基于BAS-BPNN的调频无线电引信目标与扫频干扰识别方法[J]. 兵工学报, 2023, 44(8): 2391-2403.
[4]	霍健, 陈慧敏, 马云飞, 郭鹏宇, 杨旭, 孟祥盛. 基于MEMS激光雷达的车辆目标识别算法[J]. 兵工学报, 2023, 44(4): 940-948.
[5]	丁伯圣, 张睿恒, 徐立新, 陈慧敏. 基于灰度补偿和特征融合的沙尘图像修复方法[J]. 兵工学报, 2023, 44(10): 3115-3126.
[6]	王亮，陈建华，李烨. 一种基于深度学习的无人艇海上目标识别技术[J]. 兵工学报, 2022, 43(S2): 13-19.
[7]	李良福，陈卫东，高强，许开銮，刘轩，何曦，钱钧. 基于深度学习的光电系统智能目标识别[J]. 兵工学报, 2022, 43(S1): 162-168.
[8]	曹昭睿，郝永平，刘万成，白帆，孙颢洋，张慧，李宇海. 紧凑折反式仿生复眼及图像快速拼接识别算法[J]. 兵工学报, 2022, 43(8): 1845-1857.
[9]	薛灵芝, 曾向阳, 杨爽. 基于生成对抗网络的水声目标识别算法[J]. 兵工学报, 2021, 42(11): 2444-2452.
[10]	黄莎玲，朱鸿志，程祥利，谢玉斌. 自适应阈值侵彻引信层目标识别算法[J]. 兵工学报, 2020, 41(9): 1762-1771.
[11]	王健，秦春霞，杨珂，任萍. 基于导向重构与降噪稀疏自编码器的合成孔径雷达目标识别[J]. 兵工学报, 2020, 41(9): 1861-1870.
[12]	邹汝平，刘建书. 基于概率假设密度滤波与无迹Kalman滤波的多目标跟踪与识别[J]. 兵工学报, 2020, 41(8): 1502-1508.
[13]	曾溢良，蓝金辉，邹金霖. 滑动置信度约束的红外弱小目标跟踪算法研究[J]. 兵工学报, 2017, 38(9): 1771-1778.
[14]	李关防，崔杰，袁富宇. 基于线谱瞬时频率估计的被动声纳目标运动分析[J]. 兵工学报, 2017, 38(7): 1395-1401.
[15]	朱甦，薄煜明，何亮. 基于冗余字典的多特征压缩感知目标跟踪算法[J]. 兵工学报, 2017, 38(6): 1140-1146.

基于改进测地距离的等度规映射及其在声目标特征提取中的应用

An Improved Isometric Mapping Algorithm Based on New Geodesic Distance and Its Application in theFeature Extraction of Acoustic Targets

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价