面向不均匀飞机舱音信号样本的模糊支持向量机识别方法

电子与自动控制

本期目录 | 过刊浏览 | 高级检索

前一篇 | 后一篇

面向不均匀飞机舱音信号样本的模糊支持向量机识别方法

杨琳¹, 王从庆², 姜龙生²

1. 中国民航科学技术研究院, 北京 100028;
2. 南京航空航天大学自动化学院, 江苏南京 210016

收稿日期:2011-04-19 修回日期:2011-06-03 出版日期:2012-03-25 发布日期:2012-03-24
通讯作者: 王从庆 E-mail:cqwang@nuaa.edu.cn
基金资助:
国家自然科学基金(60776819)

Signal Recognition of Imbalanced Samples for CVR Based on Fuzzy SVM

YANG Lin¹, WANG Congqing², JIANG Longsheng²

1. China Academy of Civil Aviation Science and Technology, Beijing 100028, China;
2. College of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China

Received:2011-04-19 Revised:2011-06-03 Online:2012-03-25 Published:2012-03-24

摘要/Abstract

摘要： 针对驾驶舱话音记录器(CVR)中记录的舱音背景信息多而复杂、频率范围宽、非平稳等特点,通过对15种舱音信息进行傅里叶变换和小波包变换,依次提取其Mel倒谱系数(MFCC)和小波包分解系数(WPC),利用距离可分性判据对MFCC和WPC信息进行压缩融合,得到舱音信息特征向量。设计了面向不均衡样本的模糊支持向量机(FSVM),分别计算每种类别样本及其内每种舱音信息的2个隶属度,然后利用FSVM对舱音信号进行分类识别,解决了CVR信号含噪奇异样本和数目不均衡样本时识别性能较差的缺点,实验表明该方法明显优于常规支持向量机(SVM)和FSVM,分类识别率达到98.33%。

关键词: 驾驶舱话音记录器, 非话语背景声音信号, 特征提取, 不均匀舱音样本, 模糊支持向量机

Abstract: The voice signals in a cockpit voice recorder(CVR)are complex, non-stationary, and exist in a wide frequency range. By using Fourier transform and wavelet packet transform for the fifteen kinds of signals in a CVR, Mel frequency cepstrum coefficient (MFCC) and wavelet packet coefficient (WPC) are extracted as the initial characteristic samples. The characteristic vectors are determined by compression of the MFCC and WPC samples using a geometric distance classification criterion. A fuzzy support vector machine (FSVM) is designed to handle the imbalanced sample classification in the CVR, in which two different fuzzy-membership values in relation to the imbalanced samples are calculated by the extracted samples in each voice signal. The above method can improve the recognition performance of the voice signals with imbalanced samples in the presence of outliers and noise. The experimental results show that it is obviously superior to the conventional support vector machine (SVM) and FSVM with a 98.33% recognition rate.

Key words: cockpit voice recorder, non-voice acoustic signal, feature extraction, imbalanced sample, fuzzy support vector machine

中图分类号:

杨琳, 王从庆, 姜龙生. 面向不均匀飞机舱音信号样本的模糊支持向量机识别方法[J]. 航空学报, 2012, (3): 544.

YANG Lin, WANG Congqing, JIANG Longsheng. Signal Recognition of Imbalanced Samples for CVR Based on Fuzzy SVM[J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2012, (3): 544.

参考文献

[1] Grossi D R. Aviation recorder overview. NTSB Journal of Accident Investigation, 2006, 2(1): 31-42.
[2] Cheng D L, Yi C J, Yao H Y, et al. Initial study on airplane cockpit voice recorder (CVR)' sound information identification methods. Noise and Vibration Control, 2006, 6(3): 81-84. (in Chinese) 程道来, 仪垂杰, 姚红宇, 等. 飞机舱音记录器声信息识别方法的初步研究. 噪声与振动控制, 2006, 6(3): 81-84.
[3] Li X R, Du J, Zhang P. Cockpit voice system and its applications Beijing: National Defense Industry Press, 2010: 107-108. (in Chinese) 李学仁, 杜军, 张鹏. 飞机舱音系统及其应用技术. 北京: 国防工业出版社, 2010: 107-108.
[4] Yang L, Wang C Q, Wang Z G, et al. Blind separation performance of non-voice acoustic signal for CVR. Journal of Nanjing University of Aeronautics & Astronautics, 2010, 42(2): 185-190. (in Chinese) 杨琳, 王从庆, 王芝刚, 等. 飞机舱音记录器非话语信号盲分离性能. 南京航空航天大学学报, 2010, 42(2): 185-190.
[5] Cheng D L, Yi C J, Guo H R, et al. Wavelet analysis on the voice and switch sound recorded by airplane cockpit voice recorder. Computer Engineering and Applications, 2006, 42(25): 187-196. (in Chinese) 程道来, 仪垂杰, 郭海荣, 等.飞机舱音记录器中话语声和开关声小波分析. 计算机工程与应用, 2006, 42(25): 187-196.
[6] Cheng H B, Lei M, Li X R, et al. Robust speech endpoint detection in aircraft cockpit voice recorder information. Journal of Data Acquisition & Processing, 2010, 25(2): 223-227. (in Chinese) 程洪炳, 雷鸣, 李学仁, 等.飞机舱音信息鲁棒语音端点检测. 数据采集与处理, 2010, 25(2): 223-227.
[7] Yi C J, Cheng D L, Guo J X, et al. Analyses on characteristic frequencies of background sound in cockpit of airplane based on different methods. Journal of Vibration and Shock, 2007, 26(8): 109-111. (in Chinese) 仪垂杰, 程道来, 郭健翔, 等. 基于不同方法的飞机舱音背景声频谱特征的获取与分析. 振动与冲击, 2007, 26(8): 109-111.
[8] Cheng D L, Yi C J, Yao H Y. Identification and characteristics analysis for airplane overspeed audio frequency warning signal. Journal of Harbin Institute of Technology, 2009, 41(12): 216-218. (in Chinese) 程道来, 仪垂杰, 姚红宇. 飞机超速音频警告信号的提取与特征分析. 哈尔滨工业大学学报, 2009, 41(12): 216-218.
[9] Lu J, Chen Y S, Sun Z X, et al. Automatic audio classification by using hidden Markov model. Journal of Software, 2002, 13(8): 1593-1597. (in Chinese) 卢坚, 陈毅松, 孙正兴, 等. 基于隐马尔可夫模型的音频自动分类. 软件学报, 2002, 13(8): 1593-1597.
[10] Lin C F, Wang S D. Fuzzy support vector machines. IEEE Transactions on Neural Network, 2002, 13(2): 464-471.
[11] Jiang L S, Wang C Q. Based on MFCC and wavelet packet for aircraft cockpit voice feature extraction. CGNCC2010. 2010: 1032-1037. (in Chinese) 姜龙生, 王从庆. 基于MFCC和小波包变换的飞机舱音特征提取. 2010年中国导航、制导与控制学术会议. 2010: 1032-1037.
[12] Mao K Z. Feature subset selection for support vector machines through discriminative function pruning analysis. IEEE Transactions on Systems, Man and Cybernetics, 2004, 34(1): 60-67.
[13] Jiang C S, Wang C Q, Wei H K, et al. Intelligent control and application.Beijing: Science Press, 2007: 75-76. (in Chinese) 姜长生, 王从庆. 魏海坤, 等. 智能控制与应用. 北京: 科学出版社, 2007: 75-76.
[14] Veropoulos K, Campbell C, Cristianini N. Controlling the sensitivity of support vector machines. Proceedings of the International Joint Conference on AI. 1999: 55-60.
[15] Akbani R, Kwek S, Japkowicz N. Applying support vector machines to imbalanced datasets. Proceedings of the 15th European Conference on Machine Learning. 2004: 39-50.

E-mail：hkxb@buaa.edu.cn

关于我们

期刊社服务

专业学科

封面文章

友情链接

主管单位：中国科学技术协会主办单位：中国航空学会北京航空航天大学

面向不均匀飞机舱音信号样本的模糊支持向量机识别方法

Signal Recognition of Imbalanced Samples for CVR Based on Fuzzy SVM

PDF (PC)

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

[1]	郎荣玲, 魏才伦, 范亚, 高飞. 基于稀疏点匹配的协同式未知目标跟踪方法[J]. 航空学报, 2026, 47(3): 632425-632425.
[2]	唐彬, 杨小冈, 卢瑞涛, 张震宇, 宿爽. 基于图像翻译的飞行器红外/卫星异源快速匹配定位方法[J]. 航空学报, 2025, 46(23): 631961-631961.
[3]	刘奇, 裴智翔, 惠乐, 何明一, 戴玉超. 基于双分支特征聚合的无人机视觉位置识别[J]. 航空学报, 2025, 46(23): 632457-632457.
[4]	赵辰豪, 吴德伟, 何晶, 吴倩. 一种无人机视觉位姿估计的语义特征匹配算法[J]. 航空学报, 2025, 46(2): 330406-330406.
[5]	金栢成, 田阔, 黄蕾. 曲面加筋结构拓扑优化结果参数化重构方法[J]. 航空学报, 2024, 45(24): 630586-630586.
[6]	褚昭晨, 宋韬, 金忍, 林德福. 基于视觉图像的空对空多无人机目标跟踪[J]. 航空学报, 2024, 45(14): 629379-629379.
[7]	张百川, 毕文豪, 张安, 毛泽铭, 杨咪. 基于Transformer模型的空战飞行器轨迹预测误差补偿方法[J]. 航空学报, 2023, 44(9): 327413-327413.
[8]	王金瑞, 季珊珊, 张宗振, 初振云, 韩宝坤, 鲍怀谦. 并行稀疏滤波在轴承声信号下的故障诊断[J]. 航空学报, 2023, 44(4): 426887-426887.
[9]	韩淞宇, 邵海东, 姜洪开, 张笑阳. 基于提升卷积神经网络的航空发动机高速轴承智能故障诊断[J]. 航空学报, 2022, 43(9): 625479-625479.
[10]	杨特, 杨智春, 梁舒雅, 康在飞, 贾有. 平稳随机载荷的信号特征提取与深度神经网络识别[J]. 航空学报, 2022, 43(9): 225952-225952.
[11]	战庆亮, 白春锦, 张宁, 葛耀君. 基于时程卷积自编码的机翼绕流特征识别方法[J]. 航空学报, 2022, 43(11): 526531-526531.
[12]	陈呈, 赵丹, 王岳青, 邓亮, 杨超, 苏铖宇, 王昉. NNW-TopViz流场可视分析系统[J]. 航空学报, 2021, 42(9): 625747-625747.
[13]	王怡星, 韩仁坤, 刘子扬, 张扬, 陈刚. 流体力学深度学习建模技术研究进展[J]. 航空学报, 2021, 42(4): 524779-524779.
[14]	张伟伟, 寇家庆, 刘溢浪. 智能赋能流体力学展望[J]. 航空学报, 2021, 42(4): 524689-524689.
[15]	张凯, 王凯迪, 杨曦, 李少毅, 王晓田. 基于DNET的空中红外目标抗干扰识别算法[J]. 航空学报, 2021, 42(2): 324223-324223.