[1]齐晓轩,都 丽,张国山.小波包近似熵特征的机动车声识别方法[J].南京理工大学学报(自然科学版),2020,44(01):67-73.[doi:10.14177/j.cnki.32-1397n.2020.44.01.011]
 Qi Xiaoxuan,Du Li,Zhang Guoshan.Vehicle type recognition by acoustic signal based on waveletpacket decomposition and approximate entropy[J].Journal of Nanjing University of Science and Technology,2020,44(01):67-73.[doi:10.14177/j.cnki.32-1397n.2020.44.01.011]
点击复制

小波包近似熵特征的机动车声识别方法()
分享到:

《南京理工大学学报》(自然科学版)[ISSN:1005-9830/CN:32-1397/N]

卷:
44卷
期数:
2020年01期
页码:
67-73
栏目:
出版日期:
2020-02-29

文章信息/Info

Title:
Vehicle type recognition by acoustic signal based on waveletpacket decomposition and approximate entropy
文章编号:
1005-9830(2020)01-0067-07
作者:
齐晓轩12都 丽1张国山2
1.沈阳大学 信息工程学院,辽宁 沈阳 110044; 2.天津大学 电气与自动化工程学院,天津 300072
Author(s):
Qi Xiaoxuan12Du Li1Zhang Guoshan2
1.School of Information Engineering,Shenyang University,Shenyang 110044,China; 2.School of Electrical Engineering & Automation School,Tianjin University,Tianjin 300072,China
关键词:
车型识别 声信号 近似熵 小波包分解 支持向量机
Keywords:
vehicle type recognition acoustic signal approximate entropy wavelet packet decomposition support vector machine
分类号:
TP391
DOI:
10.14177/j.cnki.32-1397n.2020.44.01.011
摘要:
为了提高复杂背景噪声环境下的车型识别准确性,该文基于近似熵理论,对机动车行驶中辐射的声信号进行了研究。近似熵具有抗干扰能力强的特点,可用于提取动态背景噪声下机动车声信号的车型特征信息。首先,对声信号进行3层小波包分解; 然后,利用近似熵量化第3层上各子频带信号的不规则性,描述各频带之间不同的变化趋势并作为目标车辆的声特征。为了提高分类有效性,将分解后的8个子频带信号的近似熵邻比值作为信号的特征向量,并基于支持向量机分类器实现了车型识别。分别在正常和有风两种气候条件下进行了实验,基于小波包近似熵的车型特征均获得了较为理想的分类精度。实验结果显示,小波包近似熵特征能有效地应用于机动车的声识别且对气候的影响具有一定的鲁棒性。
Abstract:
In order to improve the accuracy of vehicle type recognition under the condition of strong noise interference,the approximate entropy theory is introduced in this paper to analyse radiated acoustic signals from moving vehicles. Approximate entropy exposes a strong anti-interference ability and can be utilized to extract the feature characteristics from acoustic signals with the interference of dynamic background noise. Firstly,a three-layer wavelet packet decomposition method is utilized to analyse acoustic signals to obtain eight sub-frequent band signals. Then,the irregularity of each sub-frequent band signal is quantified by approximate entropy. Finally,the obtained adjacent ratio of these approximate entropies are used as the feature vector of target vehicles,which is input to the support vector machine for automatic classification of vehicle types. Experiments are conducted under normal conditions as well as strong windy conditions,feature characteristics baed on the wavelet decomposition approximate entropy could get satisfactory classification accuracy. The results show that the proposed method can identify vehicle types by acousitc signals effectively and is robust especially for windy cases.

参考文献/References:

[1] 张文斌,王祁,马松林,等. 一种新颖的多传感器行驶车辆分类系统[J]. 天津大学学报,2008,41(2):194-198.
Zhang Wenbin,Wang Qi,Ma Songlin,et al. A novel multisensor system for moving vehicle classification[J]. Journal of Tianjin University,2008,41(2):194-198.
[2]马侠霖,蔡铭,丁建立. 基于频谱分析与支持向量机的车型音频识别研究[J]. 应用声学,2014,33(4):371-376.
Ma Xialin,Cai Ming,Ding Jianli. Audio identification of vehicle type based on frequency analysis and support vector machine[J]. Journal of Applied Acoustics,2014,33(4):371-376.
[3]齐晓轩,纪建伟,韩晓微. 机动车声信号能量特征的小波包提取[J]. 沈阳农业大学学报,2011,42(3):372-375.Qi Xiaoxuan,Ji Jianwei,Han Xiaowei. Energy feature extraction of vehicle acoustic signal by wavelet packet[J]. Journal of Shenyang Agricultural University,2011,42(3):372-375.
[4]田寅,董宏辉,贾利民,等. 用于车辆分类的多传感器车型特征融合算法[J]. 华南理工大学学报(自然科学版),2014,42(3):52-58.
Tian Yin,Dong Honghui,Jia Limin,et al. Multi-sensor signature fusion algorithm for vehicle type classification[J]. Journal of South China University of Technology(Natural Science Edition),2014,42(3):52-58.
[5]胡彬,赵春霞,郭剑辉,等. 基于主被动传感器融合的车辆检测[J]. 华中科技大学学报(自然科学版),2014,42(12):128-132.
Hu Bin,Zhao Chunxia,Guo Jianhui,et al. Vehicle detection using active and passive sensors[J]. Journal of Huazhong University of Science and Technology(Nature Science Edition),2014,42(12):128-132.
[6]Bensrhair A,Bertozzi M,Broggi A,et al. A cooperative approach to vision-based vehicle detection[C]//Proceedings of Intelligent Transportation System. Oakland,USA:IEEE,2001:207-212.
[7]Zu X,Zhang S,Guo F,et al. Vehicle counting and moving direction identification based on small-aperture microphone array[J]. Sensors,2017,17(5):1089-1100.
[8]Pincus S M. Approximate entropy as a measure of system complexity[J]. Proceedings of the National Academy of Sciences of the United States of America,1991,88(6):2297-2301.
[9]黄娟,黄纯,江亚群,等. 基于小波包近似熵的线路故障性质辨识方法[J]. 仪器仪表学报,2012,33(9):2009-2015.
Huang Juan,Huang Chun,Jiang Yaqun,et al. Identification method of fault characteristics in transmission lines based on wavelet packet and approximate entropy[J]. Chinese Journal of Scientific Instrument,2012,33(9):2009-2015.
[10]赵荣珍,于昊,徐继刚. 基于LMD近似熵与HMM的转子故障诊断方法[J]. 兰州理工大学学报,2012,38(6):24-29.
Zhao Rongzhen,Yu Hao,Xu Jigang. Fault diagnosis approach for rotor system based on LMD-approximate entropy and HMM[J]. Journal of Lanzhou University of Technology,2012,38(6):24-29.
[11]谢中凯,刘国华. 近似熵在混凝土结构损伤识别中的应用[J]. 浙江大学学报(工学版),2013,47(3):456-464.Xie Zhongkai,Liu Guohua. Application of approximate entropy in concrete structures damage identification[J]. Journal of Zhejiang University(Engineering Science),2013,47(3):456-464.
[12]杨青,孙佰聪,朱美臣,等. 基于小波包熵和聚类分析的滚动轴承故障诊断方法[J]. 南京理工大学学报,2013,37(4):517-523.
Yang Qing,Sun Baicong,Zhu Meichen,et al. Rolling bearing fault diagnosis method based on wavelet packet entropy and clustering analysis[J]. Journal of Nanjing University of Science and Technology,2013,37(4):517-523.
[13]向玲,郭鹏飞,高楠,et al. 基于IITD和FCM聚类的滚动轴承故障诊断[J]. 航空动力学报,2018,33(10):258-265.
Xiang Ling,Guo Pengfei,Gao Nan,et al. Rolling bearing fault diagnosis based on IITD and FCM clustring[J]. Journal of Aerospace Power,2018,33(10):258-265.
[14]杜云霄,王殊轶,Zaphlene N N,et al. 基于表面肌电信号的颈部肌肉疲劳评价算法比较研究[J]. 生物医学工程学杂志,2018,2(35):31-37.
Du Yunxiao,Wang Shuyi,Zaphlene N N,et al. Comparative study on evaluation algorithms for neck muscle fatigue based on surface electromyography signal[J]. Journal of Biomedical Engineering,2018,2(35):31-37.
[15]郭凤仪,李坤,陈昌垦,et al. 基于小波近似熵的串联电弧故障识别方法[J]. 电工技术学报,2016,31(24):168-176.
Guo Fengyi,Li Kun,Chen Changken,et al. Series arc fault identification method based on wavelet approximate entropy[J]. Transactions of China Electrotechnical Society,2016,31(24):168-176.
[16]Avic E,Akpolat Z H. Speech recognition using a wavelet packet adaptive network based fuzzy inference system[J]. Expert Systems with Applications,2006,31(4):495-503.
[17]冯凯,应展烽,吴军基,等. 基于小波包变换和峰式马尔科夫链的风速短期预测[J]. 南京理工大学学报,2014,38(5):639-643,657.
Feng Kai,Ying Zhanfeng,Wu Junji,et al. Short-term wind speed forecast based on wavelet packet decomposition and peak-type Markov chain[J]. Journal of Nanjing University of Science and Technology,2014,38(5):639-643,657.
[18]Vapnik V N. The nature of statistical learning theory[M]. New York,USA:Springer-Verlag,1995.
[19] 郑怀兵,翟济云. 基于视频分析的森林火灾烟雾检测方法[J]. 南京理工大学学报,2015,39(6):686-691,710.
Zheng Huaibing,Zhai Jiyun. Forest fire smoke detection based on video analysis[J]. Journal of Nanjing University of Science and Technology,2015,39(6):686-691,710.
[20]聂伟荣,朱继南,夏虹. 地面运动目标分类的模式特征与评价[J]. 探测与控制学报,2002,24(3):25-28,32.
Nie Weirong,Zhu Jinan,Xia Hong. Feature and evaluation of the moving target on the ground[J]. Journal of Detection & Control,2002,24(3):25-28,32.
[21]赵宏旭,杨文帅. 基于短时能量和梅尔倒谱系数的车型音频识别[J]. 科学技术与工程,2018,18(18):197-201.
Zhao Hongxu,Yang Wenshuai. Vehicle recognition by acoustic signals based on short-time energy and Mel cepstrum coefficient[J]. Science Technology and Engineering,2018,18(18):197-201.

备注/Memo

备注/Memo:
收稿日期:2016-09-29 修回日期:2019-05-12
基金项目:国家自然科学基金(61473202); 辽宁省重点研发计划项目基金(2018104012); 沈阳市高层次人才项目基金(RC180362)
作者简介:齐晓轩(1974-),女,博士,副教授,主要研究方向:信号处理、故障诊断及预测,E-mail:qi_xx@aliyun.com。
引文格式:齐晓轩,都丽,张国山. 小波包近似熵特征的机动车声识别方法[J]. 南京理工大学学报,2020,44(1):67-73.
投稿网址:http://zrxuebao.njust.edu.cn
更新日期/Last Update: 2020-02-29