改进MFCC算法在朱鹮鸣声个体识别中的应用

J4 ›› 2016, Vol. 38 ›› Issue (05): 1052-1056.

改进MFCC算法在朱鹮鸣声个体识别中的应用

王民，曹清菁，贠卫国，周军妮

（西安建筑科技大学信息与控制工程学院,陕西西安 710055）

收稿日期:2015-04-28 修回日期:2015-08-21 出版日期:2016-05-25 发布日期:2016-05-25
基金资助:
陕西省教育厅专项基金（2013JK1081）；陕西省科学技术研究发展计划项目（CXY1122(2)）；陕西省自然科学基金青年基金（2013JQ8003）；陕西省教育厅科研计划项目（12JK1007）

Application of an improved MFCC algorithm
in the identification of individual crested ibis

WANG Min，CAO Qingjing,YUN Weiguo,ZHOU Junni

(School of Information and Control Engineering,Xi’an University of Architecture and Technology,Xi’an 710055,China)

Received:2015-04-28 Revised:2015-08-21 Online:2016-05-25 Published:2016-05-25

摘要/Abstract

摘要：

朱鹮鸣声信号具有非平稳性，针对FFT不能反映信号的瞬时性以及无法摆脱基函数的问题，提出了一种基于HHT变换的MFCC参数特征提取算法，通过对信号进行EMD分解，得到每一帧固有模态函数后进行HT变换，频率合成后的边际谱通过Mel滤波器；然后取对数能量，经过DCT变换后得到改进的MFCC系数，采用高斯混合模型分别在纯净和加噪两种环境以及不同信噪比下进行朱鹮鸣声的个体识别。实验结果表明，改进算法不仅能更好地体现鸣声信号的瞬时性，朱鹮个体的平均识别率也提高了4%。

关键词: 朱鹮个体识别, FFT, MFCC算法, HHT算法

Abstract:

Given that the signals of crested ibis’s chirp are nonstationary, and the FFT cannot reflect transient signals or get rid of the basis function, we propose a feature extraction algorithm based on the HillberHuang transform (HHT) MFCC parameters. Each frame of the intrinsic mode function is obtained through the EMD decomposition of the signals, on which the HT transform is then conducted. The Hilbert marginal spectrum of the frequency synthesis passes though the Mel filter, and then the logarithmic energy is obtained. The MFCC coefficients are improved after the DCT transform, and the crested ibis individual can be recognized through the Gauss mixture model in pure and noisy environment respectively under different signaltonoise ratios. Experimental results show that the improved algorithm can reflect transient chirp signals better, and the average recognition rate of the crested ibis individuals is also increased by 4%.

Key words: crested ibis individual identification;FFT;MFCC algorithm;HHT algorithm

王民，曹清菁，贠卫国，周军妮. 改进MFCC算法在朱鹮鸣声个体识别中的应用[J]. J4, 2016, 38(05): 1052-1056.

WANG Min，CAO Qingjing,YUN Weiguo,ZHOU Junni. Application of an improved MFCC algorithm
in the identification of individual crested ibis [J]. J4, 2016, 38(05): 1052-1056.

参考文献［15］

［1］	Guo Min,Wu Xiaomin,Ren Jianshe,et al.A study on the crying feature of crested ibis(Nipponia nippon) in different growing periods ［J］.Journal of Northwest University(Natural Science Edition),2007,37(5):724728.(in Chinese)
［2］	Hu Zhengquan, Zeng Yumin, Zong Yuan, et al. Improvement of MFCC parameters extraction in speaker recognition［J］.Computer Engineering and Applications,2014,50(7):217220.(in Chinese)
［3］	Tu Binbin,Yu Fengqin.Speech emotion recognition based on improved MFCC with EMD［J］.Computer Engineering and Applications,2012,48(18):119122.(in Chinese)
［4］	Tang Baoping, Zhong Youming, Cheng Fabin, et al.Research on nonstationary signal analyzer based on HHT ［J］.Chinese Journal of Scientific Instrument,2007,28(1):2933. (in Chinese)
［5］	Wang Mingyang, Liu Zheng, Zhou Yiyu. Impulse radio signal detection based on HilberHuang transform［J］.Signal Processing,2006,22(4):581584.(in Chinese)
［6］	Li Tianyun, Cheng Siyong, Yang Mei.Power system hamonic analysis based on HillberHuang transform［J］.Proceedings of the CSEE,2008,28(4):109113.(in Chinese)
［7］	Xiang Li, Xin Li. Speech emotion recognition using novel HHTTEO based features［J］.Journal of Computers,2011,6(5):989998.
［8］	Li Lin,Ji Hongbing.Signal feature extraction based on an improved EMD method［J］.Measurement,2009,42(5):796803.
［9］	Zhang Haiyan,Fan Shixuan, Lü Donghui. HilbertHuang transform method of multimode Lamb wave extraction ［J］.Journal of Vibration, Measurement & Diagnosis,2008,28(3):216.(in Chinese)
［10］	Liu Cuiwei,Li Yuxing,Meng Lingya,et al.Timefrequency analysis of acoustic leakage signal for natural gas pipelines based on HilbertHuang transform［J］.Journal of Vibration and Shock,2014,33(16):4249.(in Chinese)
［11］	Raji A S, Firoz A S, Shijo M J, et al. Singleword query recognition a speech corpusbased method based on the combined technique of FFT and DCT［J］.Journal of Computer Science and Control Systems,2010,3(2):89.
［12］	Gai Qiang,Zhang Haiyong,Xu Xiaogang.Study of adaptive frequency multiresolution analysis of the HilberHuang transform［J］.Acta Electronica Sinica,2005,33(3):563566.(in Chinese)
［13］	Zhang Zhen, Wang Huaqing. Improved algorithm of Melfrequence cepstral coefficients in characteristics extraction based on voice signal［J］.Computer Engineering and Applications,2008,44(22):5455.(in Chinese)
［14］	Du Xiaoqing,Yu Fengqin.Speaker recognition based on vocal mechanism and human ear perceptual characteristic［J］.Computer Engineering,2013,39(11):197199.(in Chinese)
［15］	Zhu Leqing, Zhang Zhen. Automatic recognition of insect sounds using MFCC and GMM［J］.Acta Entomologica Sinica,2012,55(4):466471.(in Chinese)
	附中文参考文献：
［1］	郭敏,吴晓民,任建设,等.朱鹮不同生长发育期鸣声特征［J］.西北大学学报（自然科学版）,2007,37(5):724728.
［2］	胡政权,曾毓敏,宗原,等.说话人识别中MFCC参数提取的改进［J］.计算机工程与应用,2014,50(7):217220.
［3］	屠彬彬,于凤芹.基于EMD的改进MFCC的语音情感识别［J］.计算机工程与应用,2012,48(18):119122.
［4］	汤宝平,钟佑明,程发斌，等.基于HHT的非平稳信号分析仪的研究［J］.仪器仪表学报,2007,28(1):2933.
［5］	王明阳,柳征,周一宇.基于希尔伯特黄变换的冲击无线电信号检测［J］.信号处理,2006,22(4):581584.
［6］	李天云,程思勇,杨梅.基于希尔伯特黄变换的电力系统谐波分析［J］.中国电机工程学报,2008,28(4):109113.
［9］	张海燕,樊仕轩,吕东辉.多模式兰姆波走时提取的希尔伯特黄变换方法［J］.振动、测试与诊断,2008,28(3):216.
［10］	刘翠伟, 李玉星, 孟令雅，等. 基于希尔伯特黄变换的输气管道泄漏音波时频特性分析［J］.振动与冲击,2014,33(16):4249.
［12］	盖强,张海勇,徐晓刚.HilbertHuang变换的自适应频率多分辨分析研究［J］.电子学报,2005,33(3):563566.
［13］	张震,王化清.语音信号特征提取中Mel倒谱系MFCC的改进算法［J］.计算机工程与用,2008,44(22):5455.
［14］	杜晓青,于凤芹.基于发声机理与人耳感知特性的说话人识别［J］.计算机工程,2013,39(11):197199.
［15］	竺乐庆,张真.基于MFCC和GMM的昆虫声音自动识别［J］.昆虫学报,2012,55(4):466471.

[1]	牟胜梅杨晓东. 高吞吐率浮点FFT处理器的FPGA实现研究[J]. J4, 2008, 30(7): 98-99.
[2]	陈勤刘茵王涛. 一种基于遗传算法的FFT Snake模型图像分割方法[J]. J4, 2007, 29(5): 66-68.