[1]余凌浩,陆铁文,李 晨,等.基于子带谱熵法和PSO-GA-SVM的汽车鸣笛识别[J].南京师范大学学报(工程技术版),2021,21(02):027-33.[doi:10.3969/j.issn.1672-1292.2021.02.005]
　Yu Linghao,Lu Tiewen,Li Chen,et al.Car Whistle Recognition Based on Sub-Band SpectralEntropy Method and PSO-GA-SVM[J].Journal of Nanjing Normal University(Engineering and Technology),2021,21(02):027-33.[doi:10.3969/j.issn.1672-1292.2021.02.005]
点击复制

基于子带谱熵法和PSO-GA-SVM的汽车鸣笛识别
分享到：

南京师范大学学报（工程技术版）[ISSN:1006-6977/CN:61-1281/TN]

卷:
21卷
期数:
2021年02期
页码:
027-33
栏目:
信息与通信工程
出版日期:
2021-06-30

文章信息/Info

Title:
Car Whistle Recognition Based on Sub-Band SpectralEntropy Method and PSO-GA-SVM
文章编号:
1672-1292(2021)02-0027-07
作者:
余凌浩1; 陆铁文1; 李 晨1; 曾毓敏1; 袁 芳2
(1.南京师范大学计算机与电子信息学院/人工智能学院,江苏 南京 210023)(2.杭州爱华智能科技有限公司,浙江 杭州 311121)
Author(s):
Yu Linghao1; Lu Tiewen1; Li Chen1; Zeng Yumin1; Yuan Fang2
(1.School of Computer and Electronic Information/School of Artificial Intelligence,Nanjing Normal University,Nanjing 210023,China)(2.Hangzhou Aihua Intelligent Technology Co.,Ltd.,Hangzhou 311121,China)
关键词:
汽车鸣笛识别; 子带谱熵法; 支持向量机; 粒子群优化; 遗传算法
Keywords:
car whistle recognition; sub-band spectral entropy method; support vector machine; particle swarm optimization; genetic algorithm
分类号:
TN912.34
DOI:
10.3969/j.issn.1672-1292.2021.02.005
文献标志码:
A
摘要:
针对鸣笛抓拍系统会产生误判的问题,提出了一种基于子带谱熵法和支持向量机的汽车鸣笛识别算法. 首先,使用子带谱熵法对声音样本进行初判,将子带谱熵高于阈值的样本直接判定为非鸣笛样本. 然后,对初判为鸣笛的样本中的疑似鸣笛部分进行分割,并提取Mel频率倒谱系数作为声音的特征. 最后,使用支持向量机对分割结果进行进一步分类,并使用粒子群算法与遗传算法的融合来优化支持向量机的参数. 仿真结果表明,该算法具有较好的鲁棒性. 在对实际采集样本的鸣笛识别中,该算法也取得了较高的准确率.
Abstract:
In order to solve the problem of misjudgment in car whistle capture system,an algorithm based on sub-band spectral entropy and support vector machine is proposed in this paper. Firstly,sub-band spectral entropy method is used to preliminarily judge the sound samples. The samples whose sub-band spectral entropies are higher than a threshold value are directly determined as non-whistle samples. Then,the suspected whistle parts are segmented in the samples which are initially judged as whistle,and Mel frequency cepstrum coefficient is extracted as sound feature. Finally,the segmentation results are classified by support vector machine,and the parameters of support vector machine are optimized by the combination of particle swarm optimization and genetic algorithm. Simulation results show that the algorithm has good robustness. In the whistle recognition of actual collected samples,the algorithm also achieves a higher accuracy.

参考文献/References:

[1] 袁芳,闫建伟,张勇,等. 汽车鸣笛声实时抓拍的理论研究和系统实现[J]. 电声技术,2018,42(11):13-15,21.
[2]侍艳华,刘菁原,卞飞,等. 基于MFCC和CNN的汽车鸣笛声识别算法[J]. 电声技术,2020,44(5):30-33.
[3]郑皓,赵庶旭,屈睿涛. 一种用于城市交通的优化声音识别仿真[J]. 计算机技术与发展,2019,29(2):60-64.
[4]祁兵兵,刘金磊,张焕胜,等. 智能交通系统中声源端点识别算法[J]. 电声技术,2018,42(10):8-13,68.
[5]项彬,马石磊,陈建峰,等. 一种基于支持向量机的汽车鸣笛分类方法:201811405638.8[P]. 2019-04-12.
[6]蒋翠清,邵宏波. 基于MFCC与改进ACF的汽车声音识别算法研究[J]. 计算机技术与发展,2015,25(2):140-143.
[7]张铁成. 基于麦克风阵列的声音识别与定位算法研究[D]. 南京:南京师范大学,2016.
[8]刘建平,张一闻,刘颖. 基于麦克风阵列的汽车笛语识别及笛声定位方法[J]. 西安电子科技大学学报,2012,39(1):163-167.
[9]王琳,李成荣. 一种基于自适应谱熵的端点检测改进方法[J]. 计算机仿真,2010,27(12):373-375,395.
[10]张毅,王可佳,席兵,等. 基于子带能熵比的语音端点检测算法[J]. 计算机科学,2017,44(5):304-307,319.
[11]刘灵,孙晨晨,徐寅林,等. 基于图像预处理分类分割的盲道分割算法[J]. 南京师范大学学报(工程技术版),2020,20(1):42-48.
[12]SETTLES M,SOULE T. Breeding swarms:a GA/PSO hybrid[C]//Proceedings of the 7th Annual Conference on Genetic and Evolutionary Computation Conference,CECCO 2005. Washington DC,USA:ACM,2005:161-168.
[13]ALOMARI F,LIU G H. Novel hybrid soft computing pattern recognition system SVM-GAPSO for classification of eight different hand motions[J]. Optik,2015,126(23):4757-4762.
[14]CLERC M,KENNEDY J. The particle swarm—explosion,stability and convergence in a multidimensional complex space[J]. IEEE Transactions on Evolutionary Computation,2002,6(1):58-73.
[15]宋知用. MATLAB语音信号分析与合成[M]. 2版. 北京:北京航空航天大学出版社,2017.

备注/Memo

备注/Memo:
收稿日期:2020-09-23.
基金项目:国家重点研发计划项目(2017YFB0503500)、江苏省自然科学基金资助项目(BK20171031).
通讯作者:李晨,博士,讲师,研究方向:语音信号处理. E-mail:lichen@njnu.edu.cn
更新日期/Last Update: 2021-06-30