基于惩罚系数的人脸和声音融合识别方法

doi:10.3969/j.issn.1006-2475.2015.11.015

计算机与现代化

基于惩罚系数的人脸和声音融合识别方法

(1.西安科技大学电气与控制工程学院,陕西西安710054；2.陕西师范大学计算机科学学院,陕西西安710062)

收稿日期:2015-06-10 出版日期:2015-11-12 发布日期:2015-11-16
作者简介:温苗利(1978-)，女，河南洛阳人，西安科技大学电气与控制工程学院讲师，博士，研究方向：模式识别，图像处理，生物特征识别。

Face and Speech Recognition Fusion Method Based on Penalty Coefficient

(1. School of Electrical and Control Engineering, Xi’an University of Science and Technology, Xi'an 710054, China；
2. School of Computer Science, Shaanxi Normal University, Xi’an 710062, China)

Received:2015-06-10 Online:2015-11-12 Published:2015-11-16

摘要/Abstract

摘要： 不同的生物特征采集装置采集的样本品质各不相同，样本品质高的识别可靠性较高，同时对于同一个样本，采用不同的识别方法，好的识别方法识别可靠性也较高，因而本文提出一种利用生物特征的样本品质和识别专家可靠性的融合识别（QSVM）方法。首先根据样本品质和识别专家可靠性得到样本惩罚系数和可靠性惩罚系数，进而得到总体惩罚系数，最后利用总体惩罚系数对支持向量机识别算法进行修改。本文采用XM2VTS数据库，分别将QSVM方法与贝叶斯分类法、Fisher线性判别函数分类法、多层感知器分类法和平均融合方法、SVM方法的半和误差率（HTER）进行比较，实验结果表明QSVM方法的半和误差率较小。

关键词: 多生物特征识别, 人脸识别, 声音识别, 惩罚系数, 支持向量机

Abstract: The quality of biometric sample acquired from different acquisition devices is higher, then the reliability of recognition is higher. For the same biometric sample, recognition method is better, then the reliability of recognition is higher. So this paper proposes a multi-biometric recognition algorithm using biometric sample quality and recognition expert reliability (QSVM). First, the method obtains the sample penalty coefficient and reliability penalty coefficients from the sample quality and the expert reliability, then deduces the overall penalty coefficient, finally, uses the overall penalty coefficient to modify SVM fusion recognition algorithm. The experiment uses the XM2VTS database. We compares the Half Total Error Rate (HTER) of QSVM, Bayesian, Fisher linear discriminant, multi-layer perceptron, mean methods and SVM, the experimental results show that the HTER of QSVM fusion algorithm is lower.


Key words: multi-modal biometric recognition, face recognition, speech recognition, penalty coefficient, SVM

中图分类号:

TP391.4

温苗利1,2. 基于惩罚系数的人脸和声音融合识别方法[J]. 计算机与现代化, doi: 10.3969/j.issn.1006-2475.2015.11.015.

WEN Miao-i1,2. Face and Speech Recognition Fusion Method Based on Penalty Coefficient[J]. Computer and Modernization, doi: 10.3969/j.issn.1006-2475.2015.11.015.

参考文献

Lu Yanyun, Boukharouba K, Boonaert J, et al. Application of an incremental SVM algorithm for on-line human recognition from video surveillance using texture and color feature［J］. Neurocomputing, 2014,126:132-140.
［2］Cao Kai, Pang Liaojun, Liang Jimin, et al. Fingerprint classification by a hierarchical classifier［J］. Patter Recognition, 2013,46(12):3186-3197.
［3］Lei Yinjie, Bennamoun M, El-Sallam A A. An efficient 3D face recognition approach based on the fusion of novel local low-level features［J］. Pattern Recognition, 2013,46(1):24-37.
［4］Gutschoven B, Verlinde P. Multi-modal identity verification using support vector machines(SVM)［C］// Proceedings of the 3rd International Conference on Information Fusion. 2000,2:THB3/3-THB3/8.
［5］刘铁根，李秀艳，王云新，等. 基于SVM的多生物特征融合识别算法［J］. 纳米技术与精密工程, 2011,9(1):44-47.
［6］袁臣虎，刘铁根，李秀艳. 基于kNN-SVM的手背静脉虹膜和指纹融合身份识别［J］. 光电工程, 2013,40(4):101-105.
［7］FierrezAguilar J, Ortega-Garcia J, Gonzalez-Rodriguez J, et al. Discriminative multimodal biometric authentication based on quality measures［J］. Pattern Recognition, 2005,38(5):777-779.
［8］Fierrez-Aguilar J. Ortega-Garcia J. Gonzalez-Rodriguez J. et al. Kernel-based multimodal biometric verification using quality signals［C］// Biometric Technologies for Human Identification, Proceedings of SPIE. 2004,5404:544-554.
［9］Kittler J, Poh N, Farukasi O, et al. Quality dependent fusion of intramodal and multimodal biometric experts［C］// Proceedings of SPIE Biometric Technology for Human Identification. 2007,6539:653903-1-653903-14.
［10］Fatukasi O, Kittler J , Poh N. Quality controlled multimodal fusion of biometric experts［C］// The 12th Iberoamerican Congress on Pattern Recognition CIARP. 2007:881-890.
［11］Poh N, Bourlai T, Kittler J. A multimodal biometric test bed for quality-dependent, cost-sensitive and client-specific score-level fusion algorithms［J］. Pattern Recognition， 2009,43(3):1094-1105.
［12］Poh N, Kittler J. A unified framework for biometric expert fusion incorporating quality measures［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2012,34(1):3-18.
［13］王科俊,马慧,管凤旭,等. 基于图像采集质量评价的指纹与指静脉双模态识别决策级融合方法［J］. 模式识别与人工智能, 2012,25(4):669-675.
［14］肖宾杰. 基于图像质量加权的D-S证据理论多生物特征融合识别［J］. 计算机应用, 2012,32(1):264-268.
［15］Mohammad Rafiqul Alam, Mohammed Bennamoun, Roberto Togneri, et al. A cofience-based late fusion framework for audio-visual［J］. Pattern Recognition Letters, 2015,52:65-71.
［16］Messer K, Matas J, Kittler J, et al. XM2VTSDB: The extended M2VTS database ［C］// Proceedings of the 2rd International Conference on Audio and Video-based Person Authentication(AVBPA’99). 1999:72-77.
［17］Poh N, Bengio S. Database, protocols and tools for evaluating score-level fusion algorithms in biometric authentication［J］. Pattern Recognition, 2006,39(2):223-233.

[1]	杨博, 庄毅. 基于AOA-MSVM的控制集群故障检测方法[J]. 计算机与现代化, 2023, 0(12): 112-116.
[2]	王鑫, 肖韬睿. 基于生成对抗网络的人脸识别对抗攻击[J]. 计算机与现代化, 2023, 0(10): 115-120.
[3]	杨孙哲, 孙爱珍. 基于HSV颜色与LBP纹理特征的水稻氮素营养诊断[J]. 计算机与现代化, 2023, 0(07): 86-92.
[4]	申志, 李元 . 基于KPCA和SSA优化SVM的非线性过程故障检测#br#[J]. 计算机与现代化, 2023, 0(06): 15-20.
[5]	尹建丰, 卫鑫, 顾雄伟, 黄凯, 魏敏捷. 基于图像阈值优化及改进SVM的电表数字识别[J]. 计算机与现代化, 2023, 0(05): 106-110.
[6]	盛江岸, 陈淑荣. 融合双重注意力机制的戴口罩人脸识别方法[J]. 计算机与现代化, 2023, 0(02): 72-77.
[7]	赵延平, 王芳, 夏杨. 基于支持向量机的短文本分类方法[J]. 计算机与现代化, 2022, 0(02): 92-96.
[8]	赵聪慧, 冯庆胜. 基于DT-CWT和SVM的踏面旋转不变纹理提取算法[J]. 计算机与现代化, 2021, 0(12): 85-90.
[9]	陈翔, 邹庆年, 谢绍宇, 陈翠琼. 基于迁移学习策略的压板开关状态识别[J]. 计算机与现代化, 2021, 0(05): 120-126.
[10]	徐睿, 李志华, 韩灿灿. 基于特征融合和改进卷积神经网络的环境音识别[J]. 计算机与现代化, 2021, 0(02): 62-67.
[11]	王文蔚, 肖军弼, 程鹏, 张悦. 基于SDN的DDoS攻击防御系统[J]. 计算机与现代化, 2021, 0(02): 117-121.
[12]	张文华，张志俊. 基于SVM的新能源公交车运营里程核查方法[J]. 计算机与现代化, 2020, 0(05): 39-.
[13]	陈凯,祖莉,欧屹. 基于YOLOv3与ResNet50的摄影机器人人脸识别跟踪系统[J]. 计算机与现代化, 2020, 0(04): 30-.
[14]	郑沫利1,赵艳轲1,闫敏2,孙俊2,刘雍容1. 基于RDPSO-SVM的粮食产后储藏环节损耗智能评估方法[J]. 计算机与现代化, 2020, 0(03): 72-.
[15]	李硕1，卞青山2，刘传文2，刘鸣涛1，张林涛1 . 基于DCNN的证件照人脸验证及应用研究[J]. 计算机与现代化, 2020, 0(02): 104-.

基于惩罚系数的人脸和声音融合识别方法

Face and Speech Recognition Fusion Method Based on Penalty Coefficient

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