基于Gabor小波＋PCA＋LDA特征提取方法的人脸表情识别
基于Gabor小波＋PCA＋LDA特征提取方法的人脸表情识别20机电技术2011年12月=∑cm=∑mc×(-0509-n)U=(a-n(2×3.1415926)×vm第四步:初始模型确定下来以后,采用则将该待识别人脸表情判别为第讠类表情Baum- Welch算法对参数进行重估计,迭代调整模3人脸表情实验及分析型参数以达最优化。这样就完成了训练建模工作,获得了在粗分本文在JAFF( (Japanese Female Facial Expre类层次下代表7种人脸表情的HMM模型sso)表情数据库中进行实验。 JAFFE人脸表情库1=(A,B,n)(=1,2,…7)。由10个人,共213幅图像组成,每人都包含愤怒、23HMM人脸表情分类厌恶、恐惧、高兴、悲伤、惊讶和中性7种表情,每人每种表情2~4幅,每种表情共有27幅。对待识别的人脸表情图像,按本节22中模型本文实验与 Micael等的测试方法相似,把数训练步骤的第一步获得观察向量O,然后计算观据库中的210张图像分成3个部分,每个部分包含察向量O与已训练好的HMM模型1=(A,B,)10个人,且每个人包含7幅不同表情图像,每次使的似然概率P(O|λ)(=12,…,7)。通过用其中的一个部分训练各个表情的HMM模型,剩Forward -Backward算法来计算似然概率,如果下的两个部分用来测试。循环操作3次,然后对这P(O|4),ie(12…,7在P(O|4)(=1,2…,7)3次取平均值,得出表情识别率。本方法获取的表中最大,即情识别率见表1P(Oa )=max p(oa)表1人脸表情识别结果愤怒厌恶恐惧高兴中性悲伤惊讶识别率/%愤怒91.67厌恶5400009000恐惧559167高兴00095.00中性549000悲伤9333惊讶5693.33整体识别率92.14从表1中可以看出,本方法可以获得较好的表高的问题,提出了一种新的人脸表情识别方法。情识别率,但是,从实验中也可发现,不管是有本方法通过更加有效的 Gabor小波+PCA+LDA表没有减少分类类别数,实验中始终存在误识别。情特征提取方法,然后结合HMM模型建立7个对从实验数据库中发现,库中某些人的7种表情变化应基本表情+中性表情的模型,进行识别分类,并并不明显,即使人眼也无法确定其表情类别,以且取得了92.14%的整体识别率。同时,在实验中致算法在识别时出现错误。发现,由于人脸实验数据库中某些人的7种表情变4结论化并不明显,即使人眼也无法确定其表情类别,以致算法在识别时出现错误。本文针对目前人脸表情的整体识别率普遍不(下转第35页)第6期林彩丽等:磁流变阻尼器减振系统的分数阶微分方程研究[2] DROZDOVAD Mechanicsofviscoelasticsolids[M]. New York: John Wiley &Sons Ltd, 1998: 21-65[]LAKESR S. Viscoelastic solids[M]. London: CRC Press, 1998: 63-110[4] Stanway R, Sproston, J L, Stevens N G Non-liner modeling of an electrorheological vibration damper. J. Electrostatics,1987,20.[5]周强,瞿伟廉磁流变阻尼器的两种力学模型和试验验证地震工程与工程振动2002240).[6] Gamoto D R, Filisko F E Dynamic mechanical studies of electrorheological materials: moderate frequencies. I. Rheology,1991,35(3)[7]汪建晓孟光.磁流变阻尼器用于振动控制的理论及实验研究门]振动与冲击,2001,20(2)[8] Wen Y K Method of random vibration of hysteretic systems[]. Joumal of Engineering Mechanics Division, ASCE,102(EM2),1976[9]Spencer Jr. B F, Dyke S J, Sain M K, Carlson, J D Phenomenological model of a magnetorheological damper. J Engrg. MechASCE1997,123[10薛定宇陈阳泉高等应用数学问题的 MATLAB解DM清华大学出版社,2010[l]周云谭平.磁流变阻尼控制理论与技术DM科学出版社200712]王振滨曹广义分数微积分的两种系统建模方法系统仿真学报,2004,64):810812(上接第20页)参考文献[1] ZHAO Quan-you, PAN Bao-chang, PAN Jian- jia, et al. Facial expression recognition based on fusion of Gabor and LBPfeatures[A]. In: Proceedings of the 2008 International Conference on Wavelet Analysis and Pattern Recognition[C], HongKong,IEEE,2008(8):362-367[2]Tang Fang-qi, Deng Ben-zai. Facial Expression Recognition using AAM and Local Facial Features [A]. In: Proceedings ofthe Third Intcmational Conference on Natural Computation[C], Haikou, China, IEEE, 2007(8): 632-635[3]陈培俊基于静态图像的人脸表情识别研究成都西南交通大学,2007[4] Lades M, Vorbruggen J C, Buhmann J. Distortion invariant object recognition in the dynamic linkarchitecture[].IEEE Transactions on Computers, IEEE, 1993, 42(3): 300-311[5]彭辉张长水荣钢等基于KL变换的人脸自动识别方法清华大学学报自然科学版,2007,37(6:6770[6]DudaR O, Hart P E, Stork DG Pattern Classification(second edition). New York: Wiley, 2001[7李云霞李治柱吴亚栋基于HMM的关键词识别系统门计算机工程200,30(7:130-132谨致作者敬请本刊作者允诺:稿件中没有侵犯他人著作权或其它权利的内容并且文责自负;投寄给本刊的稿件(论文、图表、照片等)自发表之日起,其专有出版权和网络传播权即授于本刊,并许可本刊在本刊网站或本刊授权的网站上传播及中国核心期刊(遵选)数据库、中国学术期刊综合评价数据库、中文科技期刊数据库、中国台湾华艺数据库全文收录。对上述合作若有异议者,烦请来稿时向本刊申明,未作中明者,本刊将视为同意,谢谢合作。并致诚挚敬意。《机电技术》编辑部基于 Gabor小波十PCA+LDA征提取方法的人脸表情识别旧数据文接作者:李烈熊,苏两河作者单位李烈熊(福硅船政交通职№学院,福廷福州,350007),苏河(福建电力职技术学院,福建泉州刊名:机电技术英文刊名:MECHANICAL AND ELECTRICAL TECHNOLOGY年,卷(期)201,34(6)参考文献(7条)1. ZHAO Quan-you; PAN Bao-chang PAN Jian-jia Facial expression recognition based on fusion of Gabor and LBPfeatures 20082. Tang Fang-qi; Deng Ben-zai Facial Expression Recognition using AAM and Local Facial Features 20073.陈培俊基丁静态图像的人脸表情识别研究20074. Lades M: Vorbmggen J C; Buhmann J Distortion invariant object recognition in the dynamic link architecture1993(03)5.彭辉;张长水;荣钢基于K-变换的人脸自动识别方法2007(6)6. Duda R 0: Hart P E Stork D G Pattern Classificat ion(second edition) 20017.李云霞;李治柱:吴亚栋基于HM的关键词识别系统[期刊论文]计算机工程2004(07本文链接http://d.g.wanfangdata.comcn/periodiCaljdjs201106005.aspx
- 2020-12-01下载
- 积分：1
ADMM优化算法讲解
alternating direction method of multipliers优化算法讲解OutlineDual decompositionMethod of multipliersAlternating direction method of multipliersCommon patternsExamplConsensus and exchangeConclusionsDual decompositionDual problemp convex equality constrained optimization problemminimizesubject to Ax= 6e Lagrangian: L(a, g)=f(a)+y(Ac-bdual function: g(y)=infx L(, g)e dual problem: maximize g(g)recover x*=argminL(, y*)Dual decompositionDual ascentgradient method for dual problem: y+l=yk +aVg(yky ")=A c-b, where a= argmin L(a, y")b dual ascent method isk+1gminz L(a, yk/-minimization(Axk+I-b)// dual updateworks, with lots of strong assumptionsDual decompositionDual decompositione suppose f is separablef(x)=f1(x1)+…+fN(xN),x=(x1Nthen L is separable in x: L(a, y)=L1(a1, 3)+...+Ln(N, 3)-y bLi(ai, y)=fi(ai)+y Aiaie -minimization in dual ascent splits into N separate minimizationsk+1argmin Li(li, y)Which can be carried out in parallelDual decompositionDual decompositiondual decomposition(Everett, Dantzig, Wolfe, Benders 1960-65k+1argLi(ei, y)N A: k+scatterupdate i in parallel, gather Ai k+solve a large problemby iteratively solving subproblems(in parallel)dual variable update provides coordinationworks, with lots of assumptions; often slowDual decompositionOutlineDual decompositionMethod of multipliersAlternating direction method of multipliersCommon patternsExamplConsensus and exchangeConclusionsMethod of multipliersMethod of multipliersa method to robustify dual ascentb use augmented Lagrangian(Hestenes, Powell 1969),p>0(, y)=f(c)+y(Ax-b)+(p/2)Acmethod of multipliers( Hestenes, Powell; analysis in Bertsekas 1982)k+1argmin Lp(a, yD(A.(note specific dual update step length pMethod of multipliersMethod of multipliers dual update stepoptimality conditions( for differentiableAcx-b=0, Vf(a*)+A(primal and dual feasibility)Since ah+1minimizes Lp(a, y)k+1 kf(x4+1)+A7(y+p(AVxf(at)+adual update yti=y+p(k+1k+1dual feasibleprimal feasibility achieved in limit: A k+I-b>0Method of multipliers
- 2021-05-06下载
- 积分：1
