摘要
服装缝纫平整度是影响服装外观质量的重要因素,成衣加工过程中的缝纫平整度控制及评价是一个困扰了服装行业多年的问题。目前在大多数商业行为中仍然以主观评定法AATCC-88B(American Association of Textile Chemists and Colorists-Method 88B)作为服装缝纫平整性能的主要评价标准,在参考样品的基础上目视评级,即1、2级是不可接受的,3级是临界级,4、5级是可接受的。虽然这种方法具有简单,易操作等特点,但由于受主观个人因素的影响较大。因此,客观评价服装缝纫平整度的研究一直在持续,而设计有效的客观评价算法是当前服装行业急需解决的热点和难点问题。本文针对这些问题,在对面料的FAST力学性能数据进行相关分析和核主成分分析的基础上,提出基于监督模糊聚类的径向基神经网络(SFCM-RBFNN)来客观评价服装缝纫性能,取得了较好的预测效果。
首先用面料性能测试仪FAST仪器对面料的各项力性能指标进行了测试,然后对面料各项力学性能指标与缝纫性能进行相关性分析,分别分析了面料结构力学性能、面料拉伸伸长性能、面料弯曲性能、面料成形性能和面料尺寸稳定性能与面料经、纬向缝纫平整度等级之间的关系,分别作图考察面料缝纫性能随FAST各项力学指标的变化分布情况,并剔除部分与缝纫平整度相关性低的指标。
同时从面料斜向力学性能理论和实验研究出发,研究了面料各向异性与缝纫外观平整性能的关系,分析了面料的拉伸性能、弯曲性能及成型性随面料斜裁角度变化的改变情况,及它们对缝纫外观平整性能产生影响。
其次针对面料的高维非线性FAST力学性能数据,运用核主成分分析法(KPCA)将这些FAST指标数据(输入空间)影射到一个新的线性特征空间去,然后在这个空间用线性分类方法进行有效的分析。用高斯核函数对面料FAST力学性能数据做了核主成分分析试验,并提取了相应的主成分,达到了非线性数据降维的目的。
然后在模糊C均值(FCM)算法基础上进一步提高模糊聚类的精度,提出了SFCM算法,即有监督的模糊C均值算法,将输入空间样本x(面料FAST力学性能数据主成分)和输出空间的期望值y(缝纫平整度)同时引入到模糊聚类中来,并通过相应的局部线性回归模型来表征相应的模糊聚类的每一个类别,用这些局部模型的线性组合求和来计算输出y的实际值,从而达到该算法的实现。在前面KPCA分析的基础上,对面料FAST力学性能数据进行SFCM模糊聚类,聚类效果较好。该算法除了反映输入空间的聚类特性外,还反映了输出空间的逼近特性。
最后对在径向基神经网络(RBFNN)基础上对它的隐层节点基函数进行了调整,将SFCM聚类算法中的模糊分割矩阵U和聚类中心v引入基函数,从而使RBF神经网络隐层节点的宽度(感受野)和中心得到了有效的优化和控制,提高了系统的泛化能力。实验证明,本文提出的基于监督模糊聚类的径向基神经网络(SFCM-RBFNN)算法对不同结构和材料面料的缝纫平整度预测具有较好的精度和鲁棒性。
Fabric sewing ability is one of important factors that influence the quality of clothing appearance. Controlling and evaluation of seam pucker is a lasting problem of fashion industry during the course of clothing manufacture.
AATCC (American Association of Textile Chemists and Colorists) Method 88B has been commonly used for the subjective evaluation of seam pucker. According to this method, the appearance of seams are compared with photographic standards and the severity of seam pucker is graded into five classes, Class 5 being little or no pucker, and Class 1 severe pucker. Normally, Class 5 and 4 are acceptable, Class 3 is critical or borderline, and Class 2 and 1 are unacceptable. The merit of this method is directness, simplicity, low investment and easy to master, but it is influenced by uncertain factors from the evaluator and loses its objectivity easily. So, the research for objective evaluation of seam pucker is on going, and designing an effective evaluation algorithm continues to be hotspot and difficult problem of the fashion design and manufacture industry.
In this paper, a supervised fuzzy clustering RBF neural network (SFCM-RBFNN) based on kernel principal component analysis is introduced for constructing the garment seam evaluation system. Experimental results demonstrate that the proposed system could efficiently evaluate the fabric sewing ability.
Firstly, this paper use FAST(Fabric Assurance by Simple Testing) apparatuses to test the mechanic indexes of fabric, and analyze the correlation between the indexes and fabric sewing ability of warp action and weft action, such as fabric structure mechanical properties, fabric extension mechanical properties, fabric formality ability, dimensional stability. Then figures are plotted to observe the trend of seam pucker grade with these FAST mechanical properties data. According to the correlation analysis, a few low related indexes with seam pucker are eliminated.
With the fabric inclined mechanical properties theory, research was done on the relations between fabric anisotropy and sewing ability, such as extension properties, bendability properties, formality ability. This paper also plots figures to observe the trend of seam pucker grade as the fabric diagonal cutting angle.
Secondly, according to these high dimension and nonlinear FAST mechanical properties data (in input space), The kernel principal component analysis (KPCA) method was applied to project these data in a whole new linear feature space, then linear clustering method was used to analyze it effectively in the new space. This paper extracts the principal component of fabric FAST mechanical properties by KPCA method with gaussion function and reduces the dimension of the data.
Thirdly, to improve the accuracy of fuzzy clustering based on fuzzy C mean (FCM) algorithm, this paper proposes a novel supervised fuzzy C mean clustering algorithm including input space sample x (principal component of fabric FAST mechanical properties) and output space expectation y (seam pucker grade) at the same time. Using local linear regression model to represent each fuzzy cluster, the actual output y can be calculated by sum of these local linear model. And the clustering effect of the fabric FAST mechanical properties PCA data by SFCM algorithm is quite good. The SFCM algorithm proposed by this paper not only shows the clustering action in input space, but also shows the approximating action in output space.
This paper modifies the radius base function of hidden nodes of RBF neural network by importing the fuzzy partition matrix U and clustering center v of SFCM algorithm to construct a SFCM-RBF neural network. The modification makes the width (receipt field) and center of hidden nodes of RBF neural network become more effectively optimizing and controlling. Experimental results demonstrate that the proposed system could efficiently be used as an objective garment seam evaluation system with high accuracy and is robust for various structures fabric.
引文
[1] R. W. Rennell. FABER research project[J]. China Textile & Apparel, 1996, 2(3): 30-32
[2] J. Lindberg, L. W. Aesterberg, R. Svenson. Wool Fabric as Garment Constrction Material. J. Text. Inst, 1960, 51:1475-1492
[3] M. Niwa, S. Wawabata, K. Ishizuka. Recent Developments in Research Corrlating Basic Fabric Mechanical Properties and the Appearance of Men's Suits[C]. Proc. Japan-Austrilia Symposium on Objective Evaluation of Appear Fabrics, Parkville, Australia, 1983.67
[4] R. L. Shishoo, M. choroszy. Analysis of Mechanical and Dimensional Properties of Wool Fabrics Relevant to Garment Making[C]. Proc. Japan-Australia Symposium on Objective Evaluation of Appear Fabrics, Textile Machinery Society of Japan, Osaka, 1990.316
[5] J. Amirbayat. The Buckling of Flexible Sheets under Tension[J]. Part I : Theoretical Analysis, Part II: Experiment Studies, J. Text. Inst., 1991, 82(2):61-77
[6] R. C. Dhingra, R. Postle. Fabric mechanical and physical properties relevant to clothing manufacture[J]. Clothing Res. J., 1980, 8:59
[7] T. J. Mah?r, R. C. Dhingra, R. Postle. Fabric Mechanical and Physical Properties Relevant to Clothing Manufacture[J]. Part I : Fabric Overfeed, Formability, Shear and Hygral Expansion during Tailoring, International Journal of Clothing Science and Technology, 1989, 1(1):2-20
[8] J. Aemirbayat. Seams of different ply properties[J]. Part I :Seam Appearance, Part II: Seam Strength, J. Text, Inst., 1992, 82(2) :211
[9] M. Y. Kwong. Engineering principles in man's jacket pattern construction[C]. Proc. of China International Wool Textile Conference, Xi'an, 1994.214
[10] R. L. Shishoo, M. Choroszy. Analysis of mechanical and dimensional properties of wool fabrics relevant to garment making[C]. Proc. Japan-Australia Symposium on Objective Evaluation of Appear Fabrics, Parkville, Australia, Textile Machinery Society of Japan, Osaka, 1990.316
[11] P. G. Cookson, A.M. Wemyss. The Effects of Hygral Expansion on the Appearance of Structured Wool Garments[C]. Proc. Japan-Australia Symposium on Objective Evaluation of Appear Fabrics, Textile Machinery Society of Japan, Osaka, 1990.380-389
[12] P. G. Cookson. Relationships between Hygral Expansion, Relaxation Shrinkage, and Extensibility in Woven Wool Fabrics[J]. Text. Res. J., 1992, 62(1):44-51
[13] A. M. Manich, J. P. Domingues, A. Barella. Relationships between fabric sewabi 1 ity and structural, physical, and FAST properties of woven wool and wool-blend fabrics[J]. J. Text. Inst., 1998, 89(3):579-591
[14] G. Stylios, J. Sotomio. A neural network approach for the optimization of the sewing process of wool and wool mixture fabrics[C]. Proc. of 1"' China international Wool Textile Conference, Xi'an, 1994.689-693
[15] C. K. Park, T.J. Rang. Objective rat ing of seam pucker using neural networks[J]. Text.Res.J.,1997,67(7):494-502
[16]T.Aibara,T.Mabuchi,M.Izumida.Automatic Evaluation of the Appearance of Seam Puckers on Suits[J].IEICE IRANS INF & SYST,2000,(E83-D)7:1346-1352
[17]T.J.Kang,J.Y.Lee.Objective evaluation of fabric wrinkles and seam puckers using fractal geometry[J].Text.Res.J.,2000,70(6):469-475
[18]J.Fan,D.Lu,J.M.K.Macalpine.Objective Evaluation of Pucker in Three Dimensional Garment Seams[J],Text.Res.J.,1999,69(17):467-472
[19]J.Fan,F.Liu.Objective evaluation of garment seams using 3Dlaser scanning Technology[J].Text.Res.J.,2000,70(11):1025-1030
[20]朱松文,李引枝.衣着评价的研究Ⅲ-缝纫线迹的评价[J].西北纺织工学院学报,1990,2:101-106
[21]张静.线迹外观质量的模糊综合评判[J].四川纺织科技,2004,4:8-11
[22]刘侃.基于面料力学性能的服装缝纫平整度等级客观评价系统的建立[D]:[博士学位论文].上海:东华大学,2005
[23]P.G.Minazio.FAST-fabric Assurance by Simple Testing[J].International Journal of Clothing Science and Technology,1995,7(2):43-48
[24]S.S.Lai,T.W.Shyr,J.Y.Lin.Comparison between KES-FB and FAST in discrimination of fabric characteristics[J].Text.Res.J.,2002,48(2):43-49
[25]慎仁安.新型纺织测试仪器使用手册[M].北京:中国纺织出版社,2005
[26]肖学霞.基于FAST仪力学性能测试的织物手感客观评价研究[D]:[硕士学位论文].苏州:苏州大学,2005
[27]G.H.Wang,W.Y.Zhang.Comparison of Low stress Mechanical Properties of Light weight Wool and Wool Blend fabrics using the KES-F and FAST Instruments[J].Journal of China Textile University,2000,17(4):99-102
[28]K.L.Yick,K.P.S.cheng.Comparison of the Fabric Mechanical Properties Measured by the KES-F and FAST Systems[J].Textile Research Journal,1996,66(10):622-633
[29]杨建忠.轻薄织物低应力力学性能及可缝性研究[D]:[博士学位论文].上海:中国纺织大学,1999
[30]AATCC Test Method 88B-2001.Smoothness of Seam in Fabrics after Repeated Home Laundering.
[31]B.K.Behera,S.Sharma.Low stress behaviour and sewability of suiting and shirting fabrics[J].Indian Journal of Fibre & Textile Research.1998,23:233-241
[32]陈之戈,陈雁.服装面料的缝制加工性能[J].丝绸,2001,3:28-31
[33]彭华陵,朱松文.服装面料的拉伸性能与服装加放量关系的研究[J].纺织科技进展,2005.4:61-62
[34]周爱英,张明杰,张怀珠.真丝织物力学性能与服装外观造型之间的关系[J].中国纺织大学学报,1999,25(1):100-107
[35]倪红.面料的性能与服装造型及生产的关系研究[D]:[硕士学位论文].苏州:苏州大学,2003
[36]Y.P.Wang.The application of FAST system in wool garment modeling and manufacturing[C].Proc.of the 3~(rd) China International Wool Textile Conference,1986.342-345
[37]汪学骞,毕松梅,宋远丁.服装面料耐皱性研究[J].安徽机电学院学报,2001,16(1):1-5
[38]罗祖道,李思简.各向异性材料力学[M].上海:上海交通大学出版社,1994
[39]W.F.Kllay.Planar stresss-strain relationships in woven fabrics[J].J.Text,Inst.,1963,54:9-27
[40]N.G.Ly,D.H.Tester.Simple Instruments for Quality control by finishers and tailors[J].Text.Res.J.,1991,61(8):402
[41]F.T.Pierce.The handle of cloth asameasurable quantity[J].J.Text.Inst.,1930,21:377-417
[42]F.T.Pierce.The geometry of cloth structure[J].J.Text.Inst.,1937,28:45-96
[43]A.Shinohara,F.Shinohara,K.Sakaebara.Theoretical study on anisotropy of bending rigidity of woven fabric[J].Journal of Textile Machinery Society of Japan,1980,28:75-79
[44]Y.Go,A.Shinohara.On the relationship between the bending behaviors of fabrics and fabric strictures[J].Viscoelastic studies of textile fabrics,part5,1958,14:300-303
[45]Y.Go,A.Shinohara.On the relationship between the bending behaviors of fabrics and fabric strictures[J].Viscoelastic studies of textile fabrics,part5,1958,14:855-864
[46]D.N.E.Cooper.The effect of twist upon bending properties[J].J.Text.Inst.,1960,51:150-151
[47]D.N.E.Cooper.The stiffness of woven textile[J].J.Text.Inst.,1960,51:317-335
[48]杨建忠,王善元.轻薄府绸织物斜向强伸性能与接缝强力的关系[J].西安科技工程学院学报,2002,16(2):95-99
[49]杨建忠,王善元.轻薄织物斜向力学性能与缝纫起皱关系的研究[J].西北纺织工学院学报,2001,15(3):15-20
[50]杨建忠,王善元.轻薄织物屈曲变形与缝纫起皱关系的研究[J].西北纺织工学院学报,2001,15(1):5-10
[51]B.Yang,Z.Y.Zhu.A nonlinear PCA algorithm based on RBF neural networks[J].Journal of Harbin Institute Technology,2005,(12)1:101-104
[52]A.Chakrabarti,A.N.Rajagopalan,R.Chellappa.Super-Resolution of Face Images Using Kernel PeA-Based Prior[J].IEEE TRANSACTIONS ON MULTIMEDIA,2007,9(4):889-892
[53]S.W.Choi,I.B.Lee.Nonlinear dynamic process monitoring based on dynamic kernel PCA[J].Chemical Engineering Science,2004,59:5897-5908
[54]A.LIMA,H.Zen.On the use of kernel PCA for feature extraction in speech recognition[J].IEICE TRANS.INE.&SYST.,2004,87(12):2802-2811
[55]B.Thirion,Olivier,Faugeras.Dynamical components analysis of fMRI data through kernel PCA[J].Neurolmage,2003,20:34-49
[56]J.Tan.Applications of kernel PeA methods to geophysical data[D]:[博士学位论文].Virginia:George Mason University,2005
[57]M.T.Xu,P.Franti.A Heuristic K-Means Clustering Algorithm by Kernel PCA[C].International Conference on Image Processing,2004.3503-3506
[58]C.D.Lu,T.Y.Zhang.Adaptive Robust Kernel PCA Algorithm[C].IEEE ICASSP,2003.621-624
[59]张润楚.多元统计分析[M].北京:科学出版社,2006
[60]I.T.Jolliffe.Principal Component Analysis[M].2nd Edition,Springer,2002
[61]W.J.Krzanowski.Principles of Multivariate Analysis[M].Oxford University Press,1988
[62]G.A.F.Seber.Multivariate Observations[M].Wiley,1984
[63]J.E.Jackson,A User's Guide to Principal Components[M].John Wiley and Sons,1991
[64]M.Ye,Z.Yi,K.K.Tan.A Robust and Globelly Convergent PCA Learning Algorithm[J].Control and Intelligent System,2007,35(2):119-125
[65]V.D.M.Nhat,S.Y.Lee.An Improvement on PCA Algorithm for Face Recognition[C].ISNN,2005.1016-1021
[66]J.S.Taylor,N.Cristianini.模式分析的核方法[M].北京:机械工业出版社,2006
[67]J.C.Lv,Y.Zhang.Global Convergence of a PCA Learning Algorithm with a Constant Learning Rate[J].Computers and Mathematics with Applications,2006,52:1425-1438
[68]李国正.支持向量机导论[M].北京:电子工业出版社,2004
[69]孙宗海.支持向量机及其在控制中的应用研究[D]:[硕士学位论文].浙江:浙江大学,2003
[70]邓乃扬,田英杰.数据挖掘中的新方法-支持向量机[M].北京:科学出版社,2004
[71]S.X.Cheng,F.Y.Shih.An improved incremental training algorithm for support vector machines using active query[J].Pattern Recognition,2007,40:964-971
[72]H.H.Gao,H.H.Yang,X.W.Wang.Kernel PCA Based on Network Intrusion Feature Extraction and Detection using SVM[C].ICNC,2005.89-94
[73]王辉.基于核主成分分析特征提取及支持向量机的人脸识别应用研究[D]:[硕士学位论文].合肥:合肥工业大学,2006
[74]A.M.Jade,B.Srikantb,V.K.Jayaraman.Feature extraction and denoising using kernel PCA[J].Chemical Engineering Science,2003,58:4441-4448
[75]S.Dambreville,Y.Rathi,A.Tannenbaum.Shape-Based Approach to Robust Image Segmentation using Kernel PCA[C].Proc.of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition,2006.521-526
[76]D.X.Yu,W.Lei,T.Qi.A novel multi-resolution vedio representation scheme based on kernel PCA[J].Visual Compute,2006,22:357-370
[77]王士同.模糊系统模糊神经网络及应用程序设计[M].上海:上海科技文献出版社,2000.
[78]杨润玲.基于FCM类型算法的自动图像分割方法研究[D]:[硕士学位论文].两安:西安电子科技大学,2006
[79]闫兆振,自适应模糊C均值聚类算法研究[D]:[硕士学位论文].青岛:山东科技大学,2006
[80]张智星,孙春在,水谷英二.神经-模糊和软计算[M].西安:西安交通大学出版社,2006
[81]李云松.改进的模糊C均值聚类对噪声图像的分割[D]:[硕士学位论文].兰州:兰州理工大学,2007
[82]宫改云.FCM算法参数研究及其应用[D]:[硕士学位论文].西安:西安电子科技大学,2004
[83]W.Pedrycz.Conditional fuzzy clustering in the design of radial badis function neural networks[J].IEEE Trans,Neural Networks,1998,(9)4:601-612
[84]N.B.Karayiannis.Weighted fuzzy learning vector quantization and weighted generalized fuzzy c-means algorithms[C].Proc.of the IEEE Conference Fuzzy Systems,New Orleans,1996.773-779
[85]W.Pedrycz.Algorithms of fuzzy clustering with partial supervision[J].Pattern Recognition Letters,1985,3:13-20
[86]W.Pedrycz.Conditional fuzzy c-means[J].Pattern Recognition Letters,1996,17:625-632
[87]W.Pedrycz,G.Vukovich.Fuzzy Clustering with supervision[J].Pattern Recognition,2004,(37)7:1339-1349
[88]R.Zagliafen-i,A.StaianoA,D.Scala.A Supervised Fuzzy Clustering for Radial Basis Function Neural Networks Training[C].IEEE,2001.1804-1809
[89]M.Setnes.Supervised Fuzzy Clustering for Rule Extration[J].IEEE Transactions on Fuzzy Systems,2000,8(4):416-424
[90]J.Paetz.Supervised Neuro-fuzzy Clustering for Life Science Applications[C].ISBMDA,2006.p378-389
[91]刘宇红,刘桥,任强.基于模糊聚类神经网络的语音识别算法[J].计算机学报,2006,29(10):1894-1900
[92]J.Abonyi,F.Szeifert.Supervised fuzzy clustering for the identification of fuzzy classifiers[J].Pattern Recognition Letters,2003,24:2195-2207
[93]J.Abonyi,F.Szeifert.Supervised fuzzy clustering for the identification of fuzzy classifiers[J].Pattern Recognition Letters,2003,24:2195-2207
[94]J.C.Bezdek.Pattern Recognition with Fuzzy Objective Function Algorithms[M].Plenum Press,New York,USA,1981
[95]J.Moody,C.Darken.Fast Learning in Networks of Locally-Tuned Processing Units[J].Neural Computation,1989,1:281-289
[96]Simon Haykin.神经网络原理[M].北京:机械工业出版社,2004
[97]王华秋.一种改进型径向基神经网络的研究及应用[D]:[硕士学位论文].重庆:重庆大学,2003
[98]罗四维.大规模人工神经网络理论基础[M].北京:清华大学出版社,2004
[99]K.B.Kim,J.H.Cho.Recognition of Passports Using FCM-Based RBF Network[C].Australian Joint Conference on Artificial Intelligence,2005.1241-1245.
[100]李淼.基于径向基神经网络分类器的人脸识别技术[D]:[硕士学位论文].,哈尔滨:哈尔滨工业大学,2006
[101]H.Sarimveis,A.Alexandridis,G.Bafas.A fast training algorithm for RBF networks based on subtractive clustering[J].Neurocomputing,2003,51:501-505
[102]W.Li,Y.Hori.An Algorithm for Extracting Fuzzy Rules Based on RBF Neural Network[J].IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS,2006,53(4):1269-1276
[103]K.Maeda,S.Kanae,Z.J.Yang.Design of RBF Network Based on Fuzzy Clustering Method for Modeling of Respiratory System[C].ISNN,2006.746-753
[104]D.Lowe.Adaptive radial basis function non-linearities and the problem of generalization[C].Proc.1st IEEE Conf.Artificial Neural Networks,London,UK,1989.171-175
[105]孙健,申瑞民,韩鹏.一种新颖的径向基函数(RBF)网络学习算法[J].计算机学报,2003,26(11):1562-1567.
[106]朱明星,徐建华.RBF网络基函数中心选取算法的研究[J].安徽大学学报(自然科学版), 2000, (24)1:73-78
[107]J.X. Peng, K. Li. A Hybrid Forward Algorithm for RBF Neural Network Construction[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 2006, (17)6:1439-1451
[108]M. Lazaro, I. Santamaria, W. Pantaleon. A new EM-based training algorithm for RBF networks [J]. Neural Networks, 2003, 16:69-77
[109] J. X. Peng, K. Li, G. W. Irwin. A Novel Continous Forware Algorithm for RBF Neural Modelling[J]. IEEE Transactions on Automatic Control, 2007, (52)1:117-122
[110]B. Ye, C.Z. Zhu. Adaptive extended fuzzy basis function network[J]. Neural Computing & Applications, 2007, 16:197-206
[111]M. T. Musavi, W. Ahmed, K. H. Chan. On the training of radial basis function classifiers[J]. Neural Networks, 1992, 5:595-603
[112]C. M. Bishop. Neural Networks for Pattern Recognition[M]. Oxford University Press, New York, USA, 1995