服装流行色的量化与预测研究
|
摘要
服装流行色预测是指应用特定的方法判断并给出24个月后服装的色彩流行趋势定案。定案能为企业提供未来服装销售季节中色彩流行的相关信息，有益于服装企业合理制定生产方案和营销策略。服装流行色定案在传递中的滞后性和保守性制约了流行色商业价值的发挥，促动了服装流行色预测研究的发展。虽然一些理论和方法已被应用到此研究领域，并取得了一定的成果。但是，总体来说服装流行色预测研究还处于探索阶段，存在不足和争议，预测的有效性依然是纺织服装行业亟待解决的问题。
色彩具有典型的视觉特征，文化学意义上的色彩流行具有模糊性，如何将色彩的感性转化为理性，构建比较严谨的数学模型是流行色趋势研究科学性的关键。本文首先探讨了将色彩转化为数据的量化方法，然后在此基础上展开了对服装流行色预测趋势的研究。本文以国际色彩委员会(INTERCOLOR)发布的2007-2013年国际春夏女装流行色定案为研究对象，通过对INTERCOLOR所使用的色彩体系PANTONE色彩体系的分析，提出了对色彩进行量化和分类的方法；通过对服装流行色定案统计数据的分析，给出了服装流行色色相、明度和纯度特征的变化。在上述基础上，分别采用灰色系统理论和BP神经网络构建了服装流行色色相预测模型，通过设计不同长度的时间序列，探讨了各色相预测模型的预测效果。最后，对服装流行色的明度和纯度进行了预测研究，最终实现了对服装流行色预测的系统性研究。
本研究的创新点主要体现在：1）针对服装流行色预测研究中存在的色彩量化的无标准、不统一性，提出了基于PANTONE色彩体系的色彩量化和分类准则，为规范色彩的量化提供了参考，搭建了主观色彩与客观计算机色彩应用间的桥梁；2）探讨了不同长度的时间序列，即进行服装流行色预测的年份参数，对服装流行色预测模型预测效果的影响，为服装流行色流行预测研究提供了理论依据；3）提出了基于主导色相的服装流行色的明度和纯度的预测方法，丰富了服装流行色的研究理论体系。本文所做的内容主要包括：
第一章为绪论部分。对服装流行色预测知识、研究现状进行介绍和评述。通过对国内外服装流行色预测研究文献的分析，概述了这一研究领域达到的研究水平、存在的不足，继而提出本文的研究目的、研究思路和研究方法。
第二章研究了服装流行色定案的量化与分析。针对色彩的主观性和研究中存在数据来源不统一、色彩量化复杂等问题，提出以国际色彩委员会发布的2007-2013年国际春夏女装流行色定案为研究对象，以PANTONE色彩体系为色彩量化和分类的工具。确定了对PANTONE色彩体系的色相、明度、纯度区间分类的依据和标准，通过对量化数据分析获得定案的特征。
第三章探讨了基于灰色系统理论的服装流行色色相的预测研究。根据色相在定案中的重要性，以及色相数据变化的非线性特征，提出采用灰色系统理论构建服装流行色色相预测模型的研究。通过设计不同长度的时间序列：4年、5年和6年，探讨了时间序列长度对预测模型预测效果的影响，研究以平均绝对误差(MAE)作为预测效果的评价指标。但灰色GM(1,1)模型对数据光滑性要求较高，进行色相预测时不能对所有数据完成预测，其普适性受到限制。
第四章为基于BP神经网络理论的服装流行色色相的预测研究。针对灰色预测模型对服装流行色进行预测中存在的不足，构建了基于BP神经网络的服装流行色色相预测模型。通过分别设计输入层节点数为3、4和5，目标输出节点数为1的BP神经网络结构，系统地探讨了不同结构的BP神经网络模型对服装流行色色相的预测效果。
第五章和第六章分别进行了服装流行色明度和纯度的预测研究。针对已有研究中缺少色相、明度和纯度内在联系不足，提出了建立在主导色相基础上的明度均值和纯度均值的预测研究。分别以定案中10类色相对应的明度和纯度数据为研究基础，完成了服装流行色明度和纯度的预测。
第七章为结论与展望。给出了本文获得的主要结论，同时指出研究中尚存的问题以及未来研究中需要拓展的方向和思路。
本文以服装流行色定案为研究对象，提出了对色彩进行量化和分类处理的方法。在对服装流行色定案数据的分析基础上，利用灰色系统理论和神经网络模型对服装流行色定案进行了预测研究，实现了对服装流行色较高精度的预测，研究结果可以为纺织服装行业掌握未来的流行色趋势变化提供参考。
Fashion color prediction is to predict the future color trend of fashion in24months byspecial methods and to provide trend palettes. Obtained palettes could offer clothingenterprises valuable information on fashion color during the future selling seasons, which isbeneficial for drafting manufacturing and marketing plots. However, the lag and confidence ofpalettes during propagating restrict its commercial value and stimulate the development ofresearch on fashion color prediction. Though several theories and methods have been appliedinto this field and achievements are obtained, fashion color prediction is still in exploringstage. Deficiencies and controversies existed and the validity of prediction has already been apressing problem of the day in the textile and clothing industries.
Color is with typically sensitive simulation and fashion color with ambiguous expression.How to translate the sensitive color into rational data and build serious mathematicalprediction models are the key points of fashion color prediction. In this research, the methodof color quantification was discussed, based on which the prediction researches on fashioncolor were investigated. Fashion color palettes were taken as the objects. The palettes werereleased by International Commission for Color Fashion and Textiles (INTERCOLOR) forwomen’s spring/summer, from2007to2013. Quantification and classification methods oncolors were put forward. The process was based on the analyses of PANTONE color system,which was applied by the palettes. Then features of palettes on hue, lightness and chromawere obtained respectively analyzing the statistical data. Consequently, hue prediction modelswere established using grey system theory and back-propagation neural networks respectively.The predicted effects of the models were investigated by setting different time series. Finally,lightness and chroma prediction models were established on the basis of hue. The systematicprediction on fashion color was developed.
The innovations of this thesis are illustrated as follows:1) Quantification andclassification of color based on PANTONE color system were proposed in terms of thenon-standard and disunity of color quantification in fashion color prediction researches. It canprovide a valuable reference for the standardization of color and improve the communicationbetween colors on subjective perception and the objective computation application.2) Thepredicted effects of the prediction models were investigated by setting different time series,which can provide soundly basis for fashion color prediction researches.3) Prediction methodon lightness and chroma of color was put forward based on the dominant hues, which enrichesthe theory on fashion color research. The main studies of this paper were arranged as follows.
Preface of the thesis was illustrated in Chapter1. Research background, methodologyand literature was introduced. Research levels and insufficiencies in this field are summarized.Consequently, the purpose, significance and methods of this research were proposed.
Quantification and analyses of fashion color palettes were discussed in Chapter2.Inercolor palettes for women’s spring/summer, from2007to2013, were proposed as objectsand PANTONE color system was as the quantification tool, against the problems of disunityof data resources and complexity of quantification methods. Quantification and classification criteria of hue, lightness and chroma were defined. Consequently, features of palettes wereobtained by analyzing statistical data of hue, lightness and chroma.
The predictions of hues using grey system theory were investigated in Chapter3.Prediction models on hue were established by grey model, in accordance with the significanceand the nonlinear features of hue. Validities of the prediction models were discussed bysetting different time series of historical fashion color data, as4,5and6years respectively.Mean absolute error (MAE) was considered as evaluation indicator in this thesis. Resultsdemonstrated that GM (1,1) cannot predict the hues continuously due to the high requirementon historical data and limit the generality of the prediction.
The predictions on hue by using back-propagation neural networks (BPNN) wereprobed in Chapter4, in accordance with the deficiency of grey system theory. The validity ofthe prediction models were explored systematically by designing different structures of BPNNwith input neurons as3,4and5respectively.
Consequently, prediction on lightness and chroma were investigated in Chapter5and6.Prediction models on mean values of lightness and chroma, based on dominant hues, wereproposed against the absence of internal relation among the three attributes of colors.
Finally, the main achievements of the research were summed up in Chapter7of thethesis, meanwhile problems to be solved and explored in the future researches were pointedout.
Quantification and classification methods were provided based on fashion color palettesin this thesis. Fashion color prediction was discussed by using grey system theory and BPNN,which was based on analyses of statistical data of fashion color palettes. It conducted theprediction with a higher accuracy and the prediction results could provide textile industriesreference on future fashion color trend.
色彩具有典型的视觉特征，文化学意义上的色彩流行具有模糊性，如何将色彩的感性转化为理性，构建比较严谨的数学模型是流行色趋势研究科学性的关键。本文首先探讨了将色彩转化为数据的量化方法，然后在此基础上展开了对服装流行色预测趋势的研究。本文以国际色彩委员会(INTERCOLOR)发布的2007-2013年国际春夏女装流行色定案为研究对象，通过对INTERCOLOR所使用的色彩体系PANTONE色彩体系的分析，提出了对色彩进行量化和分类的方法；通过对服装流行色定案统计数据的分析，给出了服装流行色色相、明度和纯度特征的变化。在上述基础上，分别采用灰色系统理论和BP神经网络构建了服装流行色色相预测模型，通过设计不同长度的时间序列，探讨了各色相预测模型的预测效果。最后，对服装流行色的明度和纯度进行了预测研究，最终实现了对服装流行色预测的系统性研究。
本研究的创新点主要体现在：1）针对服装流行色预测研究中存在的色彩量化的无标准、不统一性，提出了基于PANTONE色彩体系的色彩量化和分类准则，为规范色彩的量化提供了参考，搭建了主观色彩与客观计算机色彩应用间的桥梁；2）探讨了不同长度的时间序列，即进行服装流行色预测的年份参数，对服装流行色预测模型预测效果的影响，为服装流行色流行预测研究提供了理论依据；3）提出了基于主导色相的服装流行色的明度和纯度的预测方法，丰富了服装流行色的研究理论体系。本文所做的内容主要包括：
第一章为绪论部分。对服装流行色预测知识、研究现状进行介绍和评述。通过对国内外服装流行色预测研究文献的分析，概述了这一研究领域达到的研究水平、存在的不足，继而提出本文的研究目的、研究思路和研究方法。
第二章研究了服装流行色定案的量化与分析。针对色彩的主观性和研究中存在数据来源不统一、色彩量化复杂等问题，提出以国际色彩委员会发布的2007-2013年国际春夏女装流行色定案为研究对象，以PANTONE色彩体系为色彩量化和分类的工具。确定了对PANTONE色彩体系的色相、明度、纯度区间分类的依据和标准，通过对量化数据分析获得定案的特征。
第三章探讨了基于灰色系统理论的服装流行色色相的预测研究。根据色相在定案中的重要性，以及色相数据变化的非线性特征，提出采用灰色系统理论构建服装流行色色相预测模型的研究。通过设计不同长度的时间序列：4年、5年和6年，探讨了时间序列长度对预测模型预测效果的影响，研究以平均绝对误差(MAE)作为预测效果的评价指标。但灰色GM(1,1)模型对数据光滑性要求较高，进行色相预测时不能对所有数据完成预测，其普适性受到限制。
第四章为基于BP神经网络理论的服装流行色色相的预测研究。针对灰色预测模型对服装流行色进行预测中存在的不足，构建了基于BP神经网络的服装流行色色相预测模型。通过分别设计输入层节点数为3、4和5，目标输出节点数为1的BP神经网络结构，系统地探讨了不同结构的BP神经网络模型对服装流行色色相的预测效果。
第五章和第六章分别进行了服装流行色明度和纯度的预测研究。针对已有研究中缺少色相、明度和纯度内在联系不足，提出了建立在主导色相基础上的明度均值和纯度均值的预测研究。分别以定案中10类色相对应的明度和纯度数据为研究基础，完成了服装流行色明度和纯度的预测。
第七章为结论与展望。给出了本文获得的主要结论，同时指出研究中尚存的问题以及未来研究中需要拓展的方向和思路。
本文以服装流行色定案为研究对象，提出了对色彩进行量化和分类处理的方法。在对服装流行色定案数据的分析基础上，利用灰色系统理论和神经网络模型对服装流行色定案进行了预测研究，实现了对服装流行色较高精度的预测，研究结果可以为纺织服装行业掌握未来的流行色趋势变化提供参考。
Fashion color prediction is to predict the future color trend of fashion in24months byspecial methods and to provide trend palettes. Obtained palettes could offer clothingenterprises valuable information on fashion color during the future selling seasons, which isbeneficial for drafting manufacturing and marketing plots. However, the lag and confidence ofpalettes during propagating restrict its commercial value and stimulate the development ofresearch on fashion color prediction. Though several theories and methods have been appliedinto this field and achievements are obtained, fashion color prediction is still in exploringstage. Deficiencies and controversies existed and the validity of prediction has already been apressing problem of the day in the textile and clothing industries.
Color is with typically sensitive simulation and fashion color with ambiguous expression.How to translate the sensitive color into rational data and build serious mathematicalprediction models are the key points of fashion color prediction. In this research, the methodof color quantification was discussed, based on which the prediction researches on fashioncolor were investigated. Fashion color palettes were taken as the objects. The palettes werereleased by International Commission for Color Fashion and Textiles (INTERCOLOR) forwomen’s spring/summer, from2007to2013. Quantification and classification methods oncolors were put forward. The process was based on the analyses of PANTONE color system,which was applied by the palettes. Then features of palettes on hue, lightness and chromawere obtained respectively analyzing the statistical data. Consequently, hue prediction modelswere established using grey system theory and back-propagation neural networks respectively.The predicted effects of the models were investigated by setting different time series. Finally,lightness and chroma prediction models were established on the basis of hue. The systematicprediction on fashion color was developed.
The innovations of this thesis are illustrated as follows:1) Quantification andclassification of color based on PANTONE color system were proposed in terms of thenon-standard and disunity of color quantification in fashion color prediction researches. It canprovide a valuable reference for the standardization of color and improve the communicationbetween colors on subjective perception and the objective computation application.2) Thepredicted effects of the prediction models were investigated by setting different time series,which can provide soundly basis for fashion color prediction researches.3) Prediction methodon lightness and chroma of color was put forward based on the dominant hues, which enrichesthe theory on fashion color research. The main studies of this paper were arranged as follows.
Preface of the thesis was illustrated in Chapter1. Research background, methodologyand literature was introduced. Research levels and insufficiencies in this field are summarized.Consequently, the purpose, significance and methods of this research were proposed.
Quantification and analyses of fashion color palettes were discussed in Chapter2.Inercolor palettes for women’s spring/summer, from2007to2013, were proposed as objectsand PANTONE color system was as the quantification tool, against the problems of disunityof data resources and complexity of quantification methods. Quantification and classification criteria of hue, lightness and chroma were defined. Consequently, features of palettes wereobtained by analyzing statistical data of hue, lightness and chroma.
The predictions of hues using grey system theory were investigated in Chapter3.Prediction models on hue were established by grey model, in accordance with the significanceand the nonlinear features of hue. Validities of the prediction models were discussed bysetting different time series of historical fashion color data, as4,5and6years respectively.Mean absolute error (MAE) was considered as evaluation indicator in this thesis. Resultsdemonstrated that GM (1,1) cannot predict the hues continuously due to the high requirementon historical data and limit the generality of the prediction.
The predictions on hue by using back-propagation neural networks (BPNN) wereprobed in Chapter4, in accordance with the deficiency of grey system theory. The validity ofthe prediction models were explored systematically by designing different structures of BPNNwith input neurons as3,4and5respectively.
Consequently, prediction on lightness and chroma were investigated in Chapter5and6.Prediction models on mean values of lightness and chroma, based on dominant hues, wereproposed against the absence of internal relation among the three attributes of colors.
Finally, the main achievements of the research were summed up in Chapter7of thethesis, meanwhile problems to be solved and explored in the future researches were pointedout.
Quantification and classification methods were provided based on fashion color palettesin this thesis. Fashion color prediction was discussed by using grey system theory and BPNN,which was based on analyses of statistical data of fashion color palettes. It conducted theprediction with a higher accuracy and the prediction results could provide textile industriesreference on future fashion color trend.
引文
1.孟雅丽.色彩的价值——专访中国流行色协会常务副会长梁勇[J].中国纺织,2007(6):124-126.
2.吴永.探索流行色的奥秘[M].北京:轻工业出版社,1986.7.
3.胡月.流行色是绝对的商机[C].第四届亚洲色彩论坛论文集.2007,87-90.
4.招霞.色彩流行趋势的发展与研究[J].涂料技术与文摘,2011,9:6-10.
5.李亮之.色彩经济与营销战略[C].中国流行色协会学术年会学术论文集,2012,66-71.
6.克里斯蒂娜.不确定时代的流行色[J].生活,2009(1):76.
7.崔唯.流行色与设计[M].北京:中国纺织出版社,2013.38-50.
8.肖彬,崔唯.新论流行色的基本特征[J].流行色,2008(17):12-17.
9.李超德.色彩在现代设计中的多重语义[J].装饰,2011,218(6):59-61.
10. Miho S. A cross-cultural study on color preference in three Asian cities[J]. Japanese PsychologicalResearch,1994,36(4):219-232.
11.蔡作意.国际流行色研究[M].杭州:浙江美术学院出版社,1989.7.
12.张连生,陆宵虹.设计色彩开发拓展应用型色彩[M].南京:南京师范大学出版社,2009.104-105.
13.王蕴强.服装色彩学[M].北京：中国纺织工业出版社，2010.127-128.
14. Brannon E L. Fashion forecasting[M]. New York: Fairchild Publications,2000.147-150.
15.国际流行色委员会[EB/OL]. http://www.intercolor.nu,2006-6-5.
16.吴永.探索流行色的奥秘[M].北京:轻工业出版社,1986.141-142.
17.流行色趋势预测发布和中国发布历史[J].流行色,2007(10):97-103.
18.徐志瑞.走在潮流浪尖上——流行预测的灵感来源[J].流行色,1997(3):50-51.
19.吴镇保,张闻彩.色彩理论与应用[M].南京:江苏美术出版社,1992.12-25.
20.傅炯,韩挺.产品流行色基本原理和研究方法[C].第四届亚洲色彩论坛论文集,2007:7.
21.袁大鹏，孔祥梅.时尚色彩——流行色初探[J].东北电力学院学报，2004,24(3):67-69.
22.日本流行色研究所.确定流行色[EB/OL]. http://www.jafca.org/trendcolor,2008-1-26.
23.董奎勇.快速反应—21世纪纺织服装企业竞争取胜之路[J].江苏纺织,2001(12):19-22.
24.阎玉秀,朱江晖.女套装上衣尺寸自动生成系统的建立与实现[J].纺织学报,2011(9),142-146.
25.徐琪.服装供应链基于RFID的仓储配送智能化管理[J].纺织学报,2010(9),137-142.
26.陈林龙.采用FPV法预测流行色[J].流行色,1991(4):8-9.
27.余平.计算机在流行色分析预测中的应用[D].天津:天津工业大学,2004.
28.宋建明.1998/99年秋冬季国际流行者色定案应用色谱分析[J].流行色,1997(5):24-26.
29.崔唯,张玮,戴艳.1996-2005年Intercolor春夏色案分析(续上期)[J].国际纺织品流行趋势,2007(3):36-38.
30.戴燕.1996-2005年国际流行色女装定案与国内女装色彩的分析与研究[D].北京:北京服装学院,2006.
31.虞台文,吴瑞卿,魏朝宏.灰色系统理论于流行色预测上的研究[J].设计学报,1998,3(2):61-71.
32.李熠,吴志明.服装流行色的色彩意向灰色预测[J].纺织学报,2009,30(4):94-100.
33.狄宏静.基于BP神经网络的春夏女装流行色预测研究[D].无锡:江南大学纺织服装学院,2011.
34.吴也哲,翟永超,孙莉.基于灰色前馈神经网络的流行色预测[J].东华大学学报(自然科学版),2011,37(2):199-204.
35. Lin J J, Sun P T, Chen J J, et al. Applying grey model to predict trend of textile fashion colors [J].Journal of the Textile Institute,2010,101(4):360-368.
36. Linton H. Color forecasting: a survey of international color marketing[M]. New York: Van NostrandReinhold,1994.23-30.
37. Stanstield J, Whitfield TW. Can future color trends be forecasted on the basis of past color trends?: anempirical investigation[J]. Color Research&Application,2005,30(3):235-242.
38. Diane T, Cassidy T. Color Forecasting [M]. Oxford: Blackwell Publishing Ltd.,2005.96-99.
39. Diane T. Personal color analysis, consumer color preferences and color forecasting for the fashion andtextile industries[J]. Color Design&Creativity,2007,1(1):1-14.
40. Mello P, Storari S, Valli B. A knowledge-based system for fashion trend forecasting[C]. IEA/AIE2008,425-434.
41. Choi T M, Hui C L, Ng S F, et al. Color trend forecasting of fashionable products with very fewhistorical data [J]. Systems, Man, and Cybernetics,2012,42(6):1003-1010.
42. Yu Y, Hui C L, Choi Ts M. An empirical study of intelligent expert systems on forecasting of fashioncolor trend[J]. Expert Systems with Applications,2012(39):4383-4389.
43.崔唯,张玮,戴艳.1996-2005年Intercolor春夏色案分析[J].国际纺织品流行趋势,2007(3):27-29.
44.陈林龙.采用FPV法预测流行色(续一)[J].流行色,1992(1):16.
45.杨东奇.对流行色理论及调研方法的思考[C].2011-2003中国流行色优秀论文集,2003:35-41.
46.崔唯,周韵.2012/13秋冬国际流行色委员会定案色卡空间位置分析[J].江苏纺织,2011(9):35-44.
47.张全,康文科,明新.基于粗集的产品色彩预测系统研究[J].计算机工程与应用,2008,44(11):31-33.
48. Wu Y Z, Sun L, Le J J. G-LMBPNN: A new fashion color prediction model[C]. InternationalConference on Computational Aspects of Social Networks,2010:501-504.
49.顾雯.基于波浪理论与预测学方法的国际男装流行色趋势预测研究[D].上海:东华大学纺织服装学院,2013.
50. Gu W, Liu X G. System of color research management for fashion color forecasting[C]. InternationalConference on Engineering Management and Service Science.2010:1-4.
51.常丽霞,高卫东,张万琴等.基于马尔科夫的国际服装流行色预测研究[[J].毛纺科技,2012,40(7):44-47.
52.廖进球,李志强.市场调查与预测[M].长沙:湖南大学出版社,2008.197.
53.色研.色彩流行趋势分析与国际因素影响[J].江苏纺织,2011(12):10-11.
54.徐志瑞.现代国际流行色应用基础[J].流行色,2004(11):10-11.
55.宋建明.色彩流行趋势研究和推广必须本土化[J].中国服装,2004(11):12.
56.黄国松.色彩设计学[M].北京:中国纺织出版社,2001.87-88.
57.田少熙.数字色彩构成[M].北京:清华大学出版社,2007.11-13.
58.曹从军.色彩管理关键技术CIE L~*a~*b~*与CMYK变换算法的研究[D].西安:西北工业大学,2008.
59.色研.色立体的主要学派及应用色彩体系[J].江苏纺织，2011(8):35-40.
60. Chang L X, Gao W D, Zhang X. Discussion on fashion color forecasting researches for textile andfashion industries[J]. Journal of Fiber Bioengineering and Informatics,2009,2(1):14-19.
61. Pantone fashion theory [EB/OL]. http://pantone.custhelp.com/app/answers/detail/a_id/1403.
62.宁宣熙，刘思峰.管理预测与决策方法[M].北京:科学出版社,2009.9.
63.李莉婷.色彩构成[M].合肥:安徽美术出版,2001.11.
64. Hsu C C, Chen C Y. Applications of improved grey prediction model for power demand forecasting[J].Energy Conversation and Management,2003(44):2241-2249.
65.刘思峰,党耀国,方志耕等.灰色系统理论及其应用[M].北京:科学出版社,2004.1.
66. Deng J L. Figure on difference information space in grey relational analysis[J]. Journal of GreySystem,2004,16(2):96-100.
67. Deng J L. On boundary of grey input b in GM(1,1)[J]. Journal of Grey System,2001,13(2):14-20.
68.邓聚龙.灰色系统基本方法[M].武汉:华中科技大学出版社,2004.10.
69.崔立志.灰色预测技术及其应用研究[D].南京:南京航空航天大学经济与管理学院,2010.
70. Liu S F. Grey forecast of drought and inundation in Henan Province[J]. Journal of Grey System,1994,6(4):279-288.
71. Hsu L C. A genetic algorithm based nonlinear grey Bernoulli model for output forecasting inintegrated circuit industry[J]. Experts Systems with Applications,2009,26(11):1-6.
72. Deng J L. The foundation of grey system theory [J]. Journal of Grey System,1997,9(1):40-48.
73.易静,杜昌廷,王润华等.应用灰色预测模型GM(1,1)对结核病发病率进行预测[J].重庆医科大学学报,2007,32(3):275-278.
74. Liu S F, Forrest. The role and position of Grey System Theory in science development[J]. Journal ofGrey System,1997,9(4):351-356.
75. Liu S F.The three axioms of buffer operator and their application[J]. Journal of Grey System,1991,3(1):39-48.
76. Zheng Z Y, Chen Y Q, Lu X S. An improved grey model and its application research on the predictionof real-time GPS satellite clock errors[J]. Chinese Astronomy and Astrophysics,2009(33):72-89.
77.邓聚龙.灰色系统基本方法(第二版)[M].武汉:华中科技大学出版社,2005.26.
78.严渊智.灰色系统使用程序[M].南京:江苏科学技术出版社,1989.4.
79.谢家发.统计学原理[M].郑州:郑州大学出版社,2008.276.
80.胡勇.灰色预测模型GM（1,1）及其在奥运金牌数预测中的应用[J].德州学院学报,2005,21(4):39-41.
81. Behera L, Kumar S, Patnaik A. On adaptive learning rate thtat guarantees convergence in feed forwardnetworks[J]. Neutral Networks,2006,17(5):1116-1125.
82. Nagy H M,Watanabe K. Prediction of sediment load concentration in rivers using artificial neuralnetwork mode[J]. Journal of Hydraulic Engineering,2002,128(56):588-596.
83. Kawato M, Furukawa K, Suzuki R. Ahierarchical neural network model for control and learning ofboundary movement[J]. Biological Cybernetics,1987,57(3):169-185.
84. Ashok K G, Suresh C, Surekha B. A fast learning algorithm for training feed forward neural networks.International[J]. Journal of Systems Science,2006,37(10):709-722.
85.王玲.基于人工神经网络的水文过程模拟研究[D].南京:河海大学水环学院,2002.
86. Ashok K G, Suresh C, Surekha B. A fast learning algorithm for training feed forward neural networks[J]. International Journal of Systems Science,2006,37(10):709-722.
87. Nagy H M, Watanabe K. Prediction of sediment load concentration in rivers using artificial neuralnetwork mode[J]. Journal of Hydraulic Engineering,2002,128(56):588-596.
88.韩力群.人工神经网络教程[M].北京:北京邮电大学出版社,2006.55-58.
89. Au K F, Choi Ts M, Yu Y. Fashion retail forecasting by evolutionary neural networks[J]. InternationalJournal of Production Economics,2008(114):615-630.
90.朱大奇,史慧.人工神经网络原理及应用[M].北京:科技出版社,2006.34-38.
91. Wang X G, Tang H. An improved back propagation algorithm to avoid the local minima problem[J].Euro-Computing,2004,56(3):455-460.
92. Chen F L, Ou Ts Y. Constructing a sales forecasting model by integrating GRA and ELM: a case studyfor retail industry[J]. International Journal of Electronic Business Management,2011,9(2):107-121.
93.吴桂芳,林青松,孙秀明等.基于小样本的神经网络参数优化选择方法[J].计算机应用,2008,28(7):1162-1671.
94.姜波.灰色系统与神经网络分析方法及其应用研究[D].武汉:华中科技大学,2004.
95. Kolmogrov A N. On the representation of continuous function of many variables by superposition ofcontinuous functions of one variable and addition[J]. American Math Society Trans,1963,28(1):55-59.
96. Hetch N R. Theory of the back propagation neural network[C]. Proceeding of InternationalConference on Neural Networks,1989,593-603.
97.吴雄华,刘亚.基于BP神经网络的熔喷非织造布工艺参数优化[J].纺织学报,2011,32(1):51-54.
98.崔唯,周钧.2011年春夏国际流行色委员会定案色卡空间位置分析[J].江苏纺织,2010(6):35-40.
99. Chang L X, Pan R R, Gao W D. Fashion color forecasting by applying an improved BP neuralnetwork[J].Journal of Donghua University (English Edition),2013,30(1):58-62.
2.吴永.探索流行色的奥秘[M].北京:轻工业出版社,1986.7.
3.胡月.流行色是绝对的商机[C].第四届亚洲色彩论坛论文集.2007,87-90.
4.招霞.色彩流行趋势的发展与研究[J].涂料技术与文摘,2011,9:6-10.
5.李亮之.色彩经济与营销战略[C].中国流行色协会学术年会学术论文集,2012,66-71.
6.克里斯蒂娜.不确定时代的流行色[J].生活,2009(1):76.
7.崔唯.流行色与设计[M].北京:中国纺织出版社,2013.38-50.
8.肖彬,崔唯.新论流行色的基本特征[J].流行色,2008(17):12-17.
9.李超德.色彩在现代设计中的多重语义[J].装饰,2011,218(6):59-61.
10. Miho S. A cross-cultural study on color preference in three Asian cities[J]. Japanese PsychologicalResearch,1994,36(4):219-232.
11.蔡作意.国际流行色研究[M].杭州:浙江美术学院出版社,1989.7.
12.张连生,陆宵虹.设计色彩开发拓展应用型色彩[M].南京:南京师范大学出版社,2009.104-105.
13.王蕴强.服装色彩学[M].北京：中国纺织工业出版社，2010.127-128.
14. Brannon E L. Fashion forecasting[M]. New York: Fairchild Publications,2000.147-150.
15.国际流行色委员会[EB/OL]. http://www.intercolor.nu,2006-6-5.
16.吴永.探索流行色的奥秘[M].北京:轻工业出版社,1986.141-142.
17.流行色趋势预测发布和中国发布历史[J].流行色,2007(10):97-103.
18.徐志瑞.走在潮流浪尖上——流行预测的灵感来源[J].流行色,1997(3):50-51.
19.吴镇保,张闻彩.色彩理论与应用[M].南京:江苏美术出版社,1992.12-25.
20.傅炯,韩挺.产品流行色基本原理和研究方法[C].第四届亚洲色彩论坛论文集,2007:7.
21.袁大鹏，孔祥梅.时尚色彩——流行色初探[J].东北电力学院学报，2004,24(3):67-69.
22.日本流行色研究所.确定流行色[EB/OL]. http://www.jafca.org/trendcolor,2008-1-26.
23.董奎勇.快速反应—21世纪纺织服装企业竞争取胜之路[J].江苏纺织,2001(12):19-22.
24.阎玉秀,朱江晖.女套装上衣尺寸自动生成系统的建立与实现[J].纺织学报,2011(9),142-146.
25.徐琪.服装供应链基于RFID的仓储配送智能化管理[J].纺织学报,2010(9),137-142.
26.陈林龙.采用FPV法预测流行色[J].流行色,1991(4):8-9.
27.余平.计算机在流行色分析预测中的应用[D].天津:天津工业大学,2004.
28.宋建明.1998/99年秋冬季国际流行者色定案应用色谱分析[J].流行色,1997(5):24-26.
29.崔唯,张玮,戴艳.1996-2005年Intercolor春夏色案分析(续上期)[J].国际纺织品流行趋势,2007(3):36-38.
30.戴燕.1996-2005年国际流行色女装定案与国内女装色彩的分析与研究[D].北京:北京服装学院,2006.
31.虞台文,吴瑞卿,魏朝宏.灰色系统理论于流行色预测上的研究[J].设计学报,1998,3(2):61-71.
32.李熠,吴志明.服装流行色的色彩意向灰色预测[J].纺织学报,2009,30(4):94-100.
33.狄宏静.基于BP神经网络的春夏女装流行色预测研究[D].无锡:江南大学纺织服装学院,2011.
34.吴也哲,翟永超,孙莉.基于灰色前馈神经网络的流行色预测[J].东华大学学报(自然科学版),2011,37(2):199-204.
35. Lin J J, Sun P T, Chen J J, et al. Applying grey model to predict trend of textile fashion colors [J].Journal of the Textile Institute,2010,101(4):360-368.
36. Linton H. Color forecasting: a survey of international color marketing[M]. New York: Van NostrandReinhold,1994.23-30.
37. Stanstield J, Whitfield TW. Can future color trends be forecasted on the basis of past color trends?: anempirical investigation[J]. Color Research&Application,2005,30(3):235-242.
38. Diane T, Cassidy T. Color Forecasting [M]. Oxford: Blackwell Publishing Ltd.,2005.96-99.
39. Diane T. Personal color analysis, consumer color preferences and color forecasting for the fashion andtextile industries[J]. Color Design&Creativity,2007,1(1):1-14.
40. Mello P, Storari S, Valli B. A knowledge-based system for fashion trend forecasting[C]. IEA/AIE2008,425-434.
41. Choi T M, Hui C L, Ng S F, et al. Color trend forecasting of fashionable products with very fewhistorical data [J]. Systems, Man, and Cybernetics,2012,42(6):1003-1010.
42. Yu Y, Hui C L, Choi Ts M. An empirical study of intelligent expert systems on forecasting of fashioncolor trend[J]. Expert Systems with Applications,2012(39):4383-4389.
43.崔唯,张玮,戴艳.1996-2005年Intercolor春夏色案分析[J].国际纺织品流行趋势,2007(3):27-29.
44.陈林龙.采用FPV法预测流行色(续一)[J].流行色,1992(1):16.
45.杨东奇.对流行色理论及调研方法的思考[C].2011-2003中国流行色优秀论文集,2003:35-41.
46.崔唯,周韵.2012/13秋冬国际流行色委员会定案色卡空间位置分析[J].江苏纺织,2011(9):35-44.
47.张全,康文科,明新.基于粗集的产品色彩预测系统研究[J].计算机工程与应用,2008,44(11):31-33.
48. Wu Y Z, Sun L, Le J J. G-LMBPNN: A new fashion color prediction model[C]. InternationalConference on Computational Aspects of Social Networks,2010:501-504.
49.顾雯.基于波浪理论与预测学方法的国际男装流行色趋势预测研究[D].上海:东华大学纺织服装学院,2013.
50. Gu W, Liu X G. System of color research management for fashion color forecasting[C]. InternationalConference on Engineering Management and Service Science.2010:1-4.
51.常丽霞,高卫东,张万琴等.基于马尔科夫的国际服装流行色预测研究[[J].毛纺科技,2012,40(7):44-47.
52.廖进球,李志强.市场调查与预测[M].长沙:湖南大学出版社,2008.197.
53.色研.色彩流行趋势分析与国际因素影响[J].江苏纺织,2011(12):10-11.
54.徐志瑞.现代国际流行色应用基础[J].流行色,2004(11):10-11.
55.宋建明.色彩流行趋势研究和推广必须本土化[J].中国服装,2004(11):12.
56.黄国松.色彩设计学[M].北京:中国纺织出版社,2001.87-88.
57.田少熙.数字色彩构成[M].北京:清华大学出版社,2007.11-13.
58.曹从军.色彩管理关键技术CIE L~*a~*b~*与CMYK变换算法的研究[D].西安:西北工业大学,2008.
59.色研.色立体的主要学派及应用色彩体系[J].江苏纺织，2011(8):35-40.
60. Chang L X, Gao W D, Zhang X. Discussion on fashion color forecasting researches for textile andfashion industries[J]. Journal of Fiber Bioengineering and Informatics,2009,2(1):14-19.
61. Pantone fashion theory [EB/OL]. http://pantone.custhelp.com/app/answers/detail/a_id/1403.
62.宁宣熙，刘思峰.管理预测与决策方法[M].北京:科学出版社,2009.9.
63.李莉婷.色彩构成[M].合肥:安徽美术出版,2001.11.
64. Hsu C C, Chen C Y. Applications of improved grey prediction model for power demand forecasting[J].Energy Conversation and Management,2003(44):2241-2249.
65.刘思峰,党耀国,方志耕等.灰色系统理论及其应用[M].北京:科学出版社,2004.1.
66. Deng J L. Figure on difference information space in grey relational analysis[J]. Journal of GreySystem,2004,16(2):96-100.
67. Deng J L. On boundary of grey input b in GM(1,1)[J]. Journal of Grey System,2001,13(2):14-20.
68.邓聚龙.灰色系统基本方法[M].武汉:华中科技大学出版社,2004.10.
69.崔立志.灰色预测技术及其应用研究[D].南京:南京航空航天大学经济与管理学院,2010.
70. Liu S F. Grey forecast of drought and inundation in Henan Province[J]. Journal of Grey System,1994,6(4):279-288.
71. Hsu L C. A genetic algorithm based nonlinear grey Bernoulli model for output forecasting inintegrated circuit industry[J]. Experts Systems with Applications,2009,26(11):1-6.
72. Deng J L. The foundation of grey system theory [J]. Journal of Grey System,1997,9(1):40-48.
73.易静,杜昌廷,王润华等.应用灰色预测模型GM(1,1)对结核病发病率进行预测[J].重庆医科大学学报,2007,32(3):275-278.
74. Liu S F, Forrest. The role and position of Grey System Theory in science development[J]. Journal ofGrey System,1997,9(4):351-356.
75. Liu S F.The three axioms of buffer operator and their application[J]. Journal of Grey System,1991,3(1):39-48.
76. Zheng Z Y, Chen Y Q, Lu X S. An improved grey model and its application research on the predictionof real-time GPS satellite clock errors[J]. Chinese Astronomy and Astrophysics,2009(33):72-89.
77.邓聚龙.灰色系统基本方法(第二版)[M].武汉:华中科技大学出版社,2005.26.
78.严渊智.灰色系统使用程序[M].南京:江苏科学技术出版社,1989.4.
79.谢家发.统计学原理[M].郑州:郑州大学出版社,2008.276.
80.胡勇.灰色预测模型GM（1,1）及其在奥运金牌数预测中的应用[J].德州学院学报,2005,21(4):39-41.
81. Behera L, Kumar S, Patnaik A. On adaptive learning rate thtat guarantees convergence in feed forwardnetworks[J]. Neutral Networks,2006,17(5):1116-1125.
82. Nagy H M,Watanabe K. Prediction of sediment load concentration in rivers using artificial neuralnetwork mode[J]. Journal of Hydraulic Engineering,2002,128(56):588-596.
83. Kawato M, Furukawa K, Suzuki R. Ahierarchical neural network model for control and learning ofboundary movement[J]. Biological Cybernetics,1987,57(3):169-185.
84. Ashok K G, Suresh C, Surekha B. A fast learning algorithm for training feed forward neural networks.International[J]. Journal of Systems Science,2006,37(10):709-722.
85.王玲.基于人工神经网络的水文过程模拟研究[D].南京:河海大学水环学院,2002.
86. Ashok K G, Suresh C, Surekha B. A fast learning algorithm for training feed forward neural networks[J]. International Journal of Systems Science,2006,37(10):709-722.
87. Nagy H M, Watanabe K. Prediction of sediment load concentration in rivers using artificial neuralnetwork mode[J]. Journal of Hydraulic Engineering,2002,128(56):588-596.
88.韩力群.人工神经网络教程[M].北京:北京邮电大学出版社,2006.55-58.
89. Au K F, Choi Ts M, Yu Y. Fashion retail forecasting by evolutionary neural networks[J]. InternationalJournal of Production Economics,2008(114):615-630.
90.朱大奇,史慧.人工神经网络原理及应用[M].北京:科技出版社,2006.34-38.
91. Wang X G, Tang H. An improved back propagation algorithm to avoid the local minima problem[J].Euro-Computing,2004,56(3):455-460.
92. Chen F L, Ou Ts Y. Constructing a sales forecasting model by integrating GRA and ELM: a case studyfor retail industry[J]. International Journal of Electronic Business Management,2011,9(2):107-121.
93.吴桂芳,林青松,孙秀明等.基于小样本的神经网络参数优化选择方法[J].计算机应用,2008,28(7):1162-1671.
94.姜波.灰色系统与神经网络分析方法及其应用研究[D].武汉:华中科技大学,2004.
95. Kolmogrov A N. On the representation of continuous function of many variables by superposition ofcontinuous functions of one variable and addition[J]. American Math Society Trans,1963,28(1):55-59.
96. Hetch N R. Theory of the back propagation neural network[C]. Proceeding of InternationalConference on Neural Networks,1989,593-603.
97.吴雄华,刘亚.基于BP神经网络的熔喷非织造布工艺参数优化[J].纺织学报,2011,32(1):51-54.
98.崔唯,周钧.2011年春夏国际流行色委员会定案色卡空间位置分析[J].江苏纺织,2010(6):35-40.
99. Chang L X, Pan R R, Gao W D. Fashion color forecasting by applying an improved BP neuralnetwork[J].Journal of Donghua University (English Edition),2013,30(1):58-62.