本研究的目的在於建立一布料設計自動化的流程，運用類神經網路進行學習，再利用遺傳演算法將設計構想做有效的大量發散，將人類的感性思維帶入電腦程式運算中，使設計流程中之構想發展階段，藉由設計自動化達到更高的效率，刺激設計師靈感發散，提高布料設計構想發展的品質，降低設計的流程與成本。

　　以往設計師進行構想發展階段時，採用的是傳統手工繪製草圖，然而此一傳統方式在追求高效率高度競爭的經濟環境下，已然不適用。產品的生命週期明顯較以往縮短，加上電腦輔助設計軟體的技術進步，更縮短了產品設計時程。本研究著眼於設計流程中之構想發展階段，導入自動化設計概念，以期能獲得大量、快速且有用的設計構想，藉此激發設計師的創造力。

　　本研究方法是利用集群分析，將蒐集而得的女性衣服布料予以分群，挑選出代表性布料40張，將其以問卷對九組感性語彙對進行評分；抽取出布料的色彩黏著度當作其色彩特徵，並利用「極座標的傅立葉－小波描述子」（Polar Fourier-Wavelet descriptor）PFW之演算法抽取出布料的紋理特徵。以特徵值為輸入層，感性語彙得分為輸出層，進行倒傳遞類神經網路訓練。再將抽取出的布料特徵，運用遺傳演算法產生大量的特徵子代，並以學習成功的倒傳遞類神經網路為其評選標準，取得符合的特徵值，再依其特徵值產生新的布料。

The purpose of the research was to build a process of computer aided design for fabric. This paper calculated human percept with computer program, to learn with Neural Network Algorithm and to obtain a lot of concepts effectively with Genetic Algorithm. In the stage of conceptual design, it can increase efficiency and productivity by automatic design, stimulate designers’ imagination, improve the quality of fabric design, and reduce the cost of design process.

In the stage of conceptual design, designers are used to sketch by hand traditionally. However, the traditional way has not been suitable in this competitive economy. Because the cycle of productions is shortened and the technology of computer aided design improves obviously, the time for product design is reduced. This paper’s emphasis on the concept of automatic design is broached in the stage of conceptual design. It is looking forward to gaining a large number, fast and useful concept to encourage designers' creation.

At first, this paper uses Cluster Analysis. Female garments fabrics assembled were divided into groups and then we chose 40 representative fabrics from these groups. The subjects were invited to measure their subjective impression of 40 different fabric images using the Semantic Differential Method (SD). Second, we extracted Color Coherent Vector（CCV）as the color features of garments and the texture features of garments using Polar Fourier-Wavelet descriptor. The input layer is the features of garments and the output layer is the value of 9 impression words. The Back-Propagation Networks （BPN）was trained to approximate the relationship between kansei features and the features of garments. Third, we obtained a large number of filial generations by Genetic Algorithm, selected suitable filial generations by the successfully trained back-propagation network, and obtained new fabrics.

1. A. Laine, and J. Fan, “Texture classification by wavelet packet signatures”, IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 15,No. 11, November 1993.

2. Aleksandra Mojsilovic and Jianying Hu, “A Method for Color Content Matching of Images, ” IEEE International Conference on Multimedia and Expo (II), pp. 649-652, 2000.

3. Rafael C. Gonzalez, Richard E. Woods, “Digital Image Processing, ” Pearson Education, 2002.

4. E. R. Dougherty and J. C. Handley, “Recursive maximum-likelihood estimation in the one-dimensional discrete Boolean random set model, ” J. Signal Process., vol. 43, pp. 1–15, 1995.

5. Flickner, M. et al., “Query by image and video content: The qbic system, ”IEEE Computers, pp. 23-31, 1995.

6. García-Sevilla Pedro and Maria Petrou, “Analysis of Irregularly Shaped Texture Regions, ” Computer Vision and Image Understanding, vol. 84, pp. 62-76, October 2001.

7. G. Chen, and T. D. Bui, “Invariant Fourier-wavelet descriptor for pattern recognition”, Pattern Recognition, Vol. 32, pp.1083-1088, 1999.

8. Greg Pass, Ramin Zabih, and Justin Miller, “Comparing images using color coherence vectors”, In Proceedings of ACM Multimedia 96, pp. 65-73, Boston, MA, USA, 1996.

9. Hideyuki Tamura, Shunji Mori, and Takashi Yamawaki, “Texture features corresponding to visual perception, ” IEEE Transactions on Systems, Man, and Cybernetics, No. 6, JUN 1978.

10.http://www.nacs.uci.edu/~wiedeman/cspace/me/inforgb.html.

11.http://www.intl-light.com/handbook/ch01.html.

12.I.K. Sethi and I. Coman,, “Image retrieval using hierarchical self-organizing feature maps,” Vision and Neural Networks Laboratory, Department of Computer Science, Wayne State University, pp.1337-1345, 1999.

13.Ishihara, S.I., Keiko, “Nagamachi. etal, An automatic builder for a Kansei Engineering expert system using self-organizing neural networks, ”International Journal of Industrial Ergonomics, 15(1): p. 13-24, 1995.

14.J. H.Holland, “Adaptation in Natural and Artificial Systems”, Bradford Books, 1975.

15.Jing Huang and Ramin Zabih, “Combining color and spatial information for content-based image retrieval”, European Conference on Digital Libraries, September 1998 (web proceedings only). http://www.cs.cornell.edu/rdz

16.J. R. Smith, and Shih-Fu Chang, “Single color extraction and image query”, In Proceedings of International Conference on Image Processing, Vol. 3, pp.528-531, 1995.

17.K. I. Laws, “Texture image segmentation, ” Ph.D. dissertation, Image Processing Inst., Univ. of Southern California, 1980.

18.Li Wang, and Dong-Chen He, “Texture classfication using texture specturm”, Pattern Recognition, Vol. 23, No. 8, pp.905-910, 1990.

19.M. D. Shieh and C. C. Cheng, “Development of an Intelligent Fabric Retrieval System using Computer-Based Kansei Algorithm, ”Journal of the Asian Design International Conference, Vol.1, 2003.

20.Minami, K.,Nakajima H. and Toyoshima, T. “Real-time Discrimination of Ventricular Tachyarrhythmia with Fourier-Transform Neural Network” IEEE Trans. On Biomed. Eng. Vol.46, No.2,pp.179-185,1999.

21.Nagamachi, M., “Kansei engineering as a powerful consumer-oriented technology for product development, ”Applied Ergonomics, 33(3): p. 289-294, 2002.

22.Nakada, K., “Kansei engineering research on the design
of construction machinery, ” International Journal of Industrial Ergonomics, 19(2): pp. 129-146, 1997.

23.P.Garcia-Sevilla and M.Petrou, “Classification of binary textures using the 1-D Boolean model” IEEE Transactions of Image Processing, Vol 8, No 10, pp. 1457-1462, 1999.

24.R. M. Haralick, “Statistical and structural approaches to texture,” Proceedings of The IEEE, Vol.67, No. 5, MAY 1979.

25.R. M. Haralick, K. Shanmugam, and I. Dinstein, “Textural features for image classification”, IEEE Transactions on Systems, Man, and Cybernetics, Vol. SMC-3, No. 6, November 1973.

26.T. Chang and C.C.J. Kuo, “Texture analysis and classification with tree-structured wavelet transform, ”IEEE Trans. IP, vol. 2, no. 4, pp. 429-441, 1993.

27.T. Ojala, and M. Pietikainen, “Unsupervised texture segmentation using feature distributions”, Pattern Recognition, Vol. 32, pp.477-486, 1999.

28.Vincenzo Di Lecce and Andrea Guerriero, “An Evaluation of the Effectiveness of Image Features for Image Retrieval,” JVCIR(10), No. 4, pp. 351-362, December 1999.

29.Wael M. Al-Hasawi, , Khaled M. El-Naggar, “A genetic based algorithm for voltage flicker measurement”,2004.

30.W. Niblack, R. Barber, and et al, “The QBIC project: Querying images by content using color, texture and shape,” In Proc. SPIE Storage and Retrieval for Image and Video Databases, Feb. 1994.

31.Xiang Sean Zhou, Thomas S. Huang, “Image Retrieval: Feature Primitives, Feature Representation, and Relevance Feedback,” IEEE Workshop on Content-Based Access of Image and Video Libraries, in Conjunction with CVPR'00, pp. 10-
14, 2000.

32.Xie, X.Y.M., “Hand image segmentation using color and
RCE neural networ, ” Computational Intelligence in Robotics and AutomationIEEE International Symposium , 34: p. 235-250, 2001.

33.X. Rui, C.-H. Chang and T. Srikanthan, “On the Initialization and Training Methods for Kohonen Self-Organizing Feature Maps in Color Image Quantization, ” DELTA , pp. 321-325, 2002.

34.尹柏元，“以遺傳演算法與權重式最近鄰居分類法篩選遺傳疾病基因之研究”， 長庚大學資訊管理研究所碩士論文，2003。

35.李宏傑,“基於紋理與色彩特徵的影像檢索系統之研究”,淡江大學電機工程學系電子電路組碩士班, 2001。

36.周明，孫樹棟，“遺傳法原理應用”，國防工業出版社，北京，1998。

37.連國珍，“數位影像處理”，儒林圖書公司，台北市，民國89 年。

38.黃代鈞，“以遺傳演算法結合貝氏分類法快速篩選與遺傳疾病相關的基因”， 長庚大學資訊管理研究所碩士論文，2003。

39.楊明燿，“利用小波與類神經網路進行心電圖特徵擷取與病症分類”，私立中原大學醫學工程學系碩士論文，1997。

40.楊朝樑，“使用主要元素分析神經網路的心電圖病症辨識系統”，國立中正大學電機工程研究所碩士論文，2002。

41.鄭家杰，“影像搜尋系統輔助配色之研究-以布料影像為例”，國立成功大學工業設計學系研究所碩士論文，2003。

42.蕭世文， “類神經網路在產品造形設計上的應用研究”, 國立成功大學工 業設計研究所, 1990。