現今電腦繡布(Machine Embroidery)生產作業流程可分為前端人工打版作業與後端刺繡機台作業兩部份，後端已有自動化電腦刺繡機台，但前端打版作業仍需大幅仰賴專業技術人力於打版軟體中製版，藉由人眼觀察以及使用各種顏色仔細描繪於各區域之不同圖案，過程耗時費力，故本論文研製自動化影像分析系統於電腦繡布製版，以自動化演算法流程解決前端人力缺乏與縮短製版時間問題。
由於電腦繡布技術一般用於大型布料，布料必需繡以重複圖案使布料色彩豐富，為解決前端辨識重複圖案問題，因此本論文設計一套電腦繡布影像自動化辨識重複圖案演算法，提出新的適應性函數(Fitness Function)用於基因演算法以建立強健的自動化重複圖案辨識演算法。為解決前端打版之自動化分色程序，探討電腦繡布影像製版之分色以非監督式聚類法(Unsupervised Clustering Method)進行分色，以聚類有效性(Cluster Validity)分析K均值(K-means)、K中心點(K-medoid)、模糊C平均(Fuzzy C-mean, FCM)、Gustafson-kessel(G-K)聚類演算法，探討適合電腦繡布之聚類演算法與有效性指標進行分色與辨識顏色數。最後建立電腦繡布影像製版之自動化技術，本論文提出一套針對電腦繡布影像進行自動化影像辨識重複圖案及分色系統，首先以彩色混合中值濾波器(Color Hybrid Median Filter) 濾除雜訊，以彩色雙向濾波器(Color Bilateral Filter)平滑影像且保留邊界，運用二維離散小波轉換(Discrete Wavelet Transform, DWT)技術降低尺度及保留低頻影像，藉由本論文提出新的適應性函數於基因演算法自動辨識重複圖案影像且以影像金字塔概念還原無失真影像，再以頻率域相似度模板比對驗證重複圖案影像，利用非監督式G-K演算法與分割指標求取電腦繡布影像顏色數、以及顏色對應區域，最後以半色調指定顏色種類。
實驗結果顯示本論文能自動化準確辨識切割出重複圖案影像與探討非監督式聚類演算法之最佳演算於電腦繡布分色中以及自動化精確分色且快速製版於電腦繡布。

The present machine embroidery production operation flow can be divided into front end manual plate making operation and back end embroidery machine operation. Although there has been automatic embroidery machine at the back end, the front end plate making operation still depends significantly on professional labor using plate making software for plate making, visual inspection, and painting different colors in different patterns of various regions. The process is time and labor-consuming. Therefore, this study developed an automatic image analysis system for machine embroidery plate making, using automatic algorithm flow to solve the shortage of front end manpower and to shorten the plate making time.
The machine embroidery technology is usually used in large size cloth, the cloth needs to be embroidered with repeated patterns to make the cloth colorful. In order to solve the front end identification of repeated patterns, this study designed an automatic repeated patterns recognition algorithm for machine embroidery image, and proposed a new fitness function for genetic algorithm to create a robust automatic repeated patterns identification algorithm. In order to perform automatic color separation procedure of front end plate making, the unsupervised clustering method was used for color separation of machine embroidery image plate making. Moreover, the cluster validity was used for analyzing K-means, K-medoid, fuzzy C-mean, Gustafson-kessel (G-K) clustering algorithm. The clustering algorithm and validity indicator suitable for machine embroidery were discussed for color separation and identification of number of colors.
This study expected to create an automated technique of machine embroidery image plate making, and proposed an automatic repeated patterns identification and color separation system for machine embroidery images. First, the noise was filtered by using a color hybrid median filter, the image was smoothed by color bilateral filter and the boundary was reserved. The scale is reduced and the low frequency images were reserved by using two-dimensional discrete wavelet transform technology. The images of repeated patterns were identified automatically by using the new fitness function of this paper in genetic algorithm and the distortionless images were restored in the concept of image pyramid. The images of repeated patterns were verified by using frequency domain similarity template matching. The number of machine embroidery image colors and regions corresponding to colors were determined by using unsupervised G-K algorithm and segmentation indicator. Finally, the color varieties were specified by halftoning.
The experimental result showed that the system can provide automatic and accurate identification and segmentation of images of repeated patterns. This study discussed the optimal calculation of unsupervised clustering algorithm in machine embroidery color separation and automatic precise color separation, as well as quick plate making in machine embroidery.

[1] Davis LS, Rosenfeld A. Noise cleaning by iterated local averaging. IEEE Trans Syst Man Cybern 1978; 7: 705–710.
[2] Lee JS. Digital image smoothing and the sigma filter. In Proceedings of Computer Vision 1983; 24: 255–269.
[3] Kenny C, Deng Y, Manjunath BS, Hewer G. Peer group image enhancement. IEEE Trans on Image Processing 2001; 10(2) : 326–334.
[4] Tukey JW. Exploratory data analysis. 1th ed. TX: Addison Wesley, 1977.
[5] Kuo CFJ, Kao CY, Chiu CH. Integrating a genetic algorithm and a self-organizing map network for an automatically separating color printed fabric system. Textile Res J 2009; 79(13): 1235–1244.
[6] Kuo CFJ, Hsu CTM, Shih CY. Automatic pattern recognition and color separation of embroidery fabrics. Textile Res J 2011; 81(8): 1–13.
[7] Haavisto P, Heinonen P, Neuvo Y. Vector FIR-median hybrid filters for multispectral signals. Electronics Letters 1988; 24(1): 7.
[8] Wong NF. Complete works of chinese art. 6th ed. Beijing: Cultural Relics Press, 1985, p.2.
[9] Li ZX. The dawn of China-book of history. 1th ed. Taipei: Reading Times, 1998, p.147.
[10] Nian BH. One hundred kinds of embroidery stitch-history and demonstration. 1th ed. Taipei: Hsiung-Shih Art Book, 2003, pp.140-143.
[11] Tomasi C, Manduchi R. Bilateral filtering for gray and color images. Sixth International Conference on Computer Vision 1998; 4(7): 839–846.
[12] Pan R, Gao W, Liu J, et al. Automatic detection of structure parameters of yarn-dyed fabric. Textile Res J 2010; 80(17) : 1819–1832.
[13] Schalkoff RJ. Digital image processing and computer vision. 1th ed. NY: John Wiley & Sons Inc, 1989, pp.279–286.
[14] Mattoccia S, Tombari F, Di Stefano L. Fast full-search equivalent template matching by enhanced bounded correlation. IEEE Trans on Image Processing 2008; 17(4): 528-538.
[15] Kuo CFJ, Shih CY. Printed fabric computerized automatic color separating system. Textile Res J 2010; 81(7): 706–713.
[16] Yang MS. Department of Mathematics, On Cluster Validity for Fuzzy Clustering. Masters thesis, Chung Yuan Christian University, 2000.
[17] Xu B, Lin S. Automatic color identification in printed fabric images by a fuzzy-neural network. AATCC Review 2002; 2(9): 42–45.
[18] Kuo CFJ, Su TL, Huang YJ. Computerized color separation system for printed fabrics by using backward-propagation neural network. Fibers and Polymers 2007; 8(5): 529–536.
[19] Pan R, Gao W, Liu J, et al. Automatic detection of the layout of color yarns for yarn-dyed f`abric via a fcm algorithm. Textile Res J 2010; 80(12): 1222–1231.
[20] Kuo CFJ, Kao CY, Chiu CH. Integrating a genetic algorithm and a self-organizing map network for an automatically separating color printed fabric system. Textile Res J 2009; 79(13): 1235–1244.
[21] Gonzalez RC, Woods RE digital image processing. 4th ed. NY: Prentice Hall, 2002, pp.282–348.
[22] Plataniotis KN, Venetsanopoulos AN. Color image processing and applications. 1th ed. NY: Springer Verlag, 2000, pp.9–13.
[23] Nixon MS, Aguado AS. Feature extraction and image processing. 2th ed. NY: Academic Press, 2008, pp.89–95.
[24] Hung PC. Information Science and Engineering, Human Detection in Nighttime Near-InfraRed Video. Master thesis, National Chiao Tung University, 2008.
[25] Ko SJ, Lee YH. Center weighted median filters and their applications to image enhancement. IEEE Trans on Circuits and Systems 1991; 38(9): 984–993.
[26] Chen T, Ma KK, Chen LH. Tri-state median filter for image denoising. IEEE T Image Process 1999; 8(12): 1834–1838.
[27] Meerwald P, Uhl A. A Survey of Wavelet-Domain Watermarking Algorithms. Proceedings of SPIE 2001; 4314: 505–516.
[28] Nievergelt Y. Wavelets made easy. 1th ed. NY: Springer Verlag, 1999.
[29] Gonzalez, Woods. Digital image processing using matlab gonzalez woods eddins. 3th ed. NY: Prentice Hall, 2009.
[30] Gu ZP. Electrical Engineering, Studies on The Laser Marking System by Image Processing. Master thesis, National Changhua University of Education, 2006.
[31] Baqai FA, Lee JH, Agar AU, et al. Digital color halftoning. IEEE Signal Processing Society 2005; 22(1): 87–96.
[32] Floyd R, Steinberg L. An adaptive algorithm for spatial grey scale. Proc. Soc. Image Display 1976; 17: 75–77.
[33] Juang JS. Electronic Engineering, Error-Diffusion Kernel Estimation and Continuous-Tone Image Restoration. Master thesis, National United University, 2006.
[34] Swain MJ, Ballard DH. Color Indexing. International Journal of Computer Vision 1991; 7(1): 11–32.
[35] Shih CY. Polymer Engineering, The Automatic Image Analysis System for Printed Fabrics. PhD thesis, National Taiwan University of Science and Technology, 2006.
[36] Trappe W, Washington LC. Introduction to cryptography with: coding theory. 4th ed. NY: Prentice Hall, 2006, pp.328–332.
[37] Holland J. Adaptation in natural and artificial systems. 1th ed. MA: MIT Press Cambridge, 1992.
[38] Tsutsui S, Ghosh A. Genetic algorithms with a robust solution searching scheme. IEEE Trans Evolutionary Computation 1997; 1(3): 201–208.
[39] Zhang Lei. Signal and Information Processing, Analysis of Feature of Target Based On Double Infrared Bands Image and Research of Method of Fusion. Master thesis, North University of China, 2007.
[40] Lee CH, Chen LH. A fast search algorithm for vector quantization using mean pyramids of codewords. IEEE Trans on Cmmuniations 1995; 43(234) : 1697–1702.
[41] Jain AK, Dubes RC. Algorithms for clustering data. 1th ed. NJ: Prentice Hall, 1988.
[42] MacQueen JB. Some methods for classification and analysis of multivariate observations. Proceedings of 5th Berkeley Symposium on Mathematical Statistics and Probability 1967; 31: 281–297.
[43] Basumallick S, Wong JSK. Design and implementation of a distributed database system. Journal of System Software 1996; 34(4): 21–29.
[44] Bezdek JC. Pattern recognition with fuzzy objective function algorithm. 1th ed. MA: Kluwer Academic, 1981.
[45] Krishnapuram R, Kim J. A note on the gustafson-kessel and adaptive fuzzy clustering algorithms. IEEE Trans on Fuzzy Systems 1999; 7(4): 453–461.
[46] Lin SY. Multimedia Engineering, The Study of Principal Curve Detection Using Cluster Ensembles. Master thesis, National Chiao Tung University, 2009.
[47] Kohonen T. Self-organized formation of topologically correct feature maps. Biological Cybernetics 1982; 43: 59–69.
[48] Bezdek JC. Pattern recognition with fuzzy objective function algorithms. 1th ed. MA: Springer, 1981.
[49] Bensaid AM, Hall LO, Bezdek JC, et al. Validity-guided (re)clustering with applications to imige segmentation. IEEE Trans on Fuzzy Systems 1996; 4: 112–123.
[50] Xie XL, Beni G. A validity measure for fuzzy clustering. IEEE Trans on Pattern Analysis and Machine Intelligence 1991; 13(8): 841–847.
[51] Dunn JC. Well seperated clusters and optimal fuzzy partitions. Cybernetics and Systems 1974; 4(1): 95–104.
[52] Rao RM, Bopardikar AS. Wavelet transform: introduction to theory and applications. J Electron Imaging 1998; 8(4): 478.
[53] Bayazit M, Ahsoy H. Using wavelets for data generation. J Applied Statistics 2001; 28(2): 157–166.
[54] Burrus CS, Gopinath RA, Guo H. Introduction to wavelets and wavelet transforms a primer. 1th ed. NJ: Prentice Hall, 1998.
[55] Zahid N, Limouri M, Essaid A. A new cluster-validity for fuzzy clustering. Pattern Recognition 1999; 32: 1089–1097.