目前的織物瑕疵檢測倚賴人工檢查，透過專業訓練的檢查員以人眼進行，但人眼檢測的方式耗時、疲勞，經常未檢測到細小的缺陷，大約只有70%的瑕疵可以被檢測員所檢測，而且人為主觀的瑕疵判斷，使得檢測標準不夠客觀，因此瑕疵檢測、分類無法完善，故本研究自行設計軟硬體設備，並搭配所開發的影像處理流程，建立梭織物之瑕疵檢測系統，期能克服產業存在的問題。
本研究提出的檢測系統，可針對梭織物常見的汙漬(Stain)、缺經(Broken end)、缺緯(Broken weft)、破洞(Hole)、粒結織物(Nep)、雙緯(Double pick)、緯縮(Kinky weft)、浮紗(Float)等八類型瑕疵進行辨識與分類。首先以CMOS工業相機在捲布機上端以20m/min進行 像素的灰階影像擷取，接著透過四個階段影像處理流程來達到瑕疵檢測與瑕疵分類。第一階段為影像前處理，利用高斯濾波器濾除影像雜訊，再透過演算法進行亮度校正降低光暈現象造成的亮度不均問題，本研究使用改良型遮罩差分演算法使校正後原始影像的標準差由12.072降至2.897，有效的解決光源不均的問題，接著利用平均值濾波器濾除背景紋理，經過平均值校正，進行直方圖的平移，使得本系統對於梭織物的紋理與顏色變化，具有強健性，再透過無瑕疵影像的平均值與標準差來建立二值化分割閥值。第二階段為利用自適應二值化將背景與瑕疵分離並將雜訊濾除，第三階段先進行形態學處理再對瑕疵進行輪廓圈選，針對輪廓圈選的範圍分別計算每個區塊的瑕疵面積、瑕疵長寬比、瑕疵平均灰階值以及瑕疵方向性四種特徵值，並對2246張影像進行影像瑕疵辨識，實驗結果顯示檢測成功率(Detection Success Rate)為96.44%、檢測率(Detection Rate)為96.35%以及誤判率(False Alarm Rate)為3.21%。第四階段進行瑕疵分類，利用支持向量機(SVM)進行分類，使用瑕疵影像230個作為訓練樣本，206個作為測試樣本，實驗結果顯示瑕疵總體辨識率為96.60 %，證實本研究所設計之軟硬體設備能有效的對梭織物進行瑕疵檢測與分類。

The present fabric defect detection depends on manual examination, implemented by professional inspectors with naked eye. However, the visual detection method results in time consumption, fatigue and sub-jective defect judgment. The inspection standard is not objective enough, and tiny defects are often missed. Only about 70% of defects can be detected by inspectors. The defect detection and classification cannot be perfect. Therefore, this study designs the software and hardware equipment, combined with the developed image processing procedure to build a defect detection system for woven fabric, hoping to overcome the problems in the industry.
The detection system proposed in this study can recognize and classify eight common defects in woven fabric, including stain, broken end, broken weft, hole, nep, double pick, kinky weft and float. First, a pixels gray image is captured by CMOS industrial camera above the batcher at 20m/min. Then the defect detection and defect classification are implemented by 4-stage image processing procedure. Stage 1 is image preprocessing, the image noise is filtered by Gaussian filter. The light source is corrected to reduce the uneven brightness re-sulted from halo formation. This study uses the improved mask dif-ference algorithm to reduce the standard deviation of the corrected original image from 12.072 to 2.897. The nonuniform light source is solved effectively. Afterwards, the background texture is filtered by averaging filter, and the mean value is corrected for histogram shifting, so that this system is robust to the texture and color changes of woven fabric. The binary segmentation threshold is established by the mean value and standard deviation of image without defect. Stage 2 uses adaptive binarization to separate the background and defects and filter the noise. In Stage 3, the morphological processing is implemented before the defect contour is circled, four eigenvalues of each block, in-cluding defect area, aspect ratio of defect, average gray level of defect and the defect orientation, are calculated according to the range of contour. The image defect recognition is implemented for 2,246 images. The experimental results show that the detection success rate is 96.44%, and the false alarm rate is 3.21%. The defect classification is implemented in Stage 4. The Support Vector Machine is used for clas-sification, 230 defect images are used as training samples, and 206 are used as test samples. The experimental results show that the overall defect recognition rate is 96.60 %, providing that the software and hardware equipment designed in this study can implement defect de-tection and classification for woven fabric effectively.

[1] Arivazhagan, S., Ganesan, L., Bama, S. (2006). Fault segmenta-tion in fabric images using Gabor wavelet transform. Machine Vision and Applications, 16(6), 356-363.
[2] Selvi, S. S. T., Nasira, G. M. (2017). An Effective Automatic Fabric Defect Detection System using Digital Image Processing. J. Environ. Nanotechnol, 6(1), 79-85.
[3] 王雨讓、王冠翔、安大中、巫佳宜、李信宏、林仁豪、林俊宏、俞玲華、梁瑋苓、陳宏一、彭振宇、黃致成、黃稚評、楊宜蓁、葉乙昌、閩潔、劉心蘋、駱春梅、蘇英傑撰稿(2017)，2017紡織產業年鑑，財團法人紡織產業綜合研究所。
[4] Sabeenian,R.S., Paramasivam, M.E., Dinesh, P. M. (2012). Computer vision based defect detection and identification in handloom silk fabrics. Computer, 42(17).
[5] Dorrity, J.L., Vachtsevanos, G. (1996, May). On-line defect de-tection for weaving systems. In Textile, Fiber and Film Industry Technical Conference, 1996 IEEE Annual, 1-6.
[6] Jing, J., Zhang, H., Li, P. (2011). Improved Gabor filters for textile defect detection. Procedia Engineering, 15, 5010-5014.
[7] Handbook, T. (2000). Hong Kong Productivity Council. The Hong Kong Cotton Spinners Association.
[8] Dhande, B., Dhotre, M. R. (2014). A Review of Fabric Defect Detection Techniques.International Journal of Computer Science Trends and Technology (IJCST) 2 (2), 41-44.
[9] Mishra, D.(2015). A survey-defect detection and classification for fabric texture defects in textile industry. International Journal of Computer Science and Information Security, 13(5), 48.
[10] Ngan, H. Y., Pang, G. K., Yung, N. H. (2008). Motif-based defect detection for patterned fabric. Pattern Recognition, 41(6), 1878-1894.
[11] Chan, C. H.,Pang, G. K. (2000). Fabric defect detection by Fourier analysis. IEEE transactions on Industry Applications, 36(5), 1267-1276.
[12] Mishra, R., Shukla, D. (2014). A survey on various defect detec-tion. Int. J. Eng. Trends Technol, 10(13), 642-648.
[13] Li, L., Huang, G. Q., Newman, S. T. (2007). Robotics and Com-puter-Integrated Manufacturing. In 16th International Conference on Flexible Automation and Intelligent Manufacturing, 23(6), 650-658.
[14] Eldessouki, M. (2018). Computer vision and its application in detecting fabric defects. In Applications of Computer Vision in Fashion and Textiles , 61-101.
[15] Hanbay, K., Talu, M. F. (2014). Kumaş hatalarının online/offline tespit sistemleri ve yöntemleri. Sakarya Üniversitesi Fen Bilimleri Enstitüsü Dergisi, 18(1), 49-69.
[16] Monaco, J. P., Madabhushi, A. (2012). Class-specific weighting for Markov random field estimation: Application to medical image segmentation. Medical image analysis, 16(8), 1477-1489.
[17] Cohen, F. S., Fan, Z., Attali, S. (1991). Automated inspection of textile fabrics using textural models. IEEE Transactions on Pat-tern Analysis and Machine Intelligence, 13(8), 803-808.
[18] Mahajan, P. M., Kolhe, S. R., Patil, P. M. (2009). A review of au-tomatic fabric defect detection techniques. Advances in Compu-tational Research, 1(2), 18-29.
[19] Kumar, A. (2008). Computer-vision-based fabric defect detection: A survey. IEEE transactions on industrial electronics, 55(1), 348-363.
[20] Mak, K. L., Peng, P., Yiu, K. F. C. (2009). Fabric defect detection using morphological filters. Image and Vision Computing, 27(10), 1585-1592.
[21] Hanbay, K., Talu, M. F., Özgüven, Ö. F. (2016). Fabric defect detection systems and methods—A systematic literature re-view. Optik-International Journal for Light and Electron Op-tics, 127(24), 11960-11973.
[22] Stojanovic, R., Mitropulos, P., Koulamas, C., Karayiannis, Y., Koubias, S., Papadopoulos, G. (2001). Real-time vision-based system for textile fabric inspection. Real-Time Imaging, 7(6), 507-518.
[23] Cho, C. S., Chung, B. M., Park, M. J. (2005). Development of real-time vision-based fabric inspection system. IEEE Transac-tions on Industrial Electronics, 52(4), 1073-1079.
[24] Çelik, H. İ., Dülger, L. C., & Topalbekiroğlu, M. (2014). Fabric defect detection using linear filtering and morphological opera-tions. Indian Journal of Fibre & Textile Research (IJFTR), 39(3), 254-259.
[25] Raheja, J. L., Kumar, S., Chaudhary, A. (2013). Fabric defect de-tection based on GLCM and Gabor filter: A comparison. Optik-International Journal for Light and Electron Optics, 124(23), 6469-6474.
[26] Xiaowei, Z., Xiujuan, F. (2016). Fabric Defect Detection based on GLCM Approach. Energy (ENG), 2, 1.
[27] Li, D. R., Wang, M., Pan, J. (2006). Auto-dodging processing and its application for optical RS images. Geomatics and Information Science of Wuhan University, 31(9), 753-756.
[28] Sun, M. (2009). Research on Key Technology of Automatical and Fast DOM Generation. Wuhan: Wuhan University.
[29] Zhang, Z., Zou, S. (2011, November). An improved algorithm of mask image dodging for aerial image. In MIPPR 2011: Remote Sensing Image Processing, Geographic Information Systems, and Other Applications (Vol. 8006, p. 80060S). International Society for Optics and Photonics.
[30] Yan, H., Zuo, Y., Chen, Y., & Chen, Y. (2016). Evaluation of the morphology structure of Meibomian Glands based on Mask dodging method. In Proc. of SPIE ,(10157), 101573H-1.
[31] Çelik, H. İ., Dülger, L. C., Topalbekiroğlu, M. (2014). Develop-ment of a machine vision system: real-time fabric defect detection and classification with neural networks. The Journal of The Tex-tile Institute, 105(6), 575-585.
[32] Basu, A., Chandra, J. K., Banerjee, P. K., Bhattacharyya, S., Datta, A. K. (2012, July). Sub image based eigen fabrics method using multi-class SVM classifier for the detection and classifica-tion of defects in woven fabric. In Computing Communication Networking Technologies (ICCCNT), 2012 Third International Conference , 1-6.
[33] Abdellah, H., Ahmed, R., Slimane, O. (2014). Defect detection and identification in textile fabric by SVM method. IOSR Journal of Engineering (IOSRJEN), 4(12), 69-77.
[34] Al-Anazi, A., Gates, I. D. (2010). A support vector machine algo-rithm to classify lithofacies and model permeability in heteroge-neous reservoirs. Engineering Geology, 114(3), 267-277.
[35] Hu, Y. J., Ku, T. H., Jan, R. H., Wang, K., Tseng, Y. C., Yang, S. F. (2012). Decision tree-based learning to predict patient controlled analgesia consumption and readjustment. BMC Medical Infor-matics and Decision Making, 12(1), 131.
[36] 吳欣儒(2011)，資料探勘技術於病人疼痛自控裝置之應用與分析，國立交通大學資訊科學與工程研究所碩士學位論文。
[37] 鐘國亮(2008)，影像處理與電腦視覺(第四版)，台灣台北:東華書局。
[38] Gonzalez, R. C., Woods, R. E. (2007). Digital Image Processing, 3rd edition, Pearson Education, Inc.
[39] Shewhart, W. A.(1931).Economic control of quality manufactured product.New York: Van Nostrand.
[40] Cortes, C., Vapnik, V. (1995). Support-vector networks. Machine learning, 20(3), 273-297.
[41] Vapnik, V. N.(1998).Statistical learning theory., New York: John Wiley Sons, Inc..
[42] Cover, T. M. (1965). Geometrical and statistical properties of systems of linear inequalities with applications in pattern recog-nition. IEEE transactions on electronic computers, (3), 326-334.
[43] Hsu, C. W., Lin, C. J. (2002). A comparison of methods for mul-ticlass support vector machines. IEEE transactions on Neural Networks, 13(2), 415-425.
[44] Ngan, H. Y., Pang, G. K., Yung, N. H. (2011). Automated fabric defect detection—a review. Image and Vision Computing, 29(7), 442-458.