传统服饰云肩实物图像主色的智能检测

纺织学报 ›› 2017, Vol. 38 ›› Issue (11): 110-115.

传统服饰云肩实物图像主色的智能检测

收稿日期:2016-09-01 修回日期:2017-07-24 出版日期:2017-11-15 发布日期:2017-11-15

Intelligent inspection of dominant colors for Chinese traditional folk Yunjian

Received:2016-09-01 Revised:2017-07-24 Online:2017-11-15 Published:2017-11-15

摘要/Abstract

摘要：

为解决获取传统服饰实物图像色彩耗时且缺乏准确性问题，以传统服饰的云肩为例，借助均值漂移（Mean-shift）聚类法，提出了一种检测传统服饰实物图像颜色的方法。运用单反数码相机进行实物图像采集；对所得初始图像的R、G、B3个颜色通道进行去噪处理；再将图像RGB颜色空间的特征向量转换至CIE L*a*b*颜色空间中，利用大津法阈值原理（自适应阈值算法）分割被测图像中云肩实物与背景；最后采用Mean-shift聚类算法，将被测图像的颜色像素分割为若干有效的集群，同时从这些集群中提取云肩主要色彩。实验结果表明，该算法可较为准确地从云肩图像中提取主色，且当Mean-shift聚类算法的带宽被设定为0.05时，分类颜色结果更为准确。

关键词: 云肩, 主色, 均值漂移聚类算法, CIE L*a*b*颜色模型, 颜色检测

Abstract:

In order to solve the problems of time-consuming and poor accuracy during the process of achievement colors from the images of tarditional costumes, an effective color clustering method was developed for Chinese traditional costumes Yunjian image Mean-shift clustering algorithm. A digital Single Lens Reflex camera is used to capture the costumes images. The sub-images in the three color channels are filtered by median filter separately. Thirdly, the filtered images are segmented based on the background color in the Lab color space, and the object costumes is separated from the background. Fourthly, the pixels of the costume image are classified into several clusters by Mean shift clustering algorithm, and the dominate colors are extracted from the classification results. The experimental results demonstrate that the proposed method can extract the dominant colors from costumes images with great accuracy when the bandwidth of Mean shift clustering algorithm is set as 0.05.

Key words: Yunjian, dominant color, Mean-shift clustering algorithm, CIE L*a*b* color model, color inspection

邢乐张杰梁惠娥李忠健刘姣姣. 传统服饰云肩实物图像主色的智能检测[J]. 纺织学报, 2017, 38(11): 110-115.

参考文献

[1]薛再年，王闪闪，崔荣荣，鲁南民间云肩及其制作技艺探究[J]，纺织学报，2013,34(3):109-115. XUE Zainian, WANG Shanshan, CUI Rongrong, Folk Yunjian in South Shandong and its workmanship, [J], Journal of Textile Research, 2013,34(3):109-115. [2]邢乐，梁惠娥，刘水，民间服饰云肩中人物纹样的语义考析[J]，艺术设计研究，2015,24(01): 54-60. XING Le, LIANG Huie, LIU Shui, Figure patterns of folk costume-Yunjian and its significance[J], Art & Design Research, 2015,24(01): 54-60. [3]梁惠娥，邢乐，中国最美云肩情思回味之文化[M].郑州:河南文艺出版社，2013：143-144. LIANG Huie, XING Le, Chinese Traditional Folk Dress Adornments-Yunjian Culture Research[M]. Zhengzhou: Henan Literature & Art Publishing House, 2013：143-144. [4]ZHANG, J., XIN, B.J, SHEN, C.C., Fang, H. and Cao, Y.J., Novel Colour Clustering Method for Interlaced Multi-colored Dyed Yarn Woven Fabrics[J]. FIBRES & TEXTILES in Eastern Europe. 2015,23(03): 107-114. [5]ZHENG, D. ,A new method for automatic separation of fabric color[J]. Textile Research Journal, 2015,85, (14): 1520-1538. [6]KUO, C.F.J., JIAN, B.L., Tung, C.P. and PENG, K.C., Automatic machine embroidery image color analysis system, part II: Application of the genetic algorithm in search of a repetitive pattern image[J]. Textile Research Journal. 2012,82(11): 1099-1106. [7] BROWNRIGG D. The weighted median filter[J], Communication Association Computer Machine,1984,27(8):807-818. [8]赵高长，张磊，武风波，改进的中值滤波算法在图像去噪中的应用[J].应用光学，2011,32(4)：678-682. ZHAO Gaochang, ZHANG Lei, WU Feng-bo, Application of improved median filtering algorithm to image de-noising[J]. Journal of Applied Optics, 2011,32(4)：678-682. [9]刘国宏，郭文明，改进的中值滤波去噪算法应用分析[J]，计算机工程与应用，2010,46(10):187-189. LIU Guohong, GUO Wenming, Application of improved arithmetic of median filtering denoising[J].Computer Engineering and Applications，2010，46（10）：187-189. [10]SCHANDA. J. CIE colorimetry. In: Schanda J., Colorimetry: Understanding the CIE system[M]. Hoboken: Wiley-Interscience; 2007: 61. [11]卢雨正，高卫东，基于图像分割的拼色纺织品分色算法[J], 纺织学报，2012,33(09)：55-60. LU Yuzheng, GAO Weidong, Color separation algorithm for mixed dyed textiles based on image segmentation[J]，Journal of Textile Research, 2012,33(09)：55-60. [12]PAN, R.R., GAO, W.D., LIU, J.H., Automatic Detection of the Layout of Color Yarns for Yarn-Dyed Fabric via a FCM Algorithm[J]. Textile Research Journal. 2010,80(12): 1222-1231. [13]TSAI, D.M., LUO, J.Y., Mean Shift-Based Defect Detection in Multicrystalline Solar Wafer Surfaces[J]. Browse Journals & Magazines, Industrial Informatics IEEE Transactions on, 2011,7(01): 125-135. [14]COMANICIU. D., MEER, P., Mean shift: A robust approach toward feature space analysis[J].IEEE Trans Pattern Anal Mach Intell; 2002,24: 603–619. [15] KIRKLAND. E. J., Bilinear Interpolation, Advanced Computing in Electron Microscopy[M], 2010: 261-263. [16] HAN, Y., FENG, X., BACIU, G., Variational and PCA based natural image segmentation. Pattern Recognit[J]. 2013,46:1971–1984. [17] HEGGIE, D., WARDMAN R. and LUO, M., A comparison of the colour differences computed using the CIE94, CMC(l:c) and BFD(l:c) formulae[J]. Journal of the Society of Dyers and Colourists, 2008,112(10): 264-269. [18] ASPLAND, J.R., DUNLAP, K.L. and DAN R.T., Improved Methods for Colour Inventory Management in the Apparel Industry[J]. International Journal of Clothing Science and Technology, 1989,4(2/3), : 66-70.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed