Study of wool-spun color matching system based on artificial nerve network
Dyeing color matching is a very important process in wool-spun dyeing industry. The main theory that present computer matching color system used is based on Kubelka-Munk Theory. But the theory has preconditions and the algorithm is complex, so color matching error is comparatively great. A new method based on BP nerve network in digital computer color matching system is proposed in this article. The algorithm that used in nerve network is mainly about BP algorithm that is used to achieve the nonlinear mapping between the color characteristics and the mixing proportions of different dyes. In experiments, obvious astringency can be reached, but there are still some deviations in the extensive ability. But through experiment, it also can be found that the method based on the wavelet analysis in HSV color space is the most effective method. To wool-spun color matching intellectualization, this method owns definite reference value.
computer color matching wavelet transform nerve network color characteristics color space
Xie Chunping Feng Jie Wang Huifeng Su Xuzhon Yang Lili Huang Jingquan
Textiles and Clothing School, Southern Yangtze University, Wuxi, Jiangsu 214122, China
国际会议
2006中国国际毛纺织会议暨IWTO羊毛论坛(2006 China International Wool Textile Conference & IWTO Wool Forum)
西安
英文
203-206
2006-11-19（万方平台首次上网日期，不代表论文的发表时间）