摘要
针对陶瓷制品陶瓷墙地砖颜色特征提取的关键问题,提出了灰色聚类的改进算法。并同其他已有的模糊C均值算法(FCM)、MOP_FCM、鲁棒性的FCM、改进的鲁棒性模糊C均值算法,传统的灰色聚类算法,通过实验进行算法性能分析,颜色特征提取效果比较。实验结果表明,本文提出的改进算法,算法性能更高,颜色特征提取效果更明显,更有利于进行色差比较。
For the key issue of color feature extraction for ceramic wall and floor tile, a grey clustering algorithm was improved. The other existing algorithm was used. For example, fuzzy C mean algorithm( FCM), MOP_FCM, the robustness of FCM, improved the robustness of fuzzy C mean algorithm, the traditional grey clustering algorithm. Through the experiment analysis of algorithm performance, the effect color feature extraction was comparied. The experimental results show that the proposed algorithm has high performance, and which extraction of color feature effect was more obvious, and which the color comparison was more conducive.
出处
《中国陶瓷》
CAS
CSCD
北大核心
2015年第5期42-45,共4页
China Ceramics
基金
唐山市科技计划项目(13110210b)
河北省唐山市科技计划项目(13130250z)
关键词
模糊聚类
灰色聚类
颜色特征
色差
Fuzzy Clustering
Grey clustering
Color Feature
Color Difference
