摘要
在加权模糊c-均值(FCM)聚类算法的基础上,对分色算法进行了改进.首先进行色彩空间模型转换,然后对基于样本加权的FCM算法进行改进,对隶属度进行调整,把二维彩色直方图引入加权系数中.对于模糊c-均值算法,当隶属度接近时,分类会变得模糊,而且对于不同的样本矢量,聚类效果有所不同,本算法兼顾到了这两点.该方法已用V isua l C++6.0编程实现,效果比较理想.
An improved color segmentation algorithm is presented based on weighting fuzzy c-means （FCM） clustering algorithm. The color space of the map image is transformed firstly. Then, it is improved based on weighting of swatch FCM algorithm. The subordinate degree is adjusted and the 2-D color histogram is introduced into the weighting coefficient. Segmentation becomes fuzzy when the subordinate degrees become similar and for different sample vectors the clustering effects are different in fuzzy c-means algorithm. The proposed algorithm combines the two points. The algorithm is implemented by programming on computer with Visual C＋＋6.0 and a good result is obtained.
出处
《控制与决策》
EI
CSCD
北大核心
2006年第10期1092-1096,共5页
Control and Decision
基金
国家自然科学基金项目(60274099)
国家863计划项目(2004AA412030)
关键词
地形图分色
色彩空间转换
FCM算法
加权模糊c-均值算法
Topographic map segmentation
Color space transformation
Fuzzy c -means algorithm
Weighting fuzzy c-means algorithm
