融合纹理特征的两阶段聚类分割算法
摘 要
提出一种融合纹理特征的两阶段聚类分割算法。首先,选择纹理特征、差分均值和颜色分量这3个特征,组成一个分割所用的特征矢量;然后,使用直方图对特征矢量进行初始聚类中心和类别数的估算;最后,通过模糊C均值算法对特征矢量进行聚类。该算法有效地克服了模糊C均值(FCM)容易陷入局部最优的缺陷,使聚类结果更加精确。实验结果表明该方法比一些现存方法的分割效果要好。
关键词
Texture feature fusion-based two-stage clustering segmentation algorithm
Wang Gaihua1,2, Li Dehua1(1.Institute for pattern recognition& Artificial Intelligence, Huazhong University of Science & Technology, Wuhan 430074, China;2.The College of Physics and Information Science of Xinjiang Normal University, Urumqi 830053, China) Abstract
It proposes texture feature fusion-based two-stage clustering segmentation algorithm. First, we choose texture feature, the average of difference and color component as feature vector for segmentation. Then,at the stage of segmentation, aim to the disadvantages of Fuzzy c-means, it computes the clustering center and the number of clustering center based on histogram. Finally, we use feature vector to cluster through Fuzzy c-means. Compared with some well-known methods, the proposed method has a better segmental result.
Keywords
|