结合模糊聚类与支持向量机的图像分割（英文）

提出一种新的混合的图像分割方法,利用模糊C均值聚类与支持向量机两种方法相结合。此方法首先将图像的空间分布信息作为支持向量机的特征分量,再用模糊C均值聚类获得的分类结果作为支持向量机所需的初始训练样本,并对图像的所有像素点进行分类,同一类中的像素点形成一个分割区域,以此获得图像分割。实验表明,此将模糊C均值与支持向量机结合的新方法获得的图像分割效果较好,在一定程度上解决了支持向量机特征维数过大所导致的维数灾难问题。

Image Segmentation Combined FCM and SVM

Propose a new hybrid methods for image segmentation combined Support Vector Machine （ SVM） with C mean fuzzy cluste-ring. This method takes the spatial distributed information as component characteristics of the SVM,and the classification results from fuzzy clustering as the initial training samples of the SVM. Then the pixels of the image are classified by SVM and the pixels in the same class form a segmental region to obtain image segmentation. The experimental results show that the new methods combing fuzzy cluste-ring and SVM can get better results and to a certain extent solve the dimension disaster problem caused by large dimension of SVM.