中文摘要：

针对模糊C均值聚类算法存在仅考虑以类内距离作为算法测度的不足,通过融入聚类中心之间的类间距离,提出一种将类内和类间距离相结合的模糊C-均值聚类算法并将其应用于图像分割。在目标函数中将类内距离与类间距离之差作为样本聚类依据,使其考虑到类内紧密度与类间离散度,通过调节有关参数使类内紧密度和类间离散度达到最优值,提高图像分割的准确性和鲁棒性。大量人工合成图像和实际遥感图像分割测试结果表明,改进的类内类间聚类算法是有效的,尤其是对噪声较大的图像进行分割时,其效果明显优于其它模糊聚类算法分割效果。

英文摘要：

Aiming at the defect that the fuzzy C-means clustering algorithm only considered intra-class as the algorithm measure,a kind of fuzzy C-means clustering algorithm was proposed and applied to image segmentation by combining the inter-class distance between clustering centers.The difference of intra-class distance and inter-class distance within the object function was taken as the clustering measure,which not only considered intra-class compactness,but also took the inter-class separation into account.The compactness of clusters and separation within them achieved optimal values by adjusting the relevant parameters,and the accuracy and robustness of image segmentation were improved.Results of segmentation tests of a large number of synthetic images and real remote sensing images show that the proposed algorithm is effective,especially for noisy images,the effect is better than other fuzzy clustering algorithms.