中文摘要：

本文提出了一种适用于彩色图像分割的遗传模糊C均值聚类（GAFCM）算法。该算法使用Ohta等人提出的彩色特征集中的第一个分量作为图像像素的一维特征向量，并利用由像素空间到特征空间的映射来改进目标函数，从而大大降低了运算量；使用对特征空间结构没有特殊要求的特征距离代替欧氏距离，从而克服了特征空间结构对聚类结果的影响；使用引入FCM优化的遗传算法来搜索最优解，从而提高了搜索速度。实验表明，该算法不但能很好地分割彩色图像，而且具有运算量小、收敛速度快的优点。

英文摘要：

An improved Genetic Fuzzy C-means Clustering （GAFCM） algorithm is proposed for color image segmentation. The first component of color feature set discovered by Ohta is chosen as the one-dimensional eigenvector. In order to reduce the computational complexity, the mapping from pixel space to eigenvector space is used for modifying the object function. Feature distance which is applied to any structure of eigenvector space is applied instead of Euclidian distance to overcome the influence caused by structure of eigenvector space. FCM optimization is introduced to genetic algorithm to accelerate the searching speed. Experiments show that the algorithm has better effect and lower computational complexity on color image segmentation.