基于二维直方图加权的高斯核FCM图像分割

摘要

This paper proposed an image segmentation method based on the two-dimensional histogram weighting and Gaussian kernel (GK) FCM (2DWGKFCM).Due to noise,the traditional FCM method can't distinguish the target and the background in image segmentation.While the introduction of twodimensional weighting coefficient can adjust the clustering centers,which is proved can improve the segmentation efficiency.Besides,the introduction of GK improves the segmentation quality by improving the partition accuracy where the sample feature space is of linear inseparability.The results showed that the method can not only improve the image segmentation efficiency,but also qualified with better robustness and effectiveness.%文中提出了基于二维直方图加权的高斯核FCM图像分割方法,该方法在传统FCM算法中融入了二维直方图的加权系数及高斯核函数.在图像分割中,由于噪声干扰,传统的FCM算法在许多情况下并不能有效区分目标和背景,而利用二维直方图中像素点与其领域间相关的空间信息,将其作为加权系数用来调整聚类中心,可有效提高分割效率.此外,算法结合高斯核函数用于改善FCM算法在样本特征空间线性不可分时带来的划分不准的缺点,改善图像分割效果.实验结果表明,此算法在提高图像分割效率的基础上,比传统算法具有更好的鲁棒性与有效性.