基于加权模糊C均值聚类的图像分割算法

Image Segmentation Algorithm Based on Weighted Fuzzy C-means Clustering

摘要:

模糊C均值(FCM)算法用于灰度图像分割是一种非监督模糊聚类后再标定的过程,适合灰度图像中存在着模糊和不确定性的特点。但是这种算法存在着本质上的缺陷,就是仅利用了图像的灰度信息,而没有考虑像素的空间信息,使得其对于实际的含有噪声的图像分割效果不理想。因此,提出了一种新的加权模糊C均值聚类算法,实践证明,该方法可以有效地、实时地把目标从背景中分割出来,并具有较强的鲁棒性。

Abstract:

It is a procedure of the label following an unsupervised fuzzy clustering that fuzzy C-means (FCM) algorithm is applied for gray image segmentation, and it suits for the uncertain and ambiguous characters in gray image. However, there is essential defect in the algorithm, for example, it only uses its gray distribution and no spatial information. The effect of image segmentation is not ideal for practical picture containing the noise. So a new weighed fuzzy C-means clustering method is introduced. The experimental result demonstrates that this scheme can segment effectively and real-timely the target from its background and it has much more robustness.