基于SOM神经网和K-均值算法的图像分割

提出了一种基于SOM神经网络和K-均值的图像分割算法。SOM网络将多维数据映射到低维规则网格中,可以有效地用于大型数据的挖掘;而K-均值是一种动态聚类算法,适用于中小型数据的聚类。文中算法利用SOM网络将具有相似特征的象素S点映射到一个2-D神经网上,再根据神经元间的相似性,利用K-均值算法将神经元聚类。文中将该算法用于彩色图像的分割,并给出了经SOM神经网初聚类后,不同K值下神经元聚类对图像分割的结果及与单纯K-均值分割图像进行对比。

An Image Segmentation Method Based on SOM Network and K-means Algorithm

In the paper,we have proposed a image segmentation method based on SOM network and K-means algorithm.SOM network can project multi-dimensional data on a low-dimensional regular grid that can be effectively utilized to explore properties of the large data.While K-means is a dynamic clustering algorithm for the small data.We have projected the pixels that have similar feature on 2-D neural network using SOM network.Then according to the similarity of neurons,neurons were clustered by K-means algorithm.We have applied the method to color image,and experimental results demonstrate that the method is a reliable tool for image segmentation.