基于模糊能量聚类的变分水平集遥感图像分割算法

Variational Level Set Remote Sensing Image Segmentation Algorithm Based on Fuzzy Energy Clustering

提出一种基于模糊能量聚类的变分水平集遥感图像分割算法,该算法保留了变分水平集能够综合利用区域和边界信息的特点,改善了变分水平集方法对带噪声遥感图像进行分割时存在去噪效果不明显、分割精度不高的问题。在通过变分法得到能量泛函取极小值的水平集函数演化方程的基础上,采用了连续的最优隶属度函数,得到模糊能量聚类的变分水平集。实验仿真及对比结果表明,该算法分割后的图像区域具有明显灰度差和边界区分,去噪效果良好,而且分割精度优于对比算法。

A variational level set remote sensing image segmentation algorithm based on fuzzy energy clustering is proposed. The algorithm preserves the feature of variational level set that can use region and boundary information together, while solving the problems of the method in segmenting remote sensing images with noise that the denoising effect is not obvious and the segmentation accuracy is not high. On the basis of obtaining energy functional minimum level set function evolution equation by the variational method, the optimal membership functions are used continually to get variational level set of fuzzy energy clustering.Experimental simulation and comparison results show that： 1） the image area segmented with this algorithm has obvious gray scale difference and boundary; and 2） the denoising effect is fine and the segmentation accuracy is better than the other algorithms in comparison.