Remote sensing image segmentation based on ant colony optimized fuzzy C-means clustering

Middle spatial resolution multi-spectral remote sen sing image is a kind of color image with low contra st, fuzzy boundaries and informative features. In view of the se features, the fuzzy C-means clustering algorithm is an ideal choice for image segmentation. However, fuzzy C-mea ns clustering algorithm requires a pre-specified nu mber of clusters and costs large computation time, which is easy to fall into local optimal solution. In order to overcome these shortcomings, ant colony algorithm is employe d to optimize fuzzy C-means algorithm in remote sen sing image segmentation. First, the centers and number of clus ters is determined by ant colony optimization algor ithm. Then the initialization fuzzy C-means algorithm is used for remote sensing image classification. Experimental r esults show that the ant colony optimization is an effective me thod to solve the problem of fuzzy C-means algorith m in remote sensing image segmentation and the visual interpret ation of segmentation is much improved by proposed ant colony optimized C-means clustering.