Image Segmentation Using Thresholding and Swarm Intelligence

Image segmentation is a significant technology for image process. Many segmentation methods have been brought forward to deal with image segmentation, among these methods thresholding is the simple and important one. To overcome shortcoming without using space information many thresholding methods based on 2-D histogram are often used in practical work. These methods segment images by using the gray value of the pixel and the local average gray value of it, and thus provide better results than the methods based on 1-D histogram. However, its timeconsuming computation is often an obstacle in real time application systems. In this paper, fast image segmentation methods based on swarm intelligence and 2-D Fisher criteria thresholding are presented. The proposed approaches have been implemented and tested on several real images. Experiments results indicate that the proposed methods provides improved search performance which are efficient methods to help select optimum 2D thresholds with much less computation cost and suitable for real time applications.