A Lévy Flight Based Firefly Algorithm for Multilevel Thresholding Image Segmentation

Abstract Traditional exhaustive search methods have high time complexity in solving multilevel threshold problems. To address this problem, a Levy flight based firefly algorithm called LFA is proposed. Otsu regard as its objective function. A series of standard test images were used to evaluate the performance algorithm. The Peak signal noise ratio(PSNR) and Structural similarity index (SSIM) are utilized segmented image. experimental results show that better than FA for thresholding image segmentation.