Maximum Entropy Image Segmentation Method Based on Improved Firefly Algorithm

Abstract

An adaptive parametric firefly algorithm is proposed for the premature convergence of the firefly algorithm itself and the late oscillation of the algorithm iteration. In the image threshold segmentation test experiment, the algorithm is used to optimize the maximum entropy image segmentation, and the maximum entropy and the basic firefly improved maximum entropy algorithm are compared. At the same time, two important indicators of image segmentation evaluation index regional consistency are used to evaluate the results. It shows that the experimental results of the algorithm have better intra-regional consistency and noise immunity.