Basic particle swarm optimization(PSO) can not get good optimization performance, because it is easy to get stuck into local optima. Therefore, an algorithm named improved PSO which combines proposed inertia adaptive PSO with partial particles Morlet mutation is proposed. The proposed algorithm and fuzzy entropy are applied to image segmentation, and improved PSO is used to explore fuzzy parameters of maximum fuzzy entropy, which gets the optimum fuzzy parameter combination, then obtains the segmentation threshold. By comparing the proposed algorithm with other two algorithms, the experiment results show that the proposed algorithm has the capability of good segmentation performance and low time cost, which can be use to real time and precision measure coal dust image.