Aiming at the fusion of infrared intensity-polarization image, an intelligent fusion method based on spatially weighted averaging method optimized by fireworks algorithm is proposed. Based on the optimization model, the boundary conditions of fireworks algorithm are determined. The fitness function based on comprehensive relative-entropy is established by introducing the weight of relative-entropy. Finally, the fusion experiments on three groups of infrared image “ground”, “truck” and “car” are carried out with six typical traditional fusion methods, and the fusion results are evaluated objectively and compared with the visual effects. The experimental results show that the proposed method can effectively achieve the fusion of infrared intensity map and polarization map, and retain the infrared intensity and polarization characteristics. Combining the visual effect and objective evaluation results, the method in this paper is superior to the comparison algorithm in relative-entropy, similarity of summary structure and total mutual information index.