Considering the disadvantages of the traditional method in the fuze image segmentation, puts forward a method to improve two-dimensional Otsu algorithm which can consider both the interclass variance and the between class variance. Introduce the 2-D Otsu algorithm, analysis and comparison the original segmentation with the adaptive 2-D Otsu algorithm. The experimental results show that the improved algorithm not only guarantees the maximum dispersion of target and background but also make better cohesion between them, that is obviously superior to the traditional method.