Multi-feature fusion technique is used to recognize and classify the image target in this paper. We extract the fractal feature and gray entropy from the image，then use Dempster-Shafer's Evidential Reasoning to fuse the information at the report level. Some decision strategies are used to recognize and classify the image. In experiment，we compare the results obtained from Multi-feature fusion with those obtained from single feature. The final results indicate that the Multi-feature fusion method is stable，reliable，and can efficiently improve the accuracy and the ability of fault tolerance.