In order to meet the requirement of high-precision target recognition for space-based systems, a target recognition method based on multi-band feature fusion template matching is proposed. In this paper, the multi-dimensional feature elements and classification methods of the target are firstly described and analyzed.Feature fusion processing is achieved by combining the features of the target motion, the target spectral features in multi-band and the changes of infrared radiation. Based on the changes of regional features, the dynamic wrapping template matching algorithm is adopted to complete the target identification. Finally, the parameters are trained and adjusted by combining the infrared radiation intensity sequence data of the cooperative target. The analysis results show that the proposed recognition method can realize the target model recognition well, and its performance is better than the traditional dynamic wrapping matching algorithm.