Target Recognition Based on Rough Set with Multi-Source Information

As the attributes provided by multi-source information can be used to distinguish between the different species of targets, attributes recognition becomes the most important work in target recognition. In this paper, a new method for attributes recognition was proposed with rough set theory. It used a new way to described the target with a information system consisting of four elements, reduced the attribute value according to the mission requirements, valuated the attribute based on the degree of importance, and recognized targets by the sum of valuation. Finally, a set of experiments were designed to demonstrate the effectiveness of the proposed method. Furthermore, the factors that would affect the performance of the recognition system were discussed.