Research of decision fusion for multi-source remote-sensing satellite information based on SVMs and DS evidence theory

摘要

Satellite information is an important source of the decision-level intelligence in battlefield. Research of the multi-source information decision-level fusion provides a key technical approach for distilling comprehensive satellite information and acquiring decision-level intelligences. A SVMs-DS model adopting statistics theory and uncertainty reasoning method in the article, ensures precision and reliability for decision intelligence output, by establishing multi-class classifier of intelligent performance and conflict disposing mechanism of fine fault tolerance, which lays a practical foundation for multi-source remote-sensing satellite information decision-level fusion.