Application of multi-sensor data fusion technology in target recognition

This paper presents a D-S evidence reasoning based on multi-sensor system, in which two CCD cameras and one IR sensor are used to extract features for discerning missile target, jamming light cluster and fire pile. With D-S evidence reasoning algorithm, we recognize missile target and jamming light on region square feature and cluster and fire pile on position feature, respectively. The experimental results show that the recognition probability obtained by fusing the D-S evidence of three sensors is greatly improved compared with using single sensors.