Unmanned surveillance system based on radar and vision fusion

Abstract

Unmanned surveillance system is mainly composed of objects detection and classification tasks. Traditional methods which are mostly based on vision cannot obtain the velocity and distance information of the objects. So they suffer from high false alarm rate and miss alarm rate. In this paper, a new kind of unmanned surveillance system based on radar and vision fusion is designed and realized. The main two contributions of the paper are: (1) We first introduce the objects’ area which is calculated by fusing the radar and vision data as a robust feature for objects detection and classification; (2) We proposed a new approach to detect human of different postures away from other objects. Experiments show the effectiveness and robustness of the proposed system.