A Radiation Accumulation Based Detection Algorithm for Rectilinear-Motion Infrared Dim Target with Constant Acceleration

摘要

The background radiation noise interference is a difficult technical problem for the dim signal detection. Radiation accumulation is one of effective methods to improve signal-to-noise (SNR). A novel detection algorithm based on radiation accumulation is proposed and the movement model of the target limited by uniform motion is extended to the rectilinear motion with constant acceleration. Firstly, a location space and a motion space are established. Radiation accumulation operation, which is controlled by vectors from the motion space, is applied to the original image space. Then a new image space with SNR improved is acquired. Secondly, quasi-targets in the new image space are achieved by constant false-alarm ratio (CFAR) judging. Finally, location vectors and motion vectors of quasi-targets are mapped into the two spaces, respectively. Local extremum of the spaces will confirm a true target. Meanwhile, the target's states are predicted. Experimental results show the effectiveness of proposed algorithm and demonstrate the algorithm is superior to compared algorithms on detection probability and false alarm probability.