An adaptive detection model of moving dim targets based on energy difference between frames

The paper is based on biology vision stared mechanism, applies the detective arithmetic operators of static image into the system of moving dim targets detection ,brings forward an adaptive detective model of moving dim targets based on energy difference between frames, and optimizes and selects the parameters in the mathematic model. Experiments completed by the paper proving, the adaptive detective model can detect moving dim targets real-timely and exactly. Detection and tracking of moving dim targets are the most important part of the system of watching and alarming. So it's essential to find and detect the targets in time. Because of claim of detecting targets at long bowls, the targets usually are regarded as the dot targets or pixels in the system of watching and alarming. So detection of infrared dim targets is one of key technology in the weapon system, and one important project in moving targets detection. We aim at slowing down signal-to-noise ratio in the detection of moving dim targets by the way of classical frame difference and rapid matching, bring out a new anisotropic moving detection model between frames in the study of theories of biology vision fixated mechanism, and analyze and research on selection of more parameters in the detection model. Experimental results show that the algorithm based on energy difference comparing between frames which is brought out by this paper is effective and practical to detect moving dim small targets.