A Method for FLIR Target Tracking Based on Distance Updating

Abstract

This paper deals with the problem that how to use the distance information of millimeter wave (MMW) radar in the target-tracking of forward looking infrared (FLIR) image sequences. Many conventional image based trackers use timing updating to update template, but how to confirm an appropriate updating period is a difficult problem. However, the presence of noise and image magnification as the camera approaches the target may cause a tracked template to drift. In a MMW/FLIR dual-mode seeking guidance system, MMW radar can provide the precise distance between the missile and the target. The distance information can be used to direct how to update template. Firstly, a distance updating strategy is confirmed after image blowup's influence on pertinence is analyzed. Secondly, magnified template is used to direct template updating after images are magnified by distance information. Lastly, an adaptive template updating algorithm using distance information is presented. Results acquired for real FLIR images obtained from an aircraft are presented. Experiments results show that the higher tracking accuracy and the stronger stability are obtained in the algorithm proposed than those in other methods.