Comparison of Fast Correlation Algorithms for Target Tracking

Laser Aided Detection and Ranging (LADAR) imaging systems can be used to provide high resolution imaging and tracking of moving targets at night. Central to the tracking system is a high speed correlation algorithm to determine target motion between sensor image frames. Several issues complicate the correlation calculations. These include coherent speckle noise and atmosphere induced scintillation of the illuminator beam. The Fitts correlation algorithm is commonly used because of its simplicity and speed. However it is only optimal when the shift between sensor frames is less than a pixel. In addition it can be sensitive to certain types of noise. Projection based phase only (PBPO) is another type of correlation algorithm that is also high speed and in many cases less sensitive to noise. In this paper we compare the Fitts algorithm with PBPO in terms of number of computations and noise immunity when used in a LADAR tracker.