Implementation of Image Registration Algorithms for Real-time Target Tracking Through Video Sequences

Abstract

Automatic detection and tracking of interesting targets from a sequence of images obtained from a reconnaissance platform is an interesting area of research for defence-related applications. Image registration is the basic step used in target tracking application. The paper briefly reviews some of the image registration algorithms, analyse their performance using a suitable image processing hardware, and selects the most suitable algorithm for a real-time target tracking application using cubic-spline model and spline model Kalman filter for the prediction of an occluded target. The algorithms developed are implemented in a ground-based image exploitation system (GIES) developed at the Aeronautical Development Establishment for unmanned aerial vehicle application, and the results presented for the images obtained during actual flight trial.