Optimal multiframe detection and tracking in digital image sequences

Abstract

We present in this paper a Bayesian algorithm for optimal multiframe detection and tracking of small extended targets in two-dimensional (2D) finite resolution images. The algorithm integrates detection and tracking into a single framework using as data a sequence of cluttered sensor snapshots. Performance studies using Monte Carlo simulations show substantial improvements when the proposed Bayes tracker is compared to the association of a correlation filter and a linearized Kalman-Bucy filter. Likewise, there are significant detection performance gains of up to 6 dB in peak signalto-noise ratio (PSNR) when the multiframe Bayes detector is compared to a single frame likelihood ratio test (LRT) detector.

Sequential Monte Carlo Methods for Joint Detection and Tracking of Multiaspect Targets in Infrared Radar Images

“…Different targets could be assumed to move independently of each other when present and to disappear only when they move out of the target grid as discussed in Section 2. Likewise, a change in target configuration hypotheses would result in new targets appearing in uniformly random locations as in (5).…”

“…As an alternative to the conventional approaches, we introduced in [5,6] a Bayesian algorithm for joint multiframe detection and tracking of known targets, fully incorporating the statistical models for target motion and background clutter and overcoming the limitations of the usual association of single-frame correlation detectors and Kalman filter trackers in scenarios of stealthy targets. An improved version of the algorithm in [5,6] was later introduced in [7] to enable joint detection and tracking of targets with unknown and randomly changing aspect.The algorithms in [5][6][7] were however limited by the need to use discrete-valued stochastic models for both target motion and target aspect changes, with the "absent target" hypothesis treated as an additional dummy aspect state. A conventional hidden Markov model (HMM) filter was used then to perform joint minimum probability of error multiframe detection and maximum a posteriori (MAP) tracking for targets that were declared present in each frame.…”

“…To better capture target motion, we drop in this paper the previous constraint in [5][6][7] and, as in the later sections of [4], allow the unknown 2D position and velocity of the target to be continuous-valued random variables. The unknown target aspect is still modeled however as a discrete random variable defined on a finite set I, where each symbol is a pointer to a possibly rotated, scaled, and/or sheared version of the target's reference template.…”