|
Low contrast, noisiness, and low resolution are the fundamental
properties of infrared images. It is hard to detect small
targets infrared image. Such that the motion characteristic of
small image sequences can be utilized to track and detect
targets. We approach using dynamic programming for small target
tracking and detection in infrared image sequences in this
thesis. Dynamic programming, like divide-and-conquer method,
solves optimization problems by combining the solutions to sub-
problems. primary advantages of dynamic programming are its
sensitivity to targets along with its robustness to target
maneuvers and sensor instabilities. Such enhancements are
achieved by performing association and detection in a single
optimization procedure. To that the target motion and noise
behavior can be accurately dynamic programming is shown to be
optimal for this task. proposed approach can track two intact
trajectories, for crossing Several experiments are conducted to
examine how the proposed performs. The experiments indicate the
satisfactory results.
|