Abstract
In the scale-changing targets detection based on passive millimeter wave image sequences, a novel tracking before detection (TBD) algorithm using features fusion model is proposed. In passive millimeter wave imaging (PMWI) system, the features description of targets is more difficult than that of visible targets. Thus, a new target modeling based on gray histogram and moment invariants is applied, to solve the problem of the target deformation and partial occlusion effectively. The disadvantage of the traditional fixed window tracking is avoid, by conducting an adaptive window tracking method based on particle filter track before detection (PF-TBD). Experimental results show that the method can effectively solve the problems of targets modeling and rapid shape changes during the detecting and tracking procedure.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call