An improved clustering concept for MHT applications

An improved clustering concept is introduced based on the premise that a cluster can contain a number of independent components. Based on the found independent components new clusters can be formed. With the introduction of the improved concept in multiple hypothesis tracking (MHT) applications, large reductions in the use of computation time and computer memory are possible. Furthermore, results are available earlier to the operator. In the paper experimental evidence for the existence of independent components is presented.