The GMCPHD tracker applied to the Clutter09 dataset

摘要: The contribution of this paper is twofold: first, it exposes the tracking community to a dataset previously used in acoustics studies and second, explores use features real clutter removal. For latter, Minimal Redundancy Maximal Relevance (MRMR) technique was chosen for feature selection due its flexibility on big data; top ranked by MRMR are sent C4.5 decision tree classification. Contacts that were not identified as given Gaussian Mixture Cardinalized Probability Hypothesis Density (GMCPHD) tracker. Several metrics show very small number can be employed satisfactory performance.