Abstract
A global space-temporal information-based sequential Monte Carlo probability hypothesis density (SMC-PHD) multi-target tracking algorithm is proposed to jointly extract multi-target peaks and tracks. This algorithm assembles particles into multiple particle clusters based on the particles' space distribution, constructs relationship between tracks and clusters, updates particle labels based on particle weights, and extracts multi-target peaks and tracks according to the evolving characteristics of the particles. Simulation results demonstrate that the proposed algorithm provides a stable tracking performance and significantly improves multi-target information extraction accuracy.
| Original language | English |
|---|---|
| Pages (from-to) | 324-333 |
| Number of pages | 10 |
| Journal | Information and Control |
| Volume | 43 |
| Issue number | 3 |
| DOIs | |
| State | Published - 2014 |
Keywords
- Global space-temporal information
- Multi-target tracking
- Peak extraction
- Probability hypothesis density
- Track management