MULTIPLE TARGET TRACKING BASED ON CONSTELLATION MATCHING AND KALMAN FILTER

PDF

Authors
  1. Wong, A.K.C.
  2. Leung, H.
Corporate Authors
Waterloo Univ, Waterloo ONT (CAN) Pattern Analysis and Machine Intelligence Lab;Defence Research Establishment Ottawa, Ottawa ONT (CAN)
Abstract
A new approach to multiple target tracking (MTT) problem is developed. The data association (DA) problem is solved by an attributed subgraph isomorphism approach called constellation matching (CM). The CM method exploits, in the most direct way, the spatial configuration of the collection of targets which are subject to temporal and spatial constraints. The CM-based tracking system combines the CM technique with the Kalman filter to track and confirm the trajectories of multiple targets. The efficiency of this new approach is demonstrated using real-life multiple target radar tracking data and the results are compared to those obtained by a multiple hypothesis tracking (MHT) system.
Report Number
AGARD-CP-555-PAP-25 — CONTAINED IN 95-01669
Date of publication
01 Jan 1995
Number of Pages
20
DSTKIM No
95-01656
CANDIS No
150058
Format(s):
Document Image stored on Optical Disk

Permanent link

Document 1 of 1

Date modified: