Data Registration and Multiple-target tracking (MTT) are an essential requirement for surveillance systems which employs a large number of sensors for information collection and environment interpretation. Prior to integration of the collected information data registration is required to ensure all data are aligned. A data fusion center is a system which integrates information and consists of (1) tracking targets using a Kalman filter and its variations; (2) selecting an adequate target model to describe target maneuvers; (3) using geometric gating scoring to associate measurements (or contacts) to tracks; (4) applying either the Bayesian reasoning method or the Dempster-Shafer theory to integration of information attributed by targets; (5) employing a modified Munkres' algorithm to solve assignment problems caused by conflicts between tracks and hypotheses; (6) implementing clustering techniques to manage hypotheses at a certain level; (7) pruning a set of hypotheses to meet practical limitations. In this research we investigate registration problems and focus on designing and developing computer-aided algorithms for data registration and image fusion such as combining information of images generated by different or dissimliar sensors. Pixel-level fusion, feature level fusion and decision level fusion are of particular interest.
Master Theses
- J. Lee, Tracking a Maneuvering Target with Kinematic Constraints Using Interactive Multiple Mode Algorithm, M.S. Thesis, Department of Computer Science and Electrical Engineering, University of Maryland Baltimore County, MD, August 1995.
- S. Kwak, Multitarget Multisensor Tracking for Data Fusion, May 2000
Publications
- Y. Kim, J. Lee and C.-I Chang, "Tracking a maneuvering target with kinematic constraints using IMM method," Proc. IASTED (Int. Assoc. of Sci. and Tech. for Development) Int. Conf. Signal and Image Processing, Las Vegas, Nevada, Nov. 20-23, 1995.
- Seo, S. Kang, C.-I Chang and H. Ko, "An efficient bias estimation method in multisensor fusion for navigation by adaptive prototype selection in a bank of Kalman filters," 1999 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems, Taipei, Taiwan, Aug. 15-19, pp. 279-284, 1999.
Any comment for this web is appreciated.
Webmaster