Object Matching Trade Study

Newland, F.T. (1996) Object Matching Trade Study

Abstract

This report outlines the latest motion analysis techniques in the fields of rigid and deformable object tracking and flow analysis. Particular examples are given for vehicle motion, ice and cloud tracking and ocean flow analysis. Image preprocessing and parameterisation methods are discussed in terms of whole-scene analysis, feature and object extraction and data associations. The application of Markov Chain Monte Carlo methods, neural networks and fuzzy logic to data extraction is also highlighted. Algorithmic similarity measures, model matching, statistical similarity analysis and neural network optimisation techniques are considered for object matching, and the application of Kalman filtering to object tracking is expounded. Where available, comparative analyses are used to highlight the benefits of each approach. Suggested approaches for the three motion types are then selected and justified from the literature, and possible future directions of research are highlighted. This report consists of a considerable literature survey of the active research areas in motion analysis, and an extensive reference list is included.

This record has no associated files available for download.

Contributors

Download statistics