Date of Award
Master of Science
Department of Electrical and Computer Engineering
Brett J. Borghetti, PhD.
An increase in sensors on the battlefield produces an abundance of collected data that overwhelms the processing capability of the DoD. Automated Visual Surveillance (AVS) seeks to use machines to better exploit increased sensor data, such as by highlighting anomalies. In this thesis, we apply AVS to overhead Full Motion Video (FMV). We seek to automate the classification of soldiers in a simulated combat scenario into their agent types. To this end, we use Multi-Dimensional Continuous Density Hidden Markov Models (MOCDHMMs), a form of HMM which models a training dataset more precisely than simple HMMs. MOCDHMMs are theoretically developed but thinly applied in literature. We discover and correct three errors which occur in HMM algorithms when applied to MOCDHMMs but not when applied to simple HMMs. We offer three fixes to the errors and show analytically why they work. To show the fixes effective, we conduct experiments on three datasets: two pilot experiment datasets and a simulated combat scenario dataset. The modified MOCDHMM algorithm gives statistically significant improvement over the standard MOCDHMM: 5% improvement in accuracy for the pilot datasets and 3% for the combat scenario dataset. In addition, results suggest that increasing the number of hidden states in an MOCDHMM classifier increases the separability of the classes but also increases classifier bias. Furthermore, we find that classification based on tracked position alone is possible and that MOCDHMM classifiers are highly resistant to noise in their training data.
DTIC Accession Number
Ross, Matthew P., "Multi-Observation Continuous Density Hidden Markov Models for Anomaly Detection in Full Motion Video" (2012). Theses and Dissertations. 1153.