Walsh, Daniel C.I.; Raftery, Adrian E. (2002, ASQ and American Statistical Association) Pennsylvania State University, University Park, PA (email@example.com); University of Washington, Seattle, WA (firstname.lastname@example.org)
[This abstract is based on the authors' abstract.] In detecting minefields from aerial images, the problem is to determine which points in the reduced image represent actual mines. It is assumed that the field consists of nearly parallel rows of mines laid out according to a probability distribution that favors evenly spaced, linear patterns. A Markov chain Monte Carlo algorithm estimates the model and obtains posterior probabilities for each point being a mine.
False alarm rate (FAR),Image,Ordnance,Poisson distribution,Probability,Sequential experiment,Markov chains