Title :
One-Class Multiple-Look Fusion: A Theoretical Comparison of Different Approaches with Examples from Infrared Video
Author_Institution :
Sandia Nat. Labs., Albuquerque, NM, USA
Abstract :
Multiple-look fusion is quickly becoming more important in statistical pattern recognition. With increased computing power and memory one can make many measurements on an object of interest using, for example, video imagery or radar. By obtaining more views of an object, a system can make decisions with lower missed detection and false alarm errors. There are many approaches for combining information from multiple looks and we mathematically compare and contrast the sequential probability ratio test, Bayesian fusion, and Dempster-Shafer theory of evidence. Using a consistent probabilistic framework we demonstrate the differences and similarities between the approaches and show results for an application in infrared video classification.
Keywords :
Bayes methods; image classification; image fusion; image sensors; inference mechanisms; infrared imaging; probability; statistical analysis; uncertainty handling; Bayesian fusion; Dempster-Shafer theory of evidence; false alarm error; infrared video classification; object measurement; one-class multiple-look fusion; probabilistic framework; radar imaging; sequential probability ratio test; statistical pattern recognition; Bayes methods; Optical fibers; Pattern recognition; Probabilistic logic; Solid modeling; Uncertainty; Vehicles; Infrared video; Multilook fusion; One class classifier;
Conference_Titel :
Computer Vision and Pattern Recognition Workshops (CVPRW), 2013 IEEE Conference on
Conference_Location :
Portland, OR
DOI :
10.1109/CVPRW.2013.58
