“Data fusion BAUE estimation of a deterministic vector, applications to image noise and blur reduction”

GUSTAVO A. ROITMAN; BRUNO CERNUSCHI FRÍAS

Buenos Aires

Congreso; 11th Argentine Symposium on Technology, AST-2010, como parte de la 39 Conferencia Argentina en Informática, JAIIO 39, Sociedad Argentina de Investigación Operativa, SADIO., 30 de Agosto al 3 de Septiembre de 2010.; 2010

Sociedad Argentina de Investigación Operativa, SADIO.

In this work we conceive centralized data fusion as a deterministicparameter estimation problem. Two different criterions arecompared: best affine unbiased fusion rule (BAUE), and Maximum Likelihoodfor Gaussian measurement noise. Estimates are described in termsof their covariance matrices, the Cramer-Rao lower bound and simulations.The developed fusion rules are suited to two different image fusioncases: noise reduction under differently exposed images, and blur reductionbased on lens response knowledge.