DocumentCode :
3530738
Title :
Aspects of object merging
Author :
Bronselaer, Antoon ; De Tré, Guy
Author_Institution :
Dept. of Telecommun. & Inf. Process., Ghent Univ., Ghent, Belgium
fYear :
2010
fDate :
12-14 July 2010
Firstpage :
1
Lastpage :
6
Abstract :
Information fusion is a research area that investigates how to combine information provided by independent sources into one piece of information. This topic has been studied for several applications leading to, amongst others, aggregation operators in bounded lattices and merge functions of propositional belief bases. In this paper, information fusion is investigated in the context of coreferent objects, which are objects that refer to the same real world entity. Some important properties of object merge functions are pointed out and object merge functions for both atomic and complex objects are investigated in a possibilistic framework. It is shown how merge functions for complex objects can be composed of merge functions for atomic objects, such that the composite function inherits the properties of the merge functions from which it is composed.
Keywords :
data handling; sensor fusion; statistical distributions; aggregation operators; atomic objects; complex objects; composite function; coreferent objects context; information fusion; merge functions; possibilistic framework; propositional belief bases; Batteries; Databases; File systems; Fuses; Information processing; Information systems; Lattices; Merging; Object detection; Uncertainty;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Fuzzy Information Processing Society (NAFIPS), 2010 Annual Meeting of the North American
Conference_Location :
Toronto, ON
Print_ISBN :
978-1-4244-7859-0
Electronic_ISBN :
978-1-4244-7857-6
Type :
conf
DOI :
10.1109/NAFIPS.2010.5548205
Filename :
5548205
Link To Document :
بازگشت