Fuzzy posterior-probabilistic fusion
2011 (English)In: Pattern Recognition, ISSN 0031-3203, E-ISSN 1873-5142, Vol. 44, no 5, 1023-1030 p.Article in journal (Refereed) PublishedText
The paradigm of the permanence of updating ratios, which is a well-proven concept in experimental engineering approximation, has recently been utilized to construct a probabilistic fusion approach for combining knowledge from multiple sources. This ratio-based probabilistic fusion, however, assumes the equal contribution of attributes of diverse evidences. This paper introduces a new framework of a fuzzy probabilistic data fusion using the principles of the permanence of ratios paradigm, and the theories of fuzzy measures and fuzzy integrals. The fuzzy sub-fusion of the proposed approach allows an effective model for incorporating evidence importance and interaction.
Place, publisher, year, edition, pages
2011. Vol. 44, no 5, 1023-1030 p.
Information fusion; Permanence of ratios; Probabilistic measures; Fuzzy measures; Fuzzy integrals
IdentifiersURN: urn:nbn:se:liu:diva-127870DOI: 10.1016/j.patcog.2010.11.014OAI: oai:DiVA.org:liu-127870DiVA: diva2:928874