Title :
Fuzzy reasoning for decision making in object recognition
Author :
Walker, Ellen L.
Author_Institution :
Dept. of Comput. Sci., Rensselaer Polytech. Inst., Troy, NY, USA
Abstract :
The area of object recognition has a long history as a subarea of computer vision, with most systems using either probabilistic or heuristic methods for reasoning about the uncertainty inherent in sensed images. The paper highlights recent work in applying fuzzy set theory to the two main processes in object recognition: grouping and matching. Fuzzy sets are a natural model for both grouping and matching, using membership values to represent the degree to which groups of image features satisfy a grouping relationship, as well as the degree to which a group of image features matches an object model. This methodology can be extended to hierarchical grouping and matching as well. Finally, we describe how fuzzy reasoning can enhance an existing sophisticated object recognition system
Keywords :
fuzzy set theory; inference mechanisms; object recognition; uncertainty handling; computer vision; decision making; fuzzy reasoning; fuzzy set theory; grouping relationship; heuristic methods; hierarchical grouping; image features; matching; membership values; natural model; object recognition; sensed images; uncertainty; Charge-coupled image sensors; Computer science; Decision making; Fuzzy reasoning; Image recognition; Image sensors; Object recognition; Robot sensing systems; Sensor phenomena and characterization; Sensor systems;
Conference_Titel :
Uncertainty Modeling and Analysis, 1995, and Annual Conference of the North American Fuzzy Information Processing Society. Proceedings of ISUMA - NAFIPS '95., Third International Symposium on
Conference_Location :
College Park, MD
Print_ISBN :
0-8186-7126-2
DOI :
10.1109/ISUMA.1995.527687
