Histogram-based scene matching measures
Ozy Sjahputera, James M. Keller, Pascal Matsakis, Paul Gader, Johan Marjamaa · 2002
Fuzzy set theory has been used to handle uncertainty in various aspects of image processing, pattern recognition and computer vision. High-level computer vision applications hold a great potential for fuzzy set theory because of its natural language capabilities. Scene description, a language-based interpretation of regions and their relationships, is one such application that has used fuzzy sets with some success. This paper extends our earlier and ongoing work in scene description in the following sense. If we have a linguistic description (from the system or from a human), and we revisit the scene, perhaps from a different orientation, can we match the scene objects and their relationships to be confident that we are indeed in the same place. We develop a scene matching methodology to accomplish this using histograms of forces between objects.