KBVision Constraint System Module Page 4
KBVision Constraint Module
Example Results
The Rule Base created for this example is delineated below:
Bush = Weighted Average of:
- Color-of-Bush = Weighted Average of:
- ExBlue-of-Bush = Tokensubset of ExBlue-Mean Attribute
- ExGreen-of-Bush = Tokensubset of ExGreen-Mean Attribute
- ExRed-of-Bush = Tokensubset of ExRed-Mean Attribute
- Shape-of-Bush = Weighted Average of:
- Compactness-of-Bush = Tokensubset of Compactness Atribute
- Size-of-Bush = Tokensubset of Pixel-Count Attribute
- Location-of-Bush = Tokensubset of Centroid.y Attribute

Weighted Average =


The illustration above shows a gray-level version of the original image with Bush Tokens overlayed in color. The Bush Tokens were further filtered based on their probability of being a Bush. This can be accomplished by Tasks in the KBVision System, or by interfacing an external Expert System.
CONSTRAINT SYSTEM Page 4 of 4
KBVision
webmaster@aai.com (Last Update 10/17/95)
KBVision and KBView are Trademarks of Amerinex Artificial Intelligence, Inc.