The Adaptive Self Parameterising Texture Region Boundary Tracker
Author(s) -
D. M. Booth,
J. E. W. Mayhew
Publication year - 1989
Publication title -
citeseer x (the pennsylvania state university)
Language(s) - English
Resource type - Conference proceedings
DOI - 10.5244/c.3.26
Subject(s) - artificial intelligence , computer science , texture (cosmology) , computer vision , boundary (topology) , mathematics , image (mathematics) , mathematical analysis
The system operates in two modes : bootstrap and feed forward mode. In bootstrap mode a portion of the image is segmented, the texture boundary is located, and all of the parameters necessary for configuring the system are found. These include the resolution of the segmentation, identification of the most effective features for distinguishing between the two textures, and an estimate of the texture's conditional probability density functions (pdfs) given the chosen feature set. In feed forward mode the location of the texture boundary is used to position the next processing window, and this is segmented using parameters supplied by the bootstrapper.
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom