Image analysis – Image segmentation
Patent
1996-06-13
1998-09-29
Mancuso, Joseph
Image analysis
Image segmentation
382199, G06K 948
Patent
active
058155961
ABSTRACT:
A method for determining region boundaries in an image comprises the steps of providing an image having a plurality of discrete points, determining for each discrete point in the image an affinity force vector representing the affinity of the point to every other point in the image, and detecting region boundaries in the image using the affinity force vectors. An apparatus for determining region boundaries in an image comprises means for providing an image having a plurality of discrete points, means for determining for each discrete point in the image an affinity force vector representing the affinity of the point to every other point in the image, and means for detecting region boundaries in the image using the affinity force vectors. Automatic spatial parameter selection for input in the affinity force vector determination and regional axis detection methods are also provided.
REFERENCES:
patent: 4658372 (1987-04-01), Witkin
patent: 4747151 (1988-05-01), Knutsson et al.
patent: 4958374 (1990-09-01), Tokita et al.
patent: 5027419 (1991-06-01), Davis
Detection and Representation of Multiscale Low-level Image Structure Using a New Transform, Mark Tabb and Narendra Ahuja, ACCV '93 Asian Conference on Computer Vision, Nov., 1993.
Van Oeffelen et al. "An algorithm for Pattern Description on the Level of Relative Proximity", Pattern Recognition v16, No. 3 pp. 341-348, 1983.
Ahuja Narendra
Tabb Mark D.
Mancuso Joseph
Narendra Ahuja
Prikockis Larry J.
LandOfFree
Multiscale image edge and region detection method and apparatus does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Multiscale image edge and region detection method and apparatus, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Multiscale image edge and region detection method and apparatus will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-694017