Image analysis – Pattern recognition – Classification
Reexamination Certificate
1998-12-09
2003-01-28
Mancuso, Joseph (Department: 2623)
Image analysis
Pattern recognition
Classification
C382S218000, C382S195000
Reexamination Certificate
active
06512850
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates generally to identifying image objects in a set of image objects, and more particularly to identifying interrelated image objects from a set of image objects.
2. Description of the Related Art
Query by image content (QBIC) is a useful technology that allows a user to query a collection of images based on visual characteristics, such as average color, color distribution, color layout, and texture. A number of publications describing QBIC exist; for example, Image Search User's Guide and Reference, Version 2.0.1 (1997), provided by International Business Machines (IBM) Corporation, Armonk, N.Y.; and U.S. Pat. No. 5,579,471, Barber et al., “IMAGE QUERY SYSTEM AND METHOD” issued on Nov. 26, 1996.
In a system employing QBIC technology, the number of image objects available for searching may be quite large, making it difficult to assess and understand the entire contents of the collection. Unfortunately, conventional technology does not yet provide a suitable means to analyze a collection of images to understand the taxonomy of those images. That is, conventional image queries alone provide very little, if any, “visual mining” capability. The ability to form relationships between images of a set may be particularly important to some businesses such as advertising agencies, animation production firms, etc.
Accordingly, it is desirable to provide means for clustering images in an analytical manner so that their underlying nature can be understood.
SUMMARY OF THE PRESENT INVENTION
A method and apparatus described herein provides a means for identifying a subset of interrelated image objects from a set of image objects. In one example, a query for identifying an interrelationship between image objects of a set of image objects is received. Each of a plurality of similarity values between image objects of the set is compared with threshold criteria from the query. A number of clusters of image object identifiers may be generated based on comparing.
REFERENCES:
patent: 5864855 (1999-01-01), Ruocco et al.
patent: 5915250 (1999-06-01), Jain et al.
patent: 6049797 (2000-04-01), Guha et al.
patent: 6104835 (2000-08-01), Han
patent: 6240423 (2001-05-01), Hirata
patent: 6240424 (2001-05-01), Hirata
patent: 6269358 (2001-07-01), Hirata
patent: WO9846998 (1998-10-01), None
Xiang et al, “Color Image Quantization by Agglomerative Clustering”; IEEE Computer graphics and Applications, ISSN: 0272-1716, vol. 14, Issue 3, pp. 44-48, May 1994.*
“A New Approach to Image Retrieval With Hierarchical Color Clustering”, Xia Wan, et al., XP-000779459 IEEE Transactions on Circuits and Systems for Video Technology, vol. 8, No. 5, Sep. 1998.
“Automatic Document Classification”, G. Salton, XP-002133538, Jan. 1989, Addison-Wesley.
Karypis et al., “Chameleon: Hierarchical Clustering Using Dynamic Modeling”,IEEE Computer, pp. 68-75, Aug., 1999.
Ganti, et al., “Mining Very Large Databases”,IEEE Computer, pp. 38-45, Aug. 1999.
Dastouri Mehrdad
Mancuso Joseph
Sawyer Law Group LLP
LandOfFree
Method of and apparatus for identifying subsets of... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method of and apparatus for identifying subsets of..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of and apparatus for identifying subsets of... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3016525