Method and system for similarity-based image classification

Details Method and system for similarity-based image classification Method and system for similarity-based image classification

2000-06-16

2001-07-31

Alam, Hosain T. (Department: 2172)

Data processing: database and file management or data structures

Database design

Data structure types

C707S793000, C709S218000, C382S305000

Reexamination Certificate

active

06269358

ABSTRACT:

BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates generally to the field of classifying and querying a database of images.
2. Description of the Related Art
Huge amounts of information are being circulated daily at an uninterrupted pace on the World-Wide Web (WWW). Additionally, in museums and photo stock agencies, millions of images are stored for on-line usage. With the explosive growth in the volume and distribution of information, more intelligent information processing and management has become critical. Various methods of accessing target data have been developed to address issues relating to image retrieval, image clustering, query interface and WWW information retrieval.
Several experimental image-clustering systems have also been proposed. In K. Hirata, et al, “The Concept of Media-based Navigation and Its Implementation on Hypermedia System ‘Miyabi’,” NEC Research & Development, Vol.35, No. 4, pp. 410-420, October 1994, the present inventor has focused on color information. Color values are extracted from the image and are mapped on to hue, lightness and saturation (HLS) color spaces. Based on the results, users can access an images directory or filter out the images for searching.
A. Del Bimbo, et al, “Shape Indexing by Structural Properties,” International Conference on Multimedia Computing and Systems, pp.370-377, June, 1997 focuses on clustering based upon shape similarity. Based on a multi-scale analysis, Del Bimbo et al. have attempted to extract the hierarchical structure of shape. Using this hierarchical structure, Del Bimbo et al. have tried to provide effective search capabilities. While this method is based on the boundary analysis, it assumes that boundary is extracted correctly. However, this is not always the case, since images extracted from the Web usually include so many elements, thus, making it difficult to extract individual objects. Del Bimbo et al. does not describe a way to solve this problem.
Image indexing using feature vectors is based on moment invariant (See, e.g., Flickner et al, “Query by Image and Video Content: The QBIC System,” Intelligent Multimedia Information Retrieval, edited by Mark T. Maybury, Chapter 1, Reprinted from IEEE Computer, 28(9): 23-31, September, 1995), or on boundary features (See R. Mehrotra et al, “Similar-Shape Retrieval in Shape Data Management,” IEEE Computer, pp. 57-62, September 1995). Such image indexing also assumes the correct extraction of the object (or input by a user). This method is very hard to apply directly to large-scale image systems, such as the WWW, requiring automatic classification.
C. Carson, et al, “Color- and Texture-Based Image Segmentation Using EM and Its Application to Image Querying and Classification,” IEEE Transaction on Pattern Analysis and Machine Intelligence, in review (http://www.cs.berkeley.edu/~carson/papers/pami.html), extracts objects from images based on color and texture. Using the combination of extracted objects and their attributes (top two colors and texture), Carson et al. tries to categorize images into several groups. Shape or positional information is not considered.
SUMMARY OF THE INVENTION
To overcome these and other difficulties, the present invention is directed to a novel method and apparatus for content based image classification and searching. A method and apparatus are described which utilize the concept of a primary object or shape. The primary object or shape corresponds to one simple composition or one simple object in the image and is defined according to the application. In accordance with the invention, a search engine classifies images in a database using the primary object as a clustering center. To perform this classification, the search engine uses boundary line similarity criteria. The compositional information of the image is described using a set of these primary objects. These classification methods improve graphical query operations as well as search speed.
A new query interface is also presented, based upon the primary objects utilizing the results of the aforementioned classification. Users need only to specify a set of primary objects to create a query. This helps reduce the user's cognitive barrier in specifying a query.


REFERENCES:
patent: 5579471 (1996-11-01), Barber et al.
patent: 5647058 (1997-07-01), Agrawal et al.
patent: 5852823 (1998-12-01), De Bonet
patent: 5893095 (1999-04-01), Jain et al.
“Geographical Image Classification and Retrieval,” Sheikhuleslami et al., Proceedings of the 5th International Workshop on Advances in Geographic Information Systems, Las Vegas, USA, Nov. 10-14, 1975, pp.58-61.*
“Region-Based Image Querying,” Carson et al., Proceedings of the 1997 IEEE Workshop on Content-Based Access of Image and Video Libraries, Jun., 1997, pp. 42-59.*
Microsoft Press Computer Dictionary, 2nd Edition, 1994, p. 326.*
“Retrieval of the Trade mark images by shape feature—the ARTISAN project,” IEE Coolquium in Intelligent Image Databases, 1996, pp. 9/1 through 9/6, Sep. 1996.*
Orphanoudakis et al., “12Cnet: Content-based Similarity Search in Geographically Distributed Repositories of Medical Images,”Computerized Medical Imaging and Graphics,vol. 20, No. 4, pp. 193-207, Jul. 1996.
Kanahara et al., “A Flexible Image Retrieval Using Explicit Visual Instruction,” Proceedings of the 1995 Third Int'l Conference on Document Analysis and Recognition, vol. 1, Aug. 1995, IEEE, pp. 175-178.
Smith et al., “SaFe: A General Framework for Integrated Spatial and Feature Image Search,” IEEE First Workshop on Multimedia Processing, Jun. 1997, pp. 301-306.
Smith et al., “Visually Searching the Web for Content,”IEEE Multimedia,vol. 4, No. 3, pp. 12-20, Jul. 1997.
A. Del Bimbo and B. Pala, “Shape Indexing by Structural Properties”, International Conference on Human Factors in Computing Systems, pp. 370-377, Jun. 1997.
Chad Carson, et al., “Color-and Texture-Based Image Segmentation Using EM and Its Application to Image Querying and Classification”, IEEE Transaction on Pattern Analysis and Machine Intelligence, in review (http://www.cs.berkeley.edu/~carson/papers/pami.html).
Kyoji Hirata et al., “The Concept of Media-Based Navigation and Its Implementation on Hypermedia System ‘Miyabi’”,NEC Research and Development,vol. 35, No. 4, pp. 410-420, Oct. 1994.
Myron Flickner, et al., “Query by Image and Video Content: The QBIC™ System”,IEEE Computer,28(9):23-31, Sep. 1995.
Rajiv Mehrotra and James E. Gary, “Similar-Shape Retrieval in Shape Data Management”,IEEE Computer,pp. 57-62, Sep. 1995.
Sougata Mukherjea, et al., “Towards a Multimedia World-Wide Web Information Retrieval Engine”, Proceedings of the Sixth International World-Wide Web Conference, pp. 177-188, Apr. 1997.

Affiliated with

Also associated with

Rating

Method and system for similarity-based image classification 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 and system for similarity-based image classification, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and system for similarity-based image classification will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2492284