Decoding image semantics using composite region templates

Author

Smith, John R. ; Chung-Sheng-Li

Author_Institution

IBM Thomas J. Watson Res. Center, Yorktown Heights, NY, USA

fYear

1998

fDate

35967

Firstpage

9

Lastpage

13

Abstract

The authors present a method for decoding image semantics using composite region templates (CRTs). The CRTs define prototypal spatial arrangements of regions and features in the images. The system classifies unknown images by matching the strings of regions extracted from the images to the templates in a CRT library. They describe the process for generating the CRTs from photographic images by automatically segmenting the images into color regions. They demonstrate that the system performs well in classifying images from ten semantic classes and in searching for images in a large collection

Keywords

decoding; feature extraction; image classification; image colour analysis; image segmentation; string matching; visual databases; automatic image segmentation; color regions; composite region templates; image searching; image semantics decoding; large collection; photographic images; region string matching; semantic classes; spatial feature arrangements; spatial region arrangements; unknown image classification; Cathode ray tubes; Color; Decoding; Image segmentation; Layout; Libraries; Milling machines; Prototypes; Rivers; Shape;

fLanguage

English

Publisher

ieee

Conference_Titel

Content-Based Access of Image and Video Libraries, 1998. Proceedings. IEEE Workshop on

Conference_Location

Santa Barbara, CA

Print_ISBN

0-8186-8544-1

Type

conf

DOI

10.1109/IVL.1998.694467

Filename

694467