
Proceedings Paper
Object recognition and matching for image retrievalFormat | Member Price | Non-Member Price |
---|---|---|
$17.00 | $21.00 |
Paper Abstract
This paper proposes a new scheme for image retrieval using high-level description, which is based on object recognition and matching. For object recognition, some meaningful regions are first extracted from the image in a lower level, and an iterative procedure for object recognition is then presented. Based on the result of recognition, further matching process is carried out in a high level. The proposed techniques have been embedded into a practical system for image retrieval and experimental results show that the performance of the image retrieval could be improved through the recognition of objects and the object based matching.
Paper Details
Date Published: 31 July 2002
PDF: 7 pages
Proc. SPIE 4875, Second International Conference on Image and Graphics, (31 July 2002); doi: 10.1117/12.477118
Published in SPIE Proceedings Vol. 4875:
Second International Conference on Image and Graphics
Wei Sui, Editor(s)
PDF: 7 pages
Proc. SPIE 4875, Second International Conference on Image and Graphics, (31 July 2002); doi: 10.1117/12.477118
Show Author Affiliations
Nikolay S. Merzlyakov, Institute for Information Transmission Problems (Russia)
Published in SPIE Proceedings Vol. 4875:
Second International Conference on Image and Graphics
Wei Sui, Editor(s)
© SPIE. Terms of Use
