Translator Disclaimer
1 August 2001 Image retrieval based on data hiding
Author Affiliations +
Abstract
With the wide spread of internet, image databases with large amount of information are used in various applications such as product briefing, personal information provider, and so on. These image databases show some difficulty in retrieving the information in real-time whereas other databases using the text information are retrieved relatively fast. In this paper, image retrieval method is proposed, in which the image is extracted using the hidden text information related to it. By invisible modification of images, the images are retrieved without any headers or separate files linked to them. This paper presents the robust data hiding method by separating the image into the edge and non-edge regions. The sum of the 8 X 8 image block is quantized by adding the pseudo noise depending on the data bit. The amount of the pseudo noise is controlled adaptively based on the gradient magnitude of the image bock. Real-time extraction of extra data by the proposed algorithm is desirable for the practical image retrieval system. Experiments with various test image sets show that the proposed data embedding algorithm is robust to joint photographic experts group compression.
© (2001) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
In Su Chang and Rae-Hong Park "Image retrieval based on data hiding", Proc. SPIE 4314, Security and Watermarking of Multimedia Contents III, (1 August 2001); https://doi.org/10.1117/12.435413
PROCEEDINGS
8 PAGES


SHARE
Advertisement
Advertisement
Back to Top