Image-Identification Methods for Camera-Equipped Mobile Phones

As the availability and distribution of content through the medium of networks is increasing, the need for information-search technology keyed to content for the purposes of marketing, searching, and managing it is also constantly increasing. This paper proposes methods of extracting image features...

Full description

Saved in:
Bibliographic Details
Main Authors: Abe, T., Takada, T., Kawamura, H., Yasuno, T., Sonehara, N.
Format: Conference Proceeding
Language:English
Subjects:
Cameras
Data mining
Feature extraction
Image retrieval
Information retrieval
Mobile communication
Mobile handsets
Pixel
Watermarking
Working environment noise
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:As the availability and distribution of content through the medium of networks is increasing, the need for information-search technology keyed to content for the purposes of marketing, searching, and managing it is also constantly increasing. This paper proposes methods of extracting image features that can be used as content ID information, and of identifying identical images as keys to these features. As a result of experiments done on images, to which noise and lighting effects had been added, results were obtained that demonstrated better identification using our methods than that with existing ones. Further, the possibility of timely, efficient ID was also demonstrated in low-resource environments using mobile phones equipped with a camera.
ISSN:1551-6245
2375-0324
DOI:10.1109/MDM.2007.79