An Annotation Watermarking Approach to Image Tagging for Relevant Information Retrieval
Jenq‐Haur Wang, Jhih-Siang Syu, Chuan Ming Liu, Yen‐Lin Chen · Applied Mechanics and Materials · 2013
In existing image search systems, image queries can be used to find similar images through content-based image retrieval (CBIR). In order to obtain more related images, users often need to provide descriptions of the image as the keywords for search engine to extract more relevant information. But it is difficult to find appropriate keywords and text description from the image content. Searching for relevant information from search engines takes a lot of time. In this paper, we propose a CBIR system which effectively finds similar images by comparing image contents and the image annotation embedded in the image. First, we use discrete wavelet transform and two-dimensional code to embed the relevant text information or tags into the image. Then, we extract color ratios and Scale-Invariant Feature Transform (SIFT) descriptors as the image features for similarity matching. The experimental results showed that our proposed approach can accurately find similar images, and extract image-related textual information to provide useful tags for users.