Image Features Extraction for Multimedia Database Content Description
Shu-Sheng Hao · 2006
The new multimedia content description standard MPEG-7 and its related framework standard MPEG-21 have been gained lots of interests by the academic and multimedia industry recently. One of the key issues that need to be determined is how to correctly extract and measure the similarity of content features. In this paper, we focus on how to effectively extract the image features and select the similar images from database. We adopt the Principle Component Analysis (PCA), Wavelet Decomposition (WD) and Moment Invariants (MI) methods to extract image features. The head-and-shoulder images (standard test images and AR database 2 ) are the main interests. The simulation results shown that our proposed algorithms can successfully select the closest images from database according to the query image.