Title :
Retrieval of Images Using Mean-Shift and Gaussian Mixtures Based on Weighted Color Histograms
Author :
Bouker, Mohamed Ali ; Hervet, Eric
Author_Institution :
Comput. Sci. Dept., Univ. of Moncton, Moncton, NB, Canada
fDate :
Nov. 28 2011-Dec. 1 2011
Abstract :
The topic of this paper is Content-Based Image Retrieval (CBIR) based on colors as a content image descriptor. The tool we developed to that purpose modelizes the colors of an image as a set of 2D Gaussian distributions based on weighted color histograms. Then, given a reference image proposed by a user, the system can automatically classify the image and provide the user with the most similar images to the reference image in its category. Experiments with Corel-1000 dataset demonstrate that our method outperforms the known implementations.
Keywords :
Gaussian distribution; content-based retrieval; image classification; image colour analysis; image retrieval; 2D Gaussian distributions; Corel-1000 dataset; Gaussian mixtures; content image descriptor; content-based image retrieval; image classification; mean-shift; reference image; weighted color histograms; Dinosaurs; Histograms; Image color analysis; Image retrieval; Indexing; Kernel; Classification; Color Histograms; Content-Based Image Retrieval; Gaussian Mixtures; Mean-Shift;
Conference_Titel :
Signal-Image Technology and Internet-Based Systems (SITIS), 2011 Seventh International Conference on
Conference_Location :
Dijon
Print_ISBN :
978-1-4673-0431-3
DOI :
10.1109/SITIS.2011.75
