An image retrieval method based on controllable feature
Xia Wang · Journal of Nanjing University of Information Science & Technology · 2010
A new image retrieval method is proposed in this paper,which uses the weighted average of color and spatial features as similarity measure between two images.Experimental results show that both precision and recall are significantly improved,compared with most common image retrieval methods;and images with similar structures and completely different colors can be retrieved efficiently.With ajustable weighted factors for different features,this method enables its users to regulate and control retrieval process according to feedback information.