Paper Title
Web And Personal Image Annotation By Mining Label Correlation With Relaxed Visual Graph Embedding

Abstract-Image processing refers to processing of a 2D picture by a computer. An image may be considered to contain sub images. Now a day the number of digital images rapidly increases. There are unlabeled images available, for that our system assigns automatic image annotation. We propose a system for image annotation by integrating label correlation mining and visual similarity mining into a join framework. We first construct a training image database which includes image visual features. A multilabel classifier is then train by simultaneously uncovering the shared structure common to different labels and the visual graph embedded label prediction matrix for image annotation. We apply the proposed framework to both image annotation and personal album labeling using the NUS-WIDE image dataset.