Contextual feature discovery and image ranking for image object retrieval and Tag refinement

Image retrieval is emerging as one of the important research area for browsing and retrieving images from a large database. Many image retrieval techniques are there but it suffers from low precision rates due to lighting conditions, noisy tags etc. One of the most important problems that we have seen in image retrieval is the semantic gap. This paper proposes a new method for improving the performance of image retrieval system and reducing the semantic gap using Auxiliary Visual Word and Tag refinement. Here the retrieval accuracy gets improved by the discovery of contextual features. This paper also provides an image pre-processing technique using Contrast Limited Adaptive Histogram Equalization algorithm and image ranking approach. It also describes about effective and efficient distance measure and a minimum distance classification for retrieval. The experimental results show that precision has been improved with the proposed method.