Content-based image retrieval is an important area of research. Here, a method to characterize visual appearance for determining global similarity in images is described. Images are filtered with Gaussian derivatives and geometric features are computed from the filtered images. The geometric features used here are curvature and phase. Two images may be said to be similar if they have similar distributions of such features. Global similarity may, therefore, be deduced by comparing histograms of these features. This allows for rapid retrieval. The system's performance on a database of about 1500 grey-level images and another database of 2000 trademark images is shown. It is also shown that the approach is scalable and examples of query results on a database of more than 63000 trademark images are provided.
展开▼