Similarity Comparison of Images Based on Earth Mover's Distance

An image retrieval system is a computer system for browsing searching and retrieving images from a large database of digital images. Image retrieval technology has been used in several applications such as fingerprint identification biodiversity information systems digital libraries crime prevention medicine and historical research. Improving the image retrieval applications is a major concern in today’s huge databases. This book presents three new dimension reduction methods to improve performance and running time of an image retrieval application using embedded Earth Mover’s Distance. We experimentally evaluate our methods and compare their results with previous methods based on EMD similarity measurement.