Pattern recognition methods in image and video databases: Past, present and future

Abstract

Image and video (multimedia) database systems have been on an increase in recent years. Several applications demand the retrieval of multimedia data from these database systems based on their content. The users of these systems and applications perceive the data in different ways and demand the ability to query the data based on their perception of the content. This need has spurred an interest to develop pattern recognition methods which can capture the visual information content and place them in a suitable form for database indexing. This paper describes some of the image and video database systems and the various pattern recognition methods used therein.