Pictorial query specification for browsing through spatially referenced image databases

A pictorial query specification technique that enables the formulation of c omplex pictorial queries for browsing through a collection of spatially ref erenced images is presented. It is distinguished from most other methods by the fact that in these methods the query image specifies a target database image in its entirety whereas in bur approach the query image specifies th e combination of objects that the target database image should contain rath er than being treated as a whole image. The query objects are represented b y shape features although other features such-as color, texture or wavelets could also be used. Using our technique, it is possible to specify which p articular objects should appear in the target images well as how many occur rences of each object are required. Moreover, it is possible to specify the minimum required certainty of matching between query-image objects and dat abase-image objects, as well as to impose spatial constraints that specify bounds on the distance between objects and the relative direction between t hem. These spatial constraints can also be used to specify other topologica l relations such as enclosure, intersection overlap, etc. Each pictorial qu ery is composed of one or more query images. Each query image is constructe d by selecting the required query objects and positioning them according to the desired spatial configuration. Boolean combinations of two or more que ry images are also possible by use of AND and OR operators. A query image m ay be negated in order to specify conditions that should not be satisfied b y the database images that are retrieved successfully. In addition, a capab ility is provided to specify whether the same instance of an object is to b e used when it appears in more than one of the query images that make up th e pictorial query, or whether two different instances are allowed. Several example queries are given that demonstrate the expressive power of this que ry specification method. An algorithm for retrieving all database images th at conform to a given pictorial query specification is presented. The user interface for using this pictorial query specification method to browse the results in a map image database application is described and illustrated v ia screen shots. (C) 1998 Academic Press.