Title :
Hierarchical interference scheme for high-level image understanding
Author_Institution :
Martin Marietta Lab., Baltimore, MD, USA
Abstract :
A multilayer hierarchical scheme, designed for parallel processing implementation, is proposed for the extraction of global attributed representations from spatial domain images. The input to this scheme is defined over an array of input cells that take values over a set of local image primitives. A global representation in the form of an attributed relational graph is extracted from the input using a technique that hierarchically groups the local primitives into a global features and relations
Keywords :
computerised pattern recognition; computerised picture processing; inference mechanisms; parallel processing; attributed relational graph; computerised picture processing; global attributed representations; high-level image understanding; image primitives; interference scheme; multilayer hierarchical scheme; parallel processing; pattern recognition; representation extraction; spatial domain images; Data mining; Image representation; Image sensors; Intelligent sensors; Interference; Machine intelligence; Pattern recognition; Pixel; Sensor arrays; Target recognition;
Conference_Titel :
Pattern Recognition, 1988., 9th International Conference on
Conference_Location :
Rome
Print_ISBN :
0-8186-0878-1
DOI :
10.1109/ICPR.1988.28389
