Title :
Parallel and distributed methods for image retrieval with dynamic feature extraction on cluster architectures
Author_Institution :
Dept. of Comput. Sci., Tech. Univ. Clausthal, Germany
Abstract :
This paper presents techniques for parallel multimedia retrieval by considering an image database as an example. The discussed cluster architecture depicts one possible solution for the performance problem. The distribution of the image data over a large number of nodes enables a parallel processing of the compute intensive operations for dynamic image retrieval. Thus, the partitioning of the data and the applied strategies for workload balancing have a decisive impact on the performance, efficiency, and the usability of such image databases
Keywords :
distributed databases; feature extraction; image retrieval; multimedia databases; parallel databases; pattern clustering; resource allocation; visual databases; cluster architecture; distributed databases; feature extraction; image database; image retrieval; multimedia database; parallel databases; parallel processing; workload balancing; Art; Computer architecture; Data mining; Feature extraction; Image databases; Image processing; Image retrieval; Information retrieval; Multimedia databases; Multimedia systems;
Conference_Titel :
Database and Expert Systems Applications, 2001. Proceedings. 12th International Workshop on
Conference_Location :
Munich
Print_ISBN :
0-7695-1230-5
DOI :
10.1109/DEXA.2001.953049
