Optimally Adapted Indexing Trees for Medical Image Databases
Conference Paper
Overview
Identity
Additional Document Info
Other
View All
Overview
abstract
2002 IEEE. Medical image databases often have high dimensional features and standard indexing trees do not perform well with them. In this paper, we propose an algorithm for eliminating nodes in the indexing tree so that the resulting tree has improved performance. The algorithm (provably) provides the least cost tree among all possible trees that can be generated by eliminating nodes. Experimental data about performance improvement in simulations and real world are provided.
name of conference
Proceedings IEEE International Symposium on Biomedical Imaging