A heuristic method for ceneralized hypercube encoding
Kirti Singh, S.K. Chang, C. C. Yang · 2005
Similarity retrieval from a pic torial data base can be made more efficient by encoding the original data base into certain convenient format. Generalized hyper cube (GH) encoding is one such technique. To optimize GH encoding, a heuristic approach has been formulated. Two optimization problems have been considered here: (1) given the handle length m, find optimal GHm encoding; (2) given the threshold density, find optimal GHm encoding such that each hypercube has den sity no less than threshold density.