A Parallel Modeling Algorithm for Semantic Image Hierarchal

摘要

A parallel modeling algorithm for semantic image hierarchal is proposed. The image semantic network frame(ISNF) structure is based on a 'weighted'' similarity measure for comparing pairs of images data, composed by two distances, the so-called color feature and texture feature. The parallel image hierarchal create (PIHC) algorithm links the image semantic information and image Features together, through related semantic of the characteristic of low-level of the pictures and concept entities on the high-level of commercial affair. In order to show the performance of the proposed PIHC algorithm, a simulation study and two illustrative applications are discussed, on which our proposed approach as well as other related techniques are implemented and compared. Extensive experiments are conducted to investigate the effectiveness of our PIHC algorithm, which is found to be consistently better than other approaches.