Extracting Emotional Semantics from Color Image using Analytical Hierarchy Process

In order to understand the emotional information of the color image, research focus has been shifted from designing sophisticated low-level feature extraction algorithms to reducing the ´semantic gap´ between the visual features and the richness of human perception. In this paper we develop an emotional-based mapping scheme using analytical hierarchy process (AHP) on the basis of HSI color space. The AHP is used to provide a systematical way to evaluate the fitness of emotional labels for interpreting the emotional messages of a color image. Experimental results are given and discussed showing the excellent performance of the proposed method.