Due to the semantic gap, we can only extract the image feature to identify indirectly the image emotional semantic. In view of the feature extraction problem of image emotional semantic identification, the image feature fusion algorithm with weights is proposed and applied to the identification of image emotional semantic in our paper. According to the effects of the extracted color, texture and shape features of image on emotional semantic, the features are weighted and fused into new feature input. Support Vector Machine is used to achieve emotional semantic identification. This algorithm is more accurate than the method that used only a kind of image features in experiments.
展开▼
