基于情感模型的感性图像检索

感性信息处理是让计算机能够模仿和识别用户的情绪、感觉和感情，实现和谐的人机交互，达到“以人为本”。通常人们用一些形容词来表达自己的情绪，本文提出了一种情感模型用来处理这些表达心理的形容词，从心理学的“维量”思想出发，采用语义量化技术和因子分析的方法建立情感空间，并分析情感空间的相似性度量方式；抽取图像的颜色和形状特征作为图像的感性特征，采用径向基函数神经网络将图像由特征空间映射到情感空间，在情感空间内实现图像的感性检索，取得了较好地实验结果。

A Kansei Image Retrieval Based on Emotion Model

The target of Kansei information processing is to simulate and recognize human sensibility, sensuality or emotion, to realize natural and harmonious human-machine interface as to achieve human-centered world. People usually use adjectives to express emotion. An emotional model based on adjectives is provided in this paper. Based on the idea of dimension?from psychology, semantic differential method and factor analysis are used to construct an orthogonal emotional space. The measurement of similarity in the emotional space is analysis. After that, the emotion model is used in image retrieval. Color and shape features extracted from the image can be used as the kansei features. A radial basis function neural network is used for mapping each image from the feature space to the emotional space. Thus an kansei image retrieval system is obtained. At last an interesting experimental result is presented.