STUDY ON THE RELATIONSHIP BETWEEN THE AROUSAL DIMENSION AND IMAGE LOW-LEVEL FEATURES

摘要：

In this paper, relationships between image low-level features and the emotion it arises for human has been studied. Firstly, color feature, shape feature and texture feature are extracted from the pictures of the image database International Affective Picture System. Secondly, the Classification Based on Association is taken to mine rules between these lowlevel features and arousal in the emotion space. After that, these rules are used to predict subjects’ emotional response to a new picture. Experimental results prove the effectiveness of our approach, and show that image low-level features do affect the arousal dimension of the emotion space to a certain extent.

关键词： image low-level features arousal CBA IAPS

作者: Yongjie Hu Shangfei Wang Rui Ding

作者单位: Department of Computer Science,University of Science and Technology of China,Hefei,Anhui 230027,P.R.China

会议类型: 国际会议

会议名称: 2008年拟人系统国际会议(2008 International Conference on Humanized Systems )(ICHS’08)

会议地点: 北京

会议语种:英文

在线出版日期: 2008-10-18（万方平台首次上网日期，不代表论文的发表时间）