摘要
针对图像底层视觉特征与高层情感语义间的语义鸿沟,分析色彩与情感的关系,建立情感空间,采用概率神经网络建立图像颜色特征与情感空间的映射,融合图像底层视觉特征和人类情感特征,提出并实现了一种感性图像检索方法。实验结果证明了该方法具有较好的效果,为图像情感语义的研究提供了一种行之有效的解决方案。
According to the semantic gap between image visual features and the high-level semantic relations among the underlying emotions, the relationship between color and emotion was analyzed. The emotion space was constructed. Probabilistic neural network was used for the mapping from the color space to the emotional space. The low-level visual features of the image and characteristics of human emotion were combined. The kansei image retrieval was proposed and implemented. The experimental results showed that the method has a good effect in providing an effective solution for the research of image emotional semantics.
出处
《莆田学院学报》
2013年第5期70-73,共4页
Journal of putian University
基金
福建江夏学院青年科研基金资助项目(2012C027)
关键词
感性图像检索
颜色特征
神经网络
情感空间
情感语义
kansei image retrieval
color feature
neural network
emotional space
emotional semantics
