摘要
图像中所蕴涵的丰富语义仅用若干低级物理特征是不能进行完整描述的,而且在语义映射时也会有信息丢失,因而产成“语义鸿沟”是在所难免的。将多特征融合,建立情感语义模型,分析情感的概念解析功能对提高智能信息检索的精度和效率是非常必要的。论文讨论了图像的颜色、纹理等特征的提取与表示,低阶图像可视化特征到高阶图像语义特征的映射过程,图像的情感语义分类,建立了情感语义模型,实现对基于情感语义图像的检索。对由2500幅数字图像组成的数据集进行了实验,并对实验结果进行分析,部分结果是令人满意的,而且提高了基于内容图像检索的精度。
The abundant semantic contained in the images can not been described completely only using some lowlevel physical features,and some information will be lost in the semantic mapping,so it is unavoidable to produce the ＂semantic gap＂.It is necessary to improve the precision and efficiency of the intellective information retrieval by syncretizing multi-features,establishing the affective semantic model and analyzing the idea-analysis function of emotion. Features extracting and expressing of image＇s color,texture,etc.,mapping process from the low-level image visual features to the high-level image semantic features,and the emotion semantic classification of the images are discussed, emotion semantic model is established,the retrieving based on affective semantic images is achieved in this paper.The data set composed of 2500 digital images is experimented with,and the experiment results have been analyzed,some of which are satisfied,and the precision based on content image retrieving has been improved.
出处
《计算机工程与应用》
CSCD
北大核心
2006年第18期82-85,共4页
Computer Engineering and Applications
