摘要
提出了一种新的利用图像语义词汇表进行图像自动标注与检索的方法。采用混合层次模型在已标注好的训练图像集上计算图像区域类与关键字的联合概率分布,并用生成的模型标注未曾观察过的测试图像集,或用来进行基于语义的图像检索。实验结果表明,该方法在标注、检索精度和效率方面均优于当前其他方法。
This paper presented a new method to automatically annotate and retrieve images using a vocabulary of image semantics. This method was based on a mixture hierarchy model, and computed a joint probability distribution for image regions classes and keywords. Then it annotated unseen test images by used model. The experiments results show that the proposed method has advantages not only in terms of annotation and retrieval accuracy, but also in terms of efficiency.
出处
《计算机应用研究》
CSCD
北大核心
2007年第7期318-320,共3页
Application Research of Computers
基金
湖南省自然科学基金资助项目(05JJ40101)
关键词
图像检索
语义图像检索
图像语义
图像自动标注
image retrieval
semantic image retrieval
image semantic
automatic image annotation
