A Novel Image Semantic Annotation Method based on Image-Concept Distribution Model

摘要：

the optimal solution to support a great annotator is very difficult to derive.Up to the present,little work can precisely tag the images since the so-called semantic gap is not easy to reduce.This paper constitutes the approach of approximating the solution to discover the visual-concept associations from the image-concept distribution.The experimental results show that our proposed annotation approach is very effective in facing the diverse relations between visual features and human concepts.

关键词： image semantic annotation entropy,imageconcept distribution model

作者: Ma Ying Zhang Laomo Gao Jixun

作者单位: School of Software Henan Institute of Engineering Henan,China

会议类型: 国际会议

会议名称: 2010 4th International Conference on Intelligent Information Techonlogy Application(第四届智能信息技术应用国际学术研讨会 IITA 2010)

会议地点: 秦皇岛

会议语种:英文

页码: 56-59

在线出版日期: 2010-11-05（万方平台首次上网日期，不代表论文的发表时间）