文献类型：会议论文

英文题名：Contextual information guided image categorization algorithm

作者：Shen, Hong[1]; Li, Tian-Gong[1]; Zhang, Zhen-Heng[1]

第一作者：沈辉

通讯作者：Shen, H.

机构：[1] Institute of Information Technology, Beijing Union University, Beijing, China

第一机构：北京联合大学智慧城市学院

会议论文集：Electrical Power Systems and Computers - Selected Papers from the 2011 International Conference on Electric and Electronics, EEIC 2011

会议日期：June 20, 2011 - June 22, 2011

会议地点：Nanchang, China

语种：英文

外文关键词：Electric power systems - Random processes - Semantics

摘要：This paper proposes a method for scene categorization by integrating region contextual information into the popular Bag-of-Visual-Words approach. The Bag-of-Visual-Words approach describes an image as a bag of discrete visual words, where the frequency distributions of these words are used for image categorization. However, the traditional visual words suffer from the problem when faced these patches with similar appearances but distinct semantic concepts. This paper introduces an improved contextual CRF model to learn each visual word simultaneously depending on itself and the rest of the visual words in the same region. The experimental results on the three wellknown datasets show that region contextual visual words indeed improves categorization performance compared to traditional visual words. ? Springer-Verlag Berlin Heidelberg 2011.