ROI Detection Based on the Visual Attention Model

摘要：

ROI (Region of Interest) detection is a precondition of ROI-based image retrieval,which has been a hot topic in CBIR.So its very important to search for the interested regions according to human attention.However,there always are two problems in existing methods.First,they are complicated and only apply to single area.Second,the attraction results dont match the vision well.To address these problems,this paper suggests a ROI detection approach based on the visual attention model,which judges the pixels salience by saliency map.Experimental results demonstrate that this approach achieves state-of-theart performance.

关键词： ROI Visual Attention Model Saliency Map image retrieval

作者: Xiaonan Chen Minqing Zhang Lin Ma JianHua Guo

作者单位: Key Laboratory of Network and Information Security,Engineering College of Chinese Armied Police Forc Equipment and Transport Department,Engineering College of Chinese Armied Police Force,Xian 710086,C Department of Communication Engineering,Engineering College of Chinese Armied Police Force,Xian 710

会议类型: 国际会议

会议名称: 2011 3rd International Conference on Computer and Network Technology(ICCNT 2011)(2011第三届IEEE计算机与网络技术国际会议)

会议地点: 太原

会议语种:英文

页码: 151-153

在线出版日期: 2011-02-26（万方平台首次上网日期，不代表论文的发表时间）