视觉注意模型在虚拟与现实交互中的研究(英文)  

Research on Image Saliency Detection Model Based on Visual Attention

作　　者：纪超[1] 刘慧英[1] 孙景峰[1] 贺胜[1] 黄民主[1] 

机构地区：[1]西北工业大学自动化学院,西安710072

出　　处：《火力与指挥控制》2014年第5期40-44,共5页Fire Control & Command Control

基　　金：航空科学基金(2012ZC53042);西北工业大学研究生创业基金资助项目(Z2013038)

摘　　要：提出将视觉注意模型应用于虚拟与现实交互中,通过显著性原理检测出真实场景中的有效区域。采用将图像在空域和频域中分别进行显著特征提取,再通过特征组合理论进行特征融合。空域中采用区域颜色对比以及局部复杂度对比特征;提出在频域内进行有效频段分割,结合幅度信息提取显著特征,再加权后合成;特征组合理论是根据细胞调解原理来处理特征之间独立与联合作用的关系。最后提出将该方法应用于虚拟与现实交互中检测真实场景中的有效区,并同其他算法进行效果对比。According to the human visual attention mechanism,this paper introduces a novel plausible model for image saliency detection.Firstly,saliency features are detected including regional color contrast and local complexity contrast based on the sparese representation principle; And proposing the Efficient Band Divided Method which unites information of amplitude with the phase spectrum for generating the saliency feature. Then refer to the principle that activity in cells responding to stimuli,.a new feature combination theory is proposed to deal with the relationship of the characteristics of independence and mutual interaction for achieving features fusion accurately. Afterwards,the proposed algorithm can effectively detect regions through some experiments,and applied in virtual and reality interaction to detect the effective region and eliminate noise area.

关 键 词：机器视觉 显著性提取 频段分割 特征组合 

分 类 号：TP911.73[自动化与计算机技术]