基于视觉组织的封闭轮廓编组模型研究  

Hierarchy Grouping Model of Closed Contour Based on Perceptual Organization

在线阅读下载全文

作  者:万月亮[1] 曹元大[1] 李钝[1] 

机构地区:[1]北京理工大学计算机科学技术学院,北京100081

出  处:《小型微型计算机系统》2008年第7期1314-1319,共6页Journal of Chinese Computer Systems

基  金:总装备部“十五”预研项目(413150801)资助

摘  要:模拟视觉组织获取轮廓的研究多基于局部视觉编组线索,使得编组线索在视觉上缺乏完整性.为此提出一种具有全局约束的层次编组模型.首先,自适应尺度小波检测图像边缘,显著性重要边缘作为轮廓编组候选集.其次,根据视觉编组规则特性提出层次编组模型,用格式塔规则如对称性和平行性作为全局线索指导格式塔规则如接近性、连续性、共线性及共曲线性等编组线索完成轮廓编组,获取符合视觉特性的轮廓.实验表明该模型在不损失视觉合理性的同时减少了编组的二义性,增强了算法的效率和健壮性.Although numerous studies have measured the strength of perceptual organization cues for closed contour, most of these concentrate on their local cues and ignore role of global cues roles in grouping stages for natural images. It appears that a stronger global constraint is required. A novel hierarchy grouping model is developed. Firstly, most saliency edges are extracted from automatic scale edge fields as input for subsequent grouping. Secondly, a hierarchy grouping model based on prin- ciples of global cue supervised is developed under the Gestalt principles. Experiments show the model reduces ambiguities in the intermediate representation, have high efficiency and robustness, while holding biological plausibility.

关 键 词:视觉组织 层次编组 对称性度量 封闭轮廓 

分 类 号:TN911[电子电信—通信与信息系统]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象