摘要
针对当前像斑随机场模型无法充分描述方位特征的不足,在RAG-MRF模型的基础上提出了一种基于综合相邻势能分析的随机场模型,采用了新型的基于统计的邻域势函数,并设计了相应的像斑空间相邻特征描述指标,以提升像斑分类及识别的效果。
As compared to a pixels' regular neighborhood system, the morphology and distribution of segments are irregular, which makes the quantity of neighborhood cliques in RAG-MRF unpredictable, and therefore the quantification of segments' orientation features becomes impossible. Aimed at the disadvantages of current segments-based random field in orientation feature, we propose a random field model based on segments spatial integrated adjacent potential energy analysis, a new neighborhood potential energy function based on statistic was used, and according spatial adjacent feature indices for segments was designed, and the results of classify or recognition was improved.
出处
《武汉大学学报(信息科学版)》
EI
CSCD
北大核心
2013年第12期1470-1474,共5页
Geomatics and Information Science of Wuhan University
基金
国家自然科学基金资助项目(41101412)
国家863计划资助项目(2013AA102401)
中央高校基本科研业务专项基金资助项目(3101009)
关键词
像斑
方位特征
空间关系
随机场
空间邻接
segments
orientation feature
spatial relation
random field
spatial adjacency