基于相对形状上下文的低分辨率遥感影像阵群目标关联算法  

Association Algorithm of Group Target in Low-resolution Remote Sensing Images Based on Relative Shape Context

作　　者：刘平[1] 周滨[2] 赵键[3] 

机构地区：[1]南阳理工学院计算机与信息工程学院,南阳473004 [2]中国人民解放军驻一二二厂军事代表室,哈尔滨150066 [3]国防科学技术大学电子科学与工程学院,长沙410073

出　　处：《计算机科学》2013年第3期305-309,共5页Computer Science

摘　　要：目标关联是遥感影像融合处理的重要步骤,本质上是目标配对问题。针对低分辨率遥感影像中阵群目标的特点,提出了一种基于点模式匹配的阵群目标关联算法。首先提出一种新的基于点集的不变特征——相对形状上下文特征,然后建立了以相对形状上下文特征的统计检验匹配测度为基础的阵群目标关联数学模型。为了求解该模型,在构造新的相容性度量函数来初始化关联概率矩阵后,利用松弛标记法通过迭代逐步更新关联概率矩阵,同时通过行列双向正则化最终得到满足一对一约束的最优关联匹配结果。通过仿真和实际数据实验验证了新算法的有效性和鲁棒性。One of the important pre-condition of identification fusion based on remote sensing images is target associa- tion,which is to determine if the information from two or more images are related to the same target. A novel and ro- bust point pattern matching method was presented for group target association in low-resolution remote sensing images. A new point set based invariant feature, Relative Shape Context （RSC）, was proposed. We used the test statistic of rela- tive shape context descriptor＇s matching scores as the foundation of mathematics model of group target association. For resolving the model, we firstly constructed the new compatibility probability matrix. Then the association probability matrix can be hing can be achieved by dual-normalization of rows and columns and on real world data show that the group association algorithm measurement and used it to initialize the association updated by relaxation labeling. The one-to-one mate- in the end. Experiments on both synthetic point-sets is effective and robust.

关 键 词：阵群目标关联 点模式匹配 相对形状上下文 松弛标记法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]