欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对震前震后合成孔径雷达（sAR）图像中发生复杂形变的目标，提出了基于稳健的加权核主成分分析（KPCA）的配准方法。首先，提出一种稳健的加权KPCA（RWKPCA）方法，不仅能获得震前震后形变目标的共同稳健核主成分（RKPCs），还可以作为异常值判别准则；其次，利用在共同RKPCs上的投影定义震前震后形变目标特征的相似性度量；最后，利用特征的相似性度量精确配准形变目标。对2008年5月12日汶川地震前后的SAR图像进行配准并与现有方法进行比较，结果表明，本文方法能够有效的得到形变目标的共同RKPCs，并得到很好的配准结果。
英文摘要：
The registration of pre- and post-earthquake synthetic aperture radar （SAR） images is a chal- lenging problem. The difficulty lies in that the variform objects to be registered often have complex deformations. To solve this problem, this paper proposes a new registration approach based on robust weighted kernel principal component analysis （KPCA）. We show how the variform objects of pre- and post-earthquake can be precisely registered using their robust kernel principal components （RKPCs）. The contribution can be divided into three parts. Firstly,a robust weighted KPCA （RWKPCA） method is developed,which can not only capture the common RKPCs of the variform objects of pre- and post- earthquake,but also act as the criterion for outlier detection. Secondly,based on the projections on com- mon RKPCs, the similarity measure of the features of the variform objects of pre- and post-earthquake is defined, and thus the matching results can be obtained. Finally, a variform objects registration approach is derived from the defined similarity measure and the matching results. Two experiments are conducted on the SAR image registration in Wenehuan earthquake, and the results show that compared with the existing methods,our method is more effective in capturing the common RKPCs of the variform objects of pre- and pos-earthquake, and thus has a better registration result.
同期刊论文项目
同项目期刊论文
Local discriminant non-negative matrix factorization feature extraction for hyperspectral image clas
期刊信息
位置：