Abstract: For pre-and post-earthquake remote-sensing images, registration is a challenging task due to the possible deformations of the objects to be registered. To overcome this problem, a registration method based on robust projective Nonnegative Matrix Factorization is proposed to precisely register the variform objects. Firstly, a Robust Projective Nonnegative Matrix Factorization（RPNMF） method is developed to capture the common projection space of the variform objects. Secondly, a registration approach is derived from the common projection space of the variform objects. Finally, two experiments are conducted to verify the effectiveness of the proposed method：one is the SAR image registration in Wenchuan earthquake on May 12, 2008, the other is change detection of Tangjiashan barrier lake. The results show that the method is very effective in capturing the common projection space of variform objects and generalizes well for registration. Meanwhile, good performance on the change detection of barrier lake is obtained.

Key words: remote-sensing image, variform object, Nonnegative Matrix Factorization（NMF）, Robust Projective Nonnegative Matrix Factorization（RPNMF）, projection space, outliers

摘要： 由于要配准的目标存在可能的形变，震前和震后遥感图像的配准变得很困难。为了解决这个问题，提出基于稳健的投影非负矩阵分解（RPNMF）的配准方法来精确的配准形变目标。给出一种稳健的投影非负矩阵分解方法来获得震前震后形变目标的共同投影空间，利用在共同投影空间的投影来配准形变目标。为验证该算法的有效性，做了两个实验：2008年5月12日汶川地震前后的SAR图像的配准；唐家山堰塞湖的变化检测。与现有方法进行比较，结果表明该方法能够有效地得到形变目标的共同投影空间，并取得了很好的配准结果；同时，堰塞湖的变化检测也得到了很好的结果。

关键词: 遥感图像, 形变目标, 非负矩阵分解, 稳健的投影非负矩阵分解, 投影空间, 异常值