A Robust Point-Matching Algorithm Based on Integrated Spatial Structure Constraint for Remote Sensing Image Registration
A Robust Point-Matching Algorithm Based on Integrated Spatial Structure Constraint for Remote Sensing Image Registration
复制标题
DOI：
10.1109/lgrs.2016.2605304
复制
发表时间：
2016-09
影响因子：
4.8
通讯作者：
Jie Jiang;Xiaolong Shi
中科院分区：
文献类型：
--
作者：
Jie Jiang;Xiaolong Shi
特征匹配是基于特征的图像配准的重要步骤，即从两组特征中找出正确的对应关系。本文提出了一种精度高、鲁棒性强的点匹配算法——综合空间结构约束。我们使用尺度不变特征变换算法建立了一组暂定对应，然后着重于增加正确对应的数量（内线）和去除不正确对应的数量（离群值）。首先，构造一个全局结构约束，即形状上下文，以增加内层数，同时提高正确率。然后，利用基于三角形面积表示的局部结构约束对每个对应点的相邻点进行去除离群点；实验结果表明，该算法具有较好的鲁棒性，无论在匹配精度上还是在匹配数量上都取得了较好的效果。
Feature matching, which refers to finding the correct correspondences from two sets of features, is an important step in feature-based image registration. In this letter, an accurate and highly robust point-matching algorithm, which is called the integrated spatial structure constraint, is proposed. We establish a set of tentative correspondences using the scale-invariant feature transform algorithm and then focus on increasing the number of correct correspondences (inliers) and removing incorrect correspondences (outliers). First, a global structure constraint, i.e., the shape context, is constructed for each correspondence out of the tentative set to increase the number of inliers and raise the correct rate simultaneously. Then, a local structure constraint based on the triangle area representation is utilized on the neighboring points of each correspondence to remove outliers. Experimental results compared with four state-of-the-art methods demonstrate that the proposed algorithm is robust and can achieve preferable results in terms of both matching accuracy and quantity of inliers.