利用混合特征的多视角遥感图像配准

发布时间：2018-04-06 20:10

本文选题：多视角　切入点：遥感图像　出处：《中国图象图形学报》2017年08期

【摘要】：目的多视角遥感图像配准是遥感图像处理领域的一项关键技术,其目的是精确获取图像间被测区域发生的几何变换关系。由于航拍视角变化以及地物的空间分布和几何形状的复杂度,多视角遥感图像间会产生非刚性畸变问题,增加了图像配准的难度,为此本文提出一种利用遥感图像SIFT(scale-invariant feature transform)特征点阵的全局和局部几何结构特征进行多视角遥感图像配准的算法。方法通过增加对SIFT点阵的几何结构特征描述以及利用SIFT点阵间全局与局部几何结构特征的互补关系,提升存在非刚性畸变的多视角遥感图像配准精度。结果实验使用谷歌地球的卫星影像数据以及无人机航拍遥感数据对本文算法进行了测试,并与3种同类算法(SIFT、SURF(speeded-up robust features)、CPD(coherent point drift))进行对比实验,本文算法在存在非刚性畸变的多视角遥感图像配准中能够有效地提升SIFT特征点阵的配准精度,从而获得更加准确的多视角遥感图像配准结果。结论本文实现了一种结合SIFT特征点阵的全局和局部几何结构特征进行多视角遥感图像配准的算法,实验结果表明了该方法对存在非刚性畸变的多视角遥感图像能够有效地进行配准,可适用于同源多视角情况下的遥感图像配准问题。
[Abstract]:Objective Multi-view remote sensing image registration is a key technology in the field of remote sensing image processing.Due to the variation of aerial view angle and the complexity of spatial distribution and geometric shape of ground objects, the problem of non-rigid distortion will occur between multi-view remote sensing images, which increases the difficulty of image registration.In this paper, an algorithm of multi-view remote sensing image registration based on the global and local geometric structure features of SIFT(scale-invariant feature transform feature matrix is proposed in this paper.Methods to improve the registration accuracy of multi-view remote sensing images with non-rigid distortion by adding geometric structure description of SIFT lattice and utilizing the complementary relationship between global and local geometric structure features of SIFT lattice.The algorithm in this paper can effectively improve the registration accuracy of SIFT feature lattice in multi-view remote sensing image registration with non-rigid distortion, and obtain more accurate registration results of multi-view remote sensing image.Conclusion in this paper, a multi-view remote sensing image registration algorithm combining the global and local geometric structure features of SIFT feature lattice is implemented.The experimental results show that this method can effectively register multi-view remote sensing images with non-rigid distortion, and can be applied to the registration of remote sensing images with the same source and multi-angle.
【作者单位】：云南师范大学信息学院;西部资源环境地理信息技术教育部工程研究中心;
【基金】：国家自然科学基金项目(41661080) 云南省教育厅科学研究项目(2015Z069) 云南师范大学博士科研启动费基金项目(01000205020503065);云南师范大学大学生科研训练基金项目~~
【分类号】：TP751

【相似文献】