基于SIFT算法的遥感图像配准研究

Research on remote sensing image registration based on SIFT algorithm

图像的自动配准是一项重要而又具有挑战性的工作,特别是对遥感图像而言。需要一种准确、鲁棒、快速的全自动配准方法。尺度不变特征变换(SIFT)算法及其改进算法在图像配准中得到了广泛地应用。然而,在遥感图像中,由于图像间的灰度映射存在显著的差异,可能很难找到足够多的正确匹配结果。采用Sobel算子进行二次梯度计算,提出了一种新的特征描述方式来克服灰度映射的差异,同时为了减少图像配准的计算量,从SIFT算法中高斯差分金字塔第二组开始进行特征检测,然后利用欧氏距离比值法进行特征匹配,最后通过快速采样一致性(FSC)算法对错误的匹配结果进行剔除。在合成孔径雷达(SAR)以及不同波段的遥感图像上进行了测试,实验结果表明,与现有的几种方法相比,该方法在配准时间和精度方面均具有很好的性能。

Automatic image registration is a vital yet challenging task,particularly for remote sensing images.A fully automatic registration approach which is accurate,robust,and fast is required.The scale-invariant feature transform algorithm and its many variants are widely used in feature-based remote sensing image registration.However,in remote sensing images,it may be difficult to find enough correct matching results due to the significant differences in the grayscale mapping between images.In this paper,Sobel operator is used to calculate the second gradient and a new feature description method is proposed to improve the effect of image registration.At the same time,in order to reduce the computational amount of image registration,feature detection is carried out from the second group of gaussian difference pyramid in SIFT algorithm.Then feature matching is carried out by Euclidean distance ratio matching method,and the wrong matching results are eliminated by fast sample consensus(FSC)algorithm.This method is tested on synthetic aperture radar(SAR)images and remote sensing images of different bands.Compared with the existing methods,the experimental results show that this method has good performance in registration time and accuracy.