摘要
针对光学和SAR图像水域辐射、几何性质差异过大而难以获得共性特征,以及大场景图像配准耗时过长的问题,提出一种基于二次Otsu阈值分割法和尺度不变特征变换法(SIFT)的光学和SAR图像配准方法.首先为了降低数据量,减弱微小目标和噪声带来的干扰,对光学图像和SAR图像进行小波变换;然后针对图像一次Otsu分割后的"过检"现象进行二次分割,获得更准确的水域分割;最后通过改进的空间一致性准则和RANSAC消除初始匹配点中的较多误配.实验结果表明,该方法可以有效地提高光学和SAR图像的配准准确度,加快配准速度.
In the optical and synthetic aperture radar(SAR)image registration,the common features are difficultto obtain because of different radiometric and geometric properties.Meanwhile,the large scene images are time-consuming.For solving above problems,a fast optical and SAR water image registration method is proposed based on second Otsu image segmentation and scale-invariant feature transform(SIFT).Firstly,wavelet transform of the optical and SAR image is applied to reduce the amount of data and the interference caused by small targets and noise.Then,in order to solve the problem of“over-inspection”after the Otsu segmentation,the second segmentation is used to achieve more accurate water segmentation.Finally,the mismatch features are eliminated by improving spatial consistency criterion and random sample consensus(RANSAC).The experimental results show that the proposed method can effectively increase the accuracy of optical and SAR image registration and the computation efficiency.
作者
曹哲
张弓
戴为龙
Cao Zhe;Zhang Gong;Dai Weilong(Key Laboratory of Radar Imaging and Microwave Photonics of the Education Ministry, College of Electronic Information Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016)
出处
《计算机辅助设计与图形学学报》
EI
CSCD
北大核心
2017年第11期1963-1970,共8页
Journal of Computer-Aided Design & Computer Graphics
基金
国家自然科学基金(61071163
61271327
61471191)
航空基金(20152052026)
南京航空航天大学研究生创新基地(实验室)开放基金项目(KFJJ20160401)
