同类传感器遥感图像自动配准的研究

Study of Automatic Registration of Remote Sensing Images in the Same Kind Sensor

遥感图像的自动配准技术是目前图像配准研究的重点。基于模板相关的匹配方法,根据同类传感器遥感图像之间灰度分布的相关性,选取数量丰富的配准控制点,利用相关系数准则进行图像间的相似性度量,可达到较高的配准精度。仿真结果表明,模板相关的匹配方法用于遥感图像的自动配准能获得很好的配准效果。

With the study of image registration recently, automatic registration of remote sensing image is becoming important. The method based template correlation matching is according to the correlation of gray-scale distribution, selects lots of registration control points, and makes use of the correlation coefficient criterion to measure the correlation of the images for high registration precision. The simulation result shows： template correlation matching acts well in the automatic registration of remote sensing images.