摘要
针对遥感图像与DEM数据之间难以找到精确同名地物点而造成的配准精度较低问题,提出了一种基于光照模型的图像配准方法。该方法首先计算DEM数据中每一个像元的方位和坡度,并结合遥感图像成像时的太阳高度角和方位角,计算图像与DEM的地形光照模型,最后通过光照模型来辅助控制点的选取,从而实现图像精确配准。实验结果表明,该方法稳定有效,特别适合地形起伏较大的区域。
Registration between remote sensing images and DEM data is difficult because it is hard to find the Ground Control Points (GCPs) between the two images. In this paper, an illumination model was introduced and served as a proxy in the image registration to increase the registration accuracy. The aspect and slope were computed for each pixel of the image based on DEM data. The solar elevation angle and azimuth were obtained from the header file of the remote-sensing image. Based on these parameters, the authors constructed the terrain illumination model based on the Lambert reflectance model, which displays very similar texture to the remote sensing'image. Thus, the GCPs can be easily identified in the two images. This method was tested by using sub ETM + image and DEM in Guangdong province. The result shows that the total RMSE of this method is 13. 373 m, which is less than one pixel. This method is effective in the registration of remote sensing image and DEM, especially in mountain areas.
出处
《国土资源遥感》
CSCD
2011年第1期83-86,共4页
Remote Sensing for Land & Resources
基金
国家自然科学基金主任基金项目(编号:41040021)
国家自然科学基金项目(编号:40901206)
中央高校基本科研业务费专项基金项目(编号:CUGL100211)
测绘遥感信息工程国家重点实验室开放研究基金项目(编号:(09)重01)联合资助
关键词
遥感图像
DEM
配准
光照模型
Remote sensing image
DEM
Registration
Illumination model