摘要
静止轨道卫星影像能有效应用于海上目标实时监测和跟踪,但其定位精度较差,需要利用影像上的岛礁控制点进行几何精校正。通常情况下,海上岛屿较为稀疏,受云雾干扰较大,控制点难以自动获取,为解决海上序列遥感影像控制点自动获取问题,本文提出一种基于决策树分类方法的控制点自动获取方法。通过对国产卫星高分四号影像进行试验,对比其他两种岛礁控制点提取方法的结果,最终精度优于其他方法。因此本文提出的技术流程能够较好的应用于海洋遥感影像的几何精校正。
The geometric correction of ocean remote sensing image is a prerequisite for its data application. With the increasing number of data, an automatic geometric correction technology needs to be researched. In this paper, to solve the problem that the sea island is sparse, cloud interference is great, the control point is difficult to obtain, an automatic correction technique based on decision tree classification is proposed. Through the experiment of GF4 image, compared with the classification results of the other two classification methods, the final precision is better than the other two methods.Therefore, the technical process proposed in this paper can be applied to the geometric correction of complex sea condition remote sensing images.
引文
[1]陈振炜,张过,宁津生,唐新明.资源三号测绘卫星自动云检测[J].测绘学报,2015(3):292-300.
[2]刘鹏宇.基于内容的图像特征提取算法的研究[D].吉林大学,2004.
[3]常学立.静止轨道高分辨率面阵相机几何处理关键技术研究[D].武汉大学,2015.
[4] Breiman L,Friedman JH,Olshen RA,etal.Classification and Regression Tree[M].Wadsworth,Inc119841.
[5]刘勇洪,牛铮,王长耀.基于MODIS数据的决策树分类方法研究与应用[J].遥感学报,2005,9(4):405-412.
[6]吴际通,谭伟,喻理飞.基于TM/ETM+影像的不同水体指数对比研究[J].测绘科学,2013,38(4):193-195.
[7]李智峰,朱谷昌,董泰锋.基于灰度共生矩阵的图像纹理特征地物分类应用[J].地质与勘探,2011,47(3):456-461.