基于卡尔曼滤波的 Landsat-8卫星影像几何精校正

摘要

Kalman filter is a recursive linear minimum variance estimation algorithm,and it has been widely used in GPS dynamic data processing,integrated navigation,communication and signal processing fields.This paper applies kalman filter into geometric precision correction of Landsat-8 imagery,and completes the correction of satellite ephemeris and attitude data.In the experiments,the comparison of accuracy between kalman filter algorithm and least square method shows that the application of kalman filter in precision correction of landsat-8 imagery can reduce the dependence of GCPs.%卡尔曼滤波是一种递推线性最小方差估计算法，被广泛应用于 GPS 动态数据处理、组合导航和通信与信号处理等领域。本文将卡尔曼滤波引入卫星影像几何精校正处理中，以 Landsat-8卫星影像为例，利用卡尔曼滤波方法和地面控制点构建运动方程和测量方程，实现了对卫星星历和姿态数据的修正。实验证明，相比于传统最小二乘平差法，该方法使用较少的控制点即可达到稳定的精度，降低了几何精校正对控制点数量的依赖，具有较高适用性。