一种新的遥感影像变化检测相对辐射校正方法

分析了相对辐射校正在变化检测中的重要性,以基于植被指数的变化检测为例,比较常用相对辐射校正方法及其对变化检测的影响,并且提出一种基于相关系数稳健的相对辐射校正的新方法。通过试验发现相对辐射校正能够减小多时相遥感图像间由于大气、照度和传感器标度等存在差异而造成的影响,提高了基于植被指数变化检测的精度。自动稳健的相对辐射校正方法能够减少辐射误差,提高植被变化检测的精度,与传统的方法相比具有结果稳定、不容易受到误差干扰的特点。

A New Method of the Relative Radiometric Normalization in Remote Sensing Image Change Detection

This paper analyzed the importance of the Relative Radiometric Normalization(RRN) in change detection.Taking the change detection based on vegetation index as an example,it compared the common used RRN methods and its effects to change detection,and put forward a new and robust RRN method based on relative coefficients.It found out that the RRN can minimized radiometric difference among images caused by inconsistencies of acquisition conditions(such as weather,season,etc) rather than change in surface reflectance;it also can reduce the vegetation index(VI) change detection error which is produced by radiometric difference.The quality of the RRN has huge reflection of the VI change detection.Compared with the traditional ways,the new method is more robust and uneasily disturbed by error.