伪不变特征法影像相对辐射归一化处理应用  

Application of the Pseudo-invariant Feature Methord for Image Relative Radiometric Normalization Process

作　　者：展昀 李丹 ZHAN Yun;LI Dan(Qingdao Surveying&Mapping Institute,Qingdao 266033,China;Qingdao Key Laboratory for Integration and Application of Marine-terrestrial Geographical Information,Qingdao 266033,China)

机构地区：[1]青岛市勘察测绘研究院,山东青岛266033 [2]青岛市海陆地理信息集成与应用重点实验室,山东青岛266033

出　　处：《测绘与空间地理信息》2023年第7期57-59,共3页Geomatics & Spatial Information Technology

摘　　要：研究运用了一种基于伪不变特征的多时相遥感影像相对辐射归一化方法,采用主成分分析获取辐射伪不变特征点(PIF),克服样本选择的主观性;依据所选取的合格PIF点,运用最小二乘法解算线性回归模型参数,对影像进行辐射归一化处理。通过多组实验结果的目视分析和定量评价,证明了该方法的有效性。The relative radiometric normalization method of multi-temporal remote sensing images is developed by using pseudo-invariant feature(PIF).The PIF points are selected by the way of principal component analysis in order to reduce the influence of the subjectivity of sample selection.According to the selected qualified PIF points,the least square method is used to solve the linear regression model parameters,and radiometric normalization is carried out on the image.Through visual analysis and quantitative evaluation of experimental results,the effectiveness of this method is proved.

关 键 词：伪不变特征 主成分分析 参考影像 相对辐射归一化 

分 类 号：P237[天文地球—摄影测量与遥感]