基于相似像元替换的遥感影像厚云去除方法研究  

Study of thick clouds removal from remotely sensed images based on similar pixel substitution approach

作　　者：赵孟银 郑小慎[1] 刘文静[2] 

机构地区：[1]天津科技大学海洋与环境学院天津市海洋资源与化学重点实验室,天津300457 [2]河海大学地球科学与工程学院,南京210098

出　　处：《计算机应用研究》2016年第11期3509-3512,共4页Application Research of Computers

基　　金：天津市自然科学基金资助项目(14JCYBJC22500)

摘　　要：去云处理是遥感图像处理以及遥感制图的重要步骤。常规的去云处理算法在去除云对影像影响的同时,通常也会伴随地物光谱信息的丢失。提出了一种基于相似像元替换遥感影像厚云去除方法。利用改进的单时相法对有云的图像进行云检测并制作云掩模图像;然后利用该图像中掩模之外区域的合适像元对掩模部分的像元进行替换;最后对边界进行后处理得到一幅过渡平滑的云去除图像。结果表明,该方法可以有效地实现厚云的去除,同时也能够有效地恢复地物的光谱信息。Cloud removing is one of the important steps in remote sensing image processing and mapping. Regular cloud remo- ving methods could remove cloud affects, however usually accompanied with lossing of spectral information. Based on similar pixel replacement, this paper proposed a new thick clouds removal method. Composed of three steps. It used an improved sin- gle-temporal cloud detection method to detect the thick cloud in the remote sensing image and generate cloud mask image. Then used appropriate similar pixels within the remaining region of the target image to fill the masked pixels. Finally, post-pro- cessing on cloud regions＇ edges in order to get a smooth result image with cloud removed. The results show that this method is highly valid in thick cloud removing and spectral information restoring.

关 键 词：厚云 云检测 相邻像元替换 波谱特征 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]