基于IBP的天气雷达超分辨率重建算法  

Super-Resolution reconstruction algorithm of weather radar based on IBP

作　　者：何建新[1] 任红萍[1] 曾强宇[1] 李学华[1] 

机构地区：[1]成都信息工程学院中国气象局大气探测重点开放实验室,成都610225

出　　处：《四川大学学报（自然科学版）》2014年第5期947-952,共6页Journal of Sichuan University(Natural Science Edition)

基　　金：国家自然科学基金资助项目(41075010);四川省教育厅重大培育项目(11ZZ015)

摘　　要：雷达基数据分辨率不够会导致灾害性天气系统预报、跟踪的准确率低.为了提高基数据的分辨率,将迭代反投影算法应用于雷达回波超分辨率重建中,并对算法进行了改进.该方法基于对天气雷达回波使用滑动窗的重建模型,并引入时间相关性约束,由低分辨率回波重建出高分辨率回波.基于天气雷达提供的实例反射率回波,给出其超分辨率重建后的结果,并对其重建效果进行估计和比较.实验表明该方法在定量估计和视觉评价上都具有良好的高分辨率重建效果.The insufficient resolution of the radar base data lead to the low forecast and tracking accuracy of severe weather. In this paper, an improved iterative back-projection algorithm was proposed to radar echo Super-Resolution reconstruction to improve the base data resolution. This approach based on the sliding window reconstruction model, using the temporal correlation constraint, through low resolution echo to reconstruct high resolution echo. Through the actual reflectivity echo provided by weather ra- dar, presents the results of super-resolution reconstruction, comparison and estimate its reconstruction performance. The experimental results indicate that the proposed algorithm can effectively reconstruct high resolution echo in terms of both quantitative measurement and visual evaluation.

关 键 词：天气雷达 超分辨率 灾害性天气 迭代反投影 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]