SAR图像并行拼接方法研究与实现——基于改进完全二叉树模型  

Parallel Research and Implementation of SAR Image Mosaic Based on Optimized Complete Binary Tree Model

作　　者：孙权[1] 张华[1] 荆于勤[1] 

机构地区：[1]重庆邮电大学移通学院,重庆401520

出　　处：《重庆工商大学学报（自然科学版）》2015年第7期75-80,共6页Journal of Chongqing Technology and Business University:Natural Science Edition

基　　金：重庆市本科高校"三特行动计划"特色专业建设项目资助(渝教高(2013)49号)

摘　　要：随着SAR成像技术的快速发展,SAR图像覆盖的地域范围与成像精度也在迅速增长,这对SAR图像拼接工作的实时性提出了挑战;文章对SAR图像序列特征进行深入分析,提出了采用基于分割的完全二叉树模型进行SAR图像的并行拼接,并使用MPI点对点通信方式对之进行了不同数量机群规模下的实现;对比实验证明,基于该方法的并行拼接工作比传统并行拼接方法在降低拼接时耗上更加有效,且并行拼接后的完整图像效果良好.The regional coverage and precision of SAR image is growing rapidly with the rapid development of SAR imaging technique,which challenges the real-time processing of SAR image mosaic. This paper deeply analyses the characteristics of the SAR image sequence and presentes a model of complete binary tree based on segmentation for the SAR image parallel mosaic,and implements what this paper proposes by using the MPI pointto-point communication mode under cluster in different size. Contrast experiments prove that compared with the traditional one,this method this paper presents is more effective in reducing the time of mosaic with better effect on the complete image after parallel mosaic.

关 键 词：SAR图像 并行拼接 分割 完全二叉树 MPI 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]