基于OpenMP的遥感影像并行ISODATA聚类研究  

Research on Parallel ISODATA Clustering for Remote Sensing Image Based on OpenMP

作　　者：刘扬[1] 王鹏[1,2] 杨瑞[1] 左宪禹[3] 张周威[2] 吴晓洋[1,2] 渠涧涛 

机构地区：[1]河南大学空间信息处理实验室,北京100101 [2]中国科学院遥感与数字地球研究所,北京100101 [3]河南大学数据与知识工程研究所,河南开封475004

出　　处：《计算机工程》2016年第7期238-243,250,共7页Computer Engineering

基　　金：国家自然科学基金资助项目(61202098);中国博士后科学基金资助面上项目(2014M552001);河南省教育厅科学技术研究基金资助重点项目(13A520071);高分辨率对地观测系统重大专项基金资助项目(Y4D0100038)

摘　　要：针对传统影像分类算法执行效率较低,无法满足海量高分辨率遥感数据实时处理需求的问题,对资源三号卫星专题产品中遥感影像的迭代自组织数据分析算法进行分析与研究,设计一种基于OpenMP的并行ISODATA聚类算法(PIsodata Omp)。采用OpenMP技术优化ISODATA算法中的样本点聚类、聚类样本中心标准差计算,实现基于共享内存的单机多核并行化处理。实验结果表明,PIsodata Omp算法能在保证分类精度不变的情况下,明显提高资源三号卫星影像数据的处理速度。The traditional image classification algorithm suffers from low efficiency and is unable to meet the requirement of mass and real-time processing of high resolution remote sensing data. Based on the analysis and research of the lterative Self-organizing Data Analysis Techniques Algorithm （ISODATA） for remote sensing images of ZY-3 satellite thematic production, this paper designs a Parallel ISODATA Based on OpenMP （PIsodataOmp）. The OpenMP technology is used for the clustering of sample points and the calculation of the standard deviation of the clustering center, realizing parallel processing on a single multi-core computer based on the shared memory. Experimental results show that, when using the PIso data Omp classification to process the image data of ZY-3 satellite, the speed of classification is improved significantly while ensuring the accuracy of classification.

关 键 词：并行聚类 迭代自组织数据分析算法 OpenMP技术 遥感影像分类 多核处理 

分 类 号：TP751.1[自动化与计算机技术—检测技术与自动化装置]