网格技术在遥感图像监督分类中的应用

在遥感图像监督分类中经常遇到在单独PC下使用复杂的遥感图像分类算法来实现对大数据量遥感图像监督分类的情况。在这种情况下，由于遥感图像监督分类算法的复杂性与单独PC计算能力的限制导致处理效率低下，并有可能出现内存溢出等状况。经过对网格技术与遥感图像监督分类的研究，给出基于网格环境的遥感图像监督分类算法处理该类问题的解决方案，并使用B/S结构为该应用创建可扩展测试与实验结果查看平台。实验结果表明，此方案有效提高大数据遥感图像分类的效率。

Application of Grid Technologies in Remote Sensing Image Supervised Classification

Supervised classification of remote sensing images often encountered in a singular PC using sophisticated remote sensing image classi-fication algorithm involved the large amount of data. In this case, the complexity of the remote sensing image supervised classification and capacity limitations of PC-individual computing, leads to process inefficiently as well as memory overflow may occur. Based on the Grid technologies and methods of remote sensing image supervised classification, gives Grid-based supervised classification algorithm for remote sensing image processing solutions and framework to such problems, and develops the B/S structure for the application. The platform created can implement the remote sensing image classification in Grid environment. The experimental result shows that the Grid approach effectively improves the efficiency of the large remote sensing data classification.