基于Hadoop云平台的无人机遥感图像分割

Remote Sensing Image Segmentation of UAV Based on Hadoop Cloud Platform

针对现有的图像分割方法对于无人机遥感图像分割速度慢实时性差的问题,本文展开了对基于Hadoop平台进行无人机遥感图像分割方法的研究。在分析了Hadoop云平台的结构特性以及其组件MapReduce编程思想的基础上,提出了一种利用OpenCV和Hadoop云平台相结合进行图像分割的方法。实验结果表明基于Hadoop云平台的无人机遥感图像分割,在保证较好的效果的前提下在分割速率上相比MATLAB有了很大提高,在图像分割的实效性上有了很大的提升。

In view of the problems that remote sensing image segmentation speed 1s slow and me real-rime perlormance is poor, tins pa-per makes a research on the remote sensing image segmentation method based on Hadoop platform. On the basis of analyzing the struc-tural characteristics of Hadoop cloud platform and its component MapReduce programming idea, it presents a method of image segmen-tation with the combination of 0penCV and Hadoop cloud platform. The experimental results show that the remote sensing image seg-mentation of UAV based on Hadoop cloud platform not only has better results than MATLAB, but also improves segmentation rate,which improves the effectiveness greatly in image segmentation.