Journal of Shanghai Jiaotong University ›› 2014, Vol. 48 ›› Issue (11): 1627-1632.

• Astronomy,Geoscience • Previous Articles     Next Articles

Parallel Resampling Method of Remote Sensing Data Based on Pre-Partitioning for Cloud Computing

CHI Ziwena,b,ZHANG Fenga,b,DU Zhenhonga,b,LIU Renyia,b   

  1. (a. Zhejiang Provincial Key Laboratory of GIS; b. Institute of Geographical Information Science, Zhejiang University, Hangzhou 310028, China)

Abstract:

Abstract: In order to solve the problem of parallel resampling of remote sensing image data in cloud computing, which is the basis for rapid publication of massive remote sensing image date, a parallel resampling method of remote sensing data based on prepartitioning was proposed in combination with the features of MapReduce parallel computing and the characteristics of remote sensing image data processing. Through the prepartitioning mechanism, the image data splitting and parallel resampling tasks can be effectively controlled, and the problem of MapReduce framework application in the unstructured remote sensing data with spatial location features processing was solved, thereby, the efficient parallel resampling of remote sensing image data in cloud computing environment is implemented. In the experiment, a parallel resampling flow on the opensource Hadoop platform was designed according to the parallel resampling method of remote sensing data based on prepartitioning. The experiment and analysis show that the parallel resampling method has a good resampling performance and is capable of achieving the efficient resampling of high resolution remote sensing image data in cloud computing environment.

Key words:  cloud computing, pre-partitioning, parallel computing, parallel resampling, remote sensing image

CLC Number: