Study on the parallel inversion method of biomass remote sensing based on MapReduce model

Study on the parallel inversion method of biomass remote sensing based on MapReduce model

Shing, Liu Zhengjun, Hao Wen

MapReduce model is a new parallel programming model based on cloud computing platform. In this paper, the MapReduce parallel programming model is applied to the parallel processing of remote sensing image, and the biomass (total biomass and edible grass quantity) of the Three Rivers source area in Qinghai Province is modis13q1 by the data sources of the 5a growth period (May-October) of 2005-2009 years. A parallel inversion method of biomass remote sensing based on this model is studied. The experimental results show that the parallel biomass remote sensing inversion results are consistent with the precision validated serial inversion results, and the parallel inversion results are accurate and credible. The efficiency of parallel inversion is greatly improved, and the parallel efficiency is increased with the increase of computing nodes.

Key words: cloud computing; MapReduce model; Biomass; Parallel computing

Temp_12121308295363.rar