A big data framework for satellite images processing using apache hadoop and rasterframes: a case study of surface water extraction in Phu Tho, Viet Nam

Abstract

—Earth data, collected from many sources such as remote sensing imagery, social media, and sensors, are growing tremendously. Among them, satellite imagery which play an important roles for monitoring environment and natural changes are increased exponentially in term of both volume and speed. This paper introduces an approach to managing and analyzing such data sources based on Apache Hadoop and RasterFrames. First, it presents the architecture and the general flow of the proposed distributed framework. Based on this, we can implement and perform efficient computations on a big data in parallel without moving data to the center computer which might lead to network congestion. Finally, the paper presents a case study that analyzes the water surface of a Vietnam region using the proposed platform.