Raptor: Large Scale Processing of Big Raster + Vector Data

Abstract

There has been an increase in the amount of spatial data in the recent years due to the advancements in remote sensing technology and the widespread use of smart phones and GPS technology. This has resulted in petabytes of satellite imagery as well as highly accurate geographical features such as city boundaries, roads, and others being made publicly available. Spatial data can generally be modeled in two representations: raster and vector. Satellite imagery is an example of raster data and is usually represented in form of multi-dimensional arrays. Vector data is represented as a set of points, lines, and polygons, and is used to represent geographical features such as regional boundaries. The growth of geospatial data has helped in new scientific discoveries in a wide range of applications that require combining raster and vector data. However, traditional systems implement algorithms that work with either raster or vector data. This paper proposes a novel approach for the concurrent processing of raster and vector data.