Towards a compact representation of temporal rasters

Big research efforts have been devoted to efficiently manage spatio-temporal data. However, most works focused on vectorial data, and much less, on raster data. This work presents a new representation for raster data that evolve along time named Temporal k^2 raster. It faces the two main issues that...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2018-10
Main Authors: Cerdeira-Pena, Ana, de Bernardo, Guillermo, Fariña, Antonio, Parama, Jose R, Silva-Coira, Fernando
Format: Article
Language:English
Subjects:
Data structures
Raster
Representations
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Big research efforts have been devoted to efficiently manage spatio-temporal data. However, most works focused on vectorial data, and much less, on raster data. This work presents a new representation for raster data that evolve along time named Temporal k^2 raster. It faces the two main issues that arise when dealing with spatio-temporal data: the space consumption and the query response times. It extends a compact data structure for raster data in order to manage time and thus, it is possible to query it directly in compressed form, instead of the classical approach that requires a complete decompression before any manipulation. In addition, in the same compressed space, the new data structure includes two indexes: a spatial index and an index on the values of the cells, thus becoming a self-index for raster data.
ISSN:2331-8422