TERPOLI (2007-2010)
Automatic recognition and characterization of polygonal terrains on Mars
Project details
Funding institution
Fundação para a Ciência e a Tecnologia (FCT), contract PTDC/CTE-SPA/65092/2006
Period
October 2007 - April 2010
Principal Contractor
Centro de Recursos Naturais e Ambiente, Instituto Superior Técnico (CERENA/IST)
Principal Investigator
Pedro Pina (IST)
Team Members
CERENA/IST - Cristina Lira, Joana Antunes, José Saraiva, Lourenço Bandeira, Nuno Benavente
Abstract
The presence of polygonal patterns on several regions of the surface of Mars became known decades ago, but it was not until the advent of images with better spatial resolution that it became established that those patterns do sometimes show similar characteristics to those that occur on periglacial regions of the Earth. This happens namely in the polar regions of Mars, where the presence of vast areas in which occur polygons with the same shapes and dimensions of the terrestrial ones has been signaled. Their analysis by human operators is a time-consuming task, since it implies their manual identification.
This project aims at the development of an automated methodology which will be capable of identifying polygonal patterns on an image, with no human intervention. This process will be implemented based on a group of images selected for the purpose, in which the polygonal patterns present will be manually identified and characterized, so that there is a ground-truth that will allow for the validation of the methodology. The automated procedure will then proceed with the full characterization of the different types of polygonal patterns that can be found, through the collection of geometric and topological data on the identified polygons. This characterization will lead to a classification, related with physical controls such as location (latitude, topography) and the nature of the materials that constitute the Martian soil (namely in what refers to the presence of ice). This data will also permit that the Martian polygonal patterns be compared with their terrestrial analogues, which will point to clues on the processes that are responsible for their formation and for their evolution through time.
After a full validation, the methodology will be applied to different areas of Mars, in order to demonstrate its potential in the collection of information that can help understand the geological evolution of the planet.