DAMATS Short Description

The need to access and interpret Satellite Image Time Series (SITS) becomes more important with the growing interest for identifying, monitoring and analyzing dynamic and precise spatiotemporal structures like cities, agricultural fields and forests or natural phenomena (floods, landslides). The goal of the DAMATS project is to develop a system that will comprise general analytical methods for the exploitation of the information contained in SITS.  The main focus shall be on the information extraction in the form of “categories of evolution” and elaboration of technologies to classify the evolutions processes of observed scenes.

The DAMATS concept was designed such that it provides access to the user via a web platform that will link him to the data repository and the core processing. A graphical interface enables the human machine interaction, including all the required settings and the visual exploration. The general workflow starts with the data searching and selecting the proper criteria towards SITS generation. Further, the SITS created must be analyzed using data mining tools. The content will be modeled accordingly, emphasizing the spatio-temporal evolution such that the obtained results to be considered as further input data for searching and analysis modules. The user may also extract relevant information from the scene, given a query example. Groups of pixels with similar temporal evolution will be highlighted, separating a distinctive class. The categories of evolution may be verified by comparison with short time analysis that means change detection between two acquisitions from the SITS. The final step refers to the exploratory data analysis in order to understand land cover transformations over the selected time period.

Project Details