MAD-MEX (Monitoring Activity Data for the Mexican REDD+ program) is a system to provide standardized annual wall-to-wall land cover information by automatic satellite image classification for the Mexican territory in a cost-beneficial manner. One of the aims of the system is to automatically produce a national land cover dataset in a standardized, consistent, transparent and transferable way, to ensure operative activity data monitoring. The integration of data, interfaces, processes, into one uniform, consistent and scalable hardware and software platform are the core components of MAD-MEX, but also, is the result of different governmental mexican organizations that follows international and national standardize guides.
At this moment, MAD-MEX processes data from two satellite: Landsat and Rapideye, and we will provide support for other satellite data in the future.
This repository contains information and tutorials about the requirements to install and use the MAD-Mex system. The folder "es" and "eng" contains the documentation in spanish and english respectively about how to install the system in two different architectures:
-
cluster
-
standalone
The first one is the deployment over a cluster of computers. It uses Sun Grid Engine, a job scheduler designed to optimize the use of computational resources or distributed processes in heterogeneous environments in order to use the resources in the most efficient way. In this type of architectures, SGE accepts, schedules and dispaches remote jobs, these jobs can be sequential, parallel or interactive.
The system can also be installed in standalone mode. This version uses a single machine in offline mode. This means the system does not need to be connected to a network.
In the database MAD-Mex commands directory there is a shell script that will install and configure the database used by the system. This step plays a main role for the sytem to work correctly. The whole functionality relies on a correct configuration of the database, without it, will not be possible to run the different commands to ingest, classify and detect changes.
We suggest that before installing and running any command, get through the documentation, in order to understand what does the system does and does not.
The MAD-Mex team.