Open Source Tools in Earth Observation: The sentinel era

The field of machine learning and pattern recognition is proving to be more and more useful in solving a wide array of problems. Open Source software libraries that incorporate machine learning algorithms have played a very important role in the widespread use of artificial intelligence in applications and scenarios where processing and analysis of large amounts of data is needed, with the accuracy and detail that extends beyond the human capabilities. Earth observation and remote sensing technologies in general, is a scientific field that can be greatly benefited from machine learning and related algorithms. The aim of our presentation is to introduce you to the use and integration of popular open source and free software for the development of spatio-temporal object monitoring services. Those types of services make use of a time-series of satellite data and prepare it in such a way that it is received as an input by the machine learning algorithms to perform a series of steps. The first step in the process is the object extraction, where useful information and specific objects and features are extracted from the input data and then, specific properties and values are assigned to them. The extracted objects are used to define models which are used as training for various learning algorithms. The output results are vital for decision making and any additional actions that need to be taken. During our presentation, we will show you various examples of services that integrate open source machine learning algorithms, such as for the implementation of CAP, land cover changes, analytics and applications in the Smart Agriculture domain. We will also discuss on how these tools can be used in conjunction with freely available satellite data in order to create high-performance services that can be fully integrated into a variety of projects and applications. Finally, we will elaborate on the many benefits that the use of open source and free machine learning libraries has, over using proprietary technologies and software to create services related to Earth Observation