Machine Learning techniques have made possible great advances in processing high-resolution satellite images, due to the development of increasingly effective and efficient algorithms that extract relevant information from these data sets.
Observing the earth system can become complex due to the presence of variables such as clouds and atmospheric phenomena; however, instead of being an impediment, it puts to test the entire mathematical apparatus through which artificial intelligence is configured, driving the development of new multivariate analysis methods.
Now, regardless of its level of complexity, the interest that has grown around Earth observation is significant and can be traced by tracking publications in Machine Learning techniques, which are becoming more and more common.
On the other hand, the numerous potential applications of this type of technologies in the observation of the planet can be used to identify elements that still escape the physical models and would allow real-time monitoring of each of the physical-chemical and geophysical changes experienced by the Earth’s surface; likewise, applications can be generated for quality control, data monitoring and correction of biases in the observation processes, which are only possible with an adequate algorithmic development and machine learning techniques.
