Short Course 4: Tensorflow 2.X and Keras for building deep learning models in remote sensing applications

The dissemination and popularization of knowledge about machine learning models applied to Remote Sensing are essential. We encourage the study of key concepts related to Artificial Neural Networks and Convolutional Neural Networks, especially for classification and segmentation applications in computer vision, focusing on Digital Image Processing in the context of Remote Sensing. Additionally, we promote the dissemination of advanced models to improve mapping tasks in satellite images and the integration of new digital image processing tools on the Google Earth Engine platform, offering increasingly sophisticated and accessible solutions for the analysis and interpretation of remote sensing data. The goal is to acquire basic knowledge of Artificial Neural Networks and Convolutional Neural Networks, learn how to build datasets for model training, validation, and testing. recognize different Deep Learning model architectures and their uses, as well as adapt, train, and test models for specific datasets and validate classification models.