In recent years, different deep learning techniques were applied to segment aerial and satellite images. Nevertheless, state of the art techniques for land cover segmentation does not provide accurate results to be used in real applications. This is a problem faced by institutions and companies that want to replace time-consuming and exhausting human work with AI technology. In this work, we propose a method that combines deep learning with a human-in-the-loop strategy to achieve expert-level results at a low cost. We use a neural network to segment the images. In parallel, another network is used to measure uncertainty for predicted pixels. Finally, we combine these neural networks with a human-in-the-loop approach to produce correct predictions as if developed by human photointerpreters. Applying this methodology shows that we can increase the accuracy of land cover segmentation tasks while decreasing human intervention.
The ADREAM building consists of an innovative architectural structure which integrates 6500 sensors for the purposes of optimized systems, energy surveillance. Amid growing needs for energy optimization research, new technological developments in renewable energies, the ADREAM project stands as a prototype example for creating a multidisciplinary platform that allows the rapid progress of the Energy Networks, Smart Grid domains. A description of the ADREAM project is presented along with the energetic, control systems installed. The methodology for the development, calibration of the building's thermal model, dynamic simulation is also highlighted. The simulations are exploited for the purpose of a thermal behavior, comfort,, energy consumption analysis. The results obtained by the applied parametric variations were used for the proposition of a building renovation investment. The energy optimization goal is enhanced by the development of an ARX model for the prediction, control of energy consumptions,, a Simulink model for the global simulation of all the energetic systems.