Power Plant Classification from Remote Imaging with Deep Learning
This enables improved land use classification and energy mix analysis from freely available satellite imagery, but it is incremental as it applies an existing deep learning method to a new domain.
The researchers tackled the problem of classifying power plant types from medium-resolution satellite images using a ResNet-50 model, achieving a mean accuracy of 90.0% for 10 types and 87.5% for cooling mechanisms.
Satellite remote imaging enables the detailed study of land use patterns on a global scale. We investigate the possibility to improve the information content of traditional land use classification by identifying the nature of industrial sites from medium-resolution remote sensing images. In this work, we focus on classifying different types of power plants from Sentinel-2 imaging data. Using a ResNet-50 deep learning model, we are able to achieve a mean accuracy of 90.0% in distinguishing 10 different power plant types and a background class. Furthermore, we are able to identify the cooling mechanisms utilized in thermal power plants with a mean accuracy of 87.5%. Our results enable us to qualitatively investigate the energy mix from Sentinel-2 imaging data, and prove the feasibility to classify industrial sites on a global scale from freely available satellite imagery.