Estimating building energy efficiency from street view imagery, aerial imagery, and land surface temperature data
This work addresses the slow and costly process of energy audits for buildings, offering a scalable solution for identifying retrofit targets, though it is incremental as it applies existing deep learning methods to a new domain.
The paper tackled the problem of estimating building energy efficiency by using remotely sensed data like street view, aerial imagery, and land surface temperature, achieving a macro-averaged F1 score of 64.64% and outperforming baseline models by 12.02 to 14.13 percentage points.
Current methods to determine the energy efficiency of buildings require on-site visits of certified energy auditors which makes the process slow, costly, and geographically incomplete. To accelerate the identification of promising retrofit targets on a large scale, we propose to estimate building energy efficiency from widely available and remotely sensed data sources only, namely street view, aerial view, footprint, and satellite-borne land surface temperature (LST) data. After collecting data for almost 40,000 buildings in the United Kingdom, we combine these data sources by training multiple end-to-end deep learning models with the objective to classify buildings as energy efficient (EU rating A-D) or inefficient (EU rating E-G). After evaluating the trained models quantitatively as well as qualitatively, we extend our analysis by studying the predictive power of each data source in an ablation study. We find that the end-to-end deep learning model trained on all four data sources achieves a macro-averaged F1 score of 64.64% and outperforms the k-NN and SVM-based baseline models by 14.13 to 12.02 percentage points, respectively. Thus, this work shows the potential and complementary nature of remotely sensed data in predicting energy efficiency and opens up new opportunities for future work to integrate additional data sources.