Drone images of 16 brick facades in Flatås, Gothenburg

The images capture both damaged and undamaged areas of the facades, focusing on areas with no or small window openings. The database was used in a case study to illustrate the PAIR methodology, which was developed in "Predicting Facade Deterioration: A Machine Learning Approach Using Drone Imagery and Microclimate Data" to predict deterioration of facade surfaces based on spatial data from the DTCC 3D model of the facade environment.

Specifically, the data was used to: - Fine-tune an object detection algorithm to detect bricks subjected to spalling. - Analyze the spatial distribution of deterioration across facades. - Leverage the data to Photographs were taken using DJI Mavic 3 drone equipped with a Hasselblad L2D-20c camera. Resolution of each photo is 5280x3956. Structure: - description.xlsx contains basic information about the facades i.e., index, address, and facade orientation - Folders contain the photographs.