Automated image recognition from aerial images—particularly in the field of satellite image analysis—provides access to large datasets that are well-suited for machine learning.
This project focuses on developing image-based methods that can automatically assess the quality of a local residential location by inputting a satellite image and predefined coordinates. The information extracted from these images represents a previously untapped source of information for existing location assessment models.
The developed methods and location models increase transparency in the real estate market for all market participants and can be used, among other things, in valuation and rent derivation.
https://research.fhstp.ac.at/projekte/immbild-lagebewertung-durch-bildanalyse