Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors

The City’s Data Portfolio: Oblique Aerial Photographs – Estimating the Potential for Green Facades Using AI

Content

A city’s data portfolio offers a wealth of untapped analytical potential. The same data can be used for a wide range of applications, and combining different data sets provides a good way to validate analytical and evaluation results through direct comparison, for example when no ‘real-world data’ is available for cross-referencing. Data such as point clouds, LoD2 or Street View data have already been used for storey detection (there is an academy here), parking space analysis and building typology. This article now explains the potential of oblique aerial images as a geodata source for assessing the potential for greening façades. It outlines how the potential index for façade greening is determined and the advantages and disadvantages of using oblique aerial images. This is followed by a brief overview of data sources used to date and their benefits for previous analyses in an urban context.

The Key Learnings

Expansion of the data portfolio

The City of Leipzig’s data portfolio has already been used for applications such as floor-by-floor detection, parking space analysis and building classification. Oblique aerial photographs as a data source further expand these possibilities, particularly for façade analysis.

A panoramic view thanks to oblique aerial photographs

As they are captured from four different directions, oblique aerial images can help to reveal all the side walls of a building. Since they contain camera parameters, this enables the conversion of real-world coordinates into pixel values, which is a prerequisite for this application.

The Potential Index as an indicator for green facades

There are a number of factors that influence green facades. Some of these can already be determined using the available data (LoD2 building model and Street View images). The various influencing factors are combined to produce a potential index for each facade wall.

Combining data makes sense

The wide range of urban data sources (Street View, orthophotos, point clouds, LoD2) each have their own strengths and weaknesses. It becomes clear just how useful combining these data sets can be for a wide range of analyses in an urban context.

Contact

Aruscha Kramm

Center for Scalable Data Analytics and Artificial Intelligence (ScaDS.AI), Department of Computer Science, University of Leipzig

digital@leipzig.de

The Partner Cities
Funded by