Neighborhood AI

Neighborhood AI is a project lead by Rolf Lyneborg Lund, Department for Sociology and Social Work at Aalborg University and solely run with the passion and interest of students and researchers alike. The goal is to explore the use of machine learning and general artificial intelligence in neighborhood research by combining computer vision with neural networks to better understand the connection between the perceived environments- and the social life within neighborhoods. All of the code will be available via this github as the project evolves.

Scroll down to see specific research topics and sub-projects

One of the first tasks of the group has been to automate image scraping via the Google Street View API and finding images of houses. To reduce the burden of the model later on, it is helpful to reduce the image load by sorting out images that has no houses in them. The overall ratio seems to be close to 25% images of houses to 75% with no houses in them. As examples, these are the typical images produced by the automated Street View API:

• 

• 

• 

It’s relatively easy to find cases where definetly are houses and where there is none. The problem arises when the cases become more intricate as seen below:

• 

• 

• 

In the first picture on the left, we have an electrical box that, if empty, could constitute as a house. In the middle, a house is maybe hiding behind the shrubbery and in the last is something that once were a house but now probably a barn. Even the simpler questions as: “What is a house?” becomes harder when wanting to automate.

Research leader

Affiliated staff

• 

• 

Affiliated students

• 

• 

• 

• 

• 

• 

• 

• 

• 

• 

• 

.