Automatic Civil Infrastructure Assessment Data Mining from Google Street View Images

The objective of the project is to demonstrate the automatic mining of Google Streetview data to assess the infrastructure of municipal assets, in particular the conditions of roadways. This is the short-term vision of the project. The long-term vision is that other assets such as bridges, tunnels, sidewalks, trees, etc. will be similarly be mined. In addition, the short-term vision is to demonstrate the ability to automatically provide an assessment that is keeping with standards provided by existing systems such as PASER (Pavement Assessment Surface Rating) or PCI (Pavement Condition Index). We have decided that the initial focus will be on providing such an output that we refer to as a qualitative assessment. Subsequently in the long term vision, the focus will be on delivering an additional quantitative assessment, for example, identifying cracks and potholes, characterizing them and localizing them, etc.

Faculty Supervisor:

John Zelek

Student:

Partner:

The Public Sector Digest Inc

Discipline:

Computer science

Sector:

Professional, scientific and technical services

University:

University of Waterloo

Program: