skip to primary navigationskip to content
 

Autonomous Image Recapture - Dr Michael Ramage

This research project will use advanced image processing to recreate geospatial viewpoints of historic and existing images of buildings and cities to programme autonomous flight paths for photographic recapture of those same images by drone. One novel aspect to our project will be the ability to reconstruct and retake heritage images and legacy photographs using a direct image to flight path algorithm.

This research project will use advanced image processing to recreate geospatial viewpoints of historic and existing images of buildings and cities to programme autonomous flight paths for photographic recapture of those same images by drone. One novel aspect to our project will be the ability to reconstruct and retake heritage images and legacy photographs using a direct image to flight path algorithm.

David Loggan (d. 1692) was engraver to Cambridge University, and is known for his bird’s-eye etchings of Cambridge and Oxford Colleges. Using computer vision and image processing techniques, this project will take Loggan’s 17th  century etchings and determine the best fit to the geospatial coordinates that he imagined them from. With these coordinates, we will program a drone to recreate Loggan’s images. The immediate outcomes are threefold:

  • Computer vision techniques to determine geospatial viewpoints – such techniques exist for photographs, but are not known to exist for aerial etchings for which multiple images might need to be stitched together.
  • Drone flight automation for reconstructive image capture integrated with existing 3D buildings/terrain
  • Direct comparison between Loggan’s and contemporary views to highlight urban growth, change and stasis

Initially the product of this research will be useful for time and space comparisons of cities and infrastructure, including urban growth, building condition surveys and similar, and we imagine other uses will grow as our research develops.

We expect that in that time we will be able to identify the viewpoints in a single Loggan image and make a similar onefrom a drone flight; as time allows we will extend to other views by Loggan. We plan to use Loggan’s iconic image of Trinity College (below left) to test the software and efficacy of our approach. Google Earth points toward a proof of concept (below right); our research would implement a higher accuracy, customisable and more specific algorithm. Our research will be extendable to places Google hasn’t gone, and ours will be specifically targeted, higher resolution and more visually useful.

2018MP_Ramage Image

 

Researchers


Department of Architecture


Department of Engineering


ARM


DJI

 

Filed under:

Welcome to the Centre for Digital Built Britain.  

The Centre for Digital Built Britain is a partnership between the Department of Business, Energy & Industrial Strategy and the University of Cambridge to deliver a smart digital economy for infrastructure and construction for the future and transform the UK construction industry’s approach to the way we plan, build, maintain and use our social and economic infrastructure.