Historical Facade Analysis – Alternative Workflows

In the world of historic preservation the utilization of the latest software and hardware isn’t exactly shunned, but it is not widely adopted either. There are numerous reasons for this, I’m sure. I could even venture to say that I know a few of them. A popular ‘certainty’ that some have is the perceivably incredible cost. I am not one to rock the boat with that thought process.

Well, OK, maybe a little.

What if we could leverage the tools already at the disposal of many CAD businesses to enhance the historical documentation and analysis process? What I would like to propose are a few options that facilitate robust information leveraging whilst providing a technologically contemporary, cost-effective solution.

Terrestrial Photogrammetry Alternatives

Marrying terrestrial scanning and photogrammetric methods for historic architectural documentation and analysis has been an important topic in papers for some time now. Within this workflow exists many challenges. Documenting historical sites with any sort of imaging techniques can prove time-consuming due to visibility obstructions such as tourists and vegetation occluding the facades (Alshawabkeh, et.al.), weather considerations (especially in the Northwest), and processing the data in and of itself.

Our team recently worked on a project requiring a comprehensive facade analysis. The background of our expertise has included aerial photogrammetry, CAD modeling, GIS, and other mapping methodologies. Terrestrial photogrammetry and the analysis thereof is a relatively new animal to these Oregonians and newer still to many architectural firms in the Portland area. Be that as it may, we found a few simple techniques to solve these documentation problems.

Our project was a historic building the size of a city block having sandstone facades. The initial approach to this project began with laser scanning the exterior walls and interior courtyard "lightwell" using a FARO Photon LS at 1/2 resolution and x3 noise compression to ensure good, measurable detail at the highest elevation of approximately 115′ from the ground. We also wanted to ensure we could stitch all of our scans into a comprehensive 3D point cloud model, which would include the interior lightwell of this city block-sized historic structure. Because it was not time or cost-effective to daisy-chain the scans through this complex building into the courtyard lightwell from the outside directly, we decided to scan the lightwell from the roof at 1:1 resolution to ensure good geometry tie points to surrounding buildings. This ultimately allowed us to tie in the lightwell facades to the exterior skin, resulting in our full 3D model. The resulting point cloud model was processed through FARO Scene’s, ‘Project Point Cloud’ algorithm clocking in at a "brief" 22 hour processing time. This gave our client the ability to tag up the 3D building model with images and work order information in FARO Scene LT 5.0.

Finally scans were exported as POD files and brought into Microstation V8i SS3 for analysis. From here using the stock Pointools Vortex Engine, our team was able to access RGB values using a color depth map to show facade lean and spalling. This was done without any modeling at all; simply and quickly defining the ACS plane was all that was required. We also provided the client the intensity backscatter imagery with Range and Offset set to 13 and 46 respectively, providing a comparative analysis from the depth map imagery.

To the right is an example of intensity RGB. Note the areas of efflorescence in the warmer hues of pink and red – a simple, but highly informational graphic, easily produced through Microstation.

For further analysis, we used the Gigapan Epic Pro panoramic photography system to capture large swaths of the exterior at 200mm focal length producing very high resolution imagery. Using Hugin 2011.4.0 open-source software, we were able to not only stitch the photos together, but remove perspective as well through an automated control point selection process. This gave the architects the ability to see missing mortar between bricks and stones from 115′-125′ away for their analysis reports and subsequent recommendations to the building owner for repairs, right from the comfort of their own office (or tablet).

We have been using new ways to document and move around the occlusion problem. I’ll leave you with a hint of what that may look like. Whoever said terrestrial photogrammetry couldn’t be fun?!

References:

Alshawabkeh, Y., Haala, N., Fritsch, D. (2009). A new true ortho-photo methodology for complex archaeological application. University of Oxford, 2009, Archaeometry,

http://dx.doi.org/10.1111/j.1475-4754.2009.00484.x