MAPS 362
Terrestrial Laser Scanning
Introduction
What was done
The project was accomplished by two second year SAIT GNT Students, Andrew Morrison and Patrick Tobitt. Our task involved us creating a 3D model of the Aldred Centre, a building on SAIT campus using terrestrial laser scanning techniques. Laser scanning involves the collection of data points using pulsed beams of light which can then be captured, visualized, extracted, analyzed, shared and represented as point cloud data, We focused on the southern facade of the building, through a main entrance and the direct interior. Our methods and results will follow below. Our intention is that a reader may learn and be inspired by our project and find inspiration about the use of the laser scanning equipment as well as processing workflows extract the data with Cyclone (the software used to process the data).
"The terrestrial laser scan turned out to be an eye opening project. The BLK 360 worked fast and efficiently. I now have the confidence on how to perform a laser scan and process the data with Cyclone. Really cool!
Andrew Morrison
"At first I had my doubts about the speed and reliability of the BLK 360s but that was soon put to rest. I was blown away by the efficiency of the machines. Using cyclone to manipulate and process the data was exciting and it help my understanding of how point clouds work."
Patrick Tobitt
Methodology
How it was done
Laser scanning is the optimal choice for collecting 3D point data about the building because of the large amount of position data about the building faster and more accurately than GPS or traditional survey methods. Laser scanning, while fast and effective at collecting a large amount of data, quickly becomes difficult to manage with large enough projects. It is also limited by environmental factors related to the weather, more specifically moisture and water, be it lakes or snow.Two surveyors, one Leica BLK 360 Laser Scanner and one building to survey. This below will go further in depth to the scouting, configuration, data collection and processing aspects of the project.
Scouting
We chose the Southern facade of the Aldred Centre, near a major entrance as the loction for our laser scan. Our area of survey was exterior facade of the building, the swinging doorway leading into the building and the inside where the cafeteria is. We also had to determine which features, outside and in, we would use to register the laser scans together. The objective involved creating a a suitable model of the interior as the result.
​
​
Configuration
The Leica BLK 360 uses an app from Autodesk called Recap Pro on the iPad iOS which is used to create, organize and manage a terrestrial scanning project as well as to directly control the laser scanner. This allows for an efficient and integrated collection method.
Data Collection
Our first four scans were of the outside of the Aldred Centre. That took place on March 23, 2018. A follow up on April 6, 2018 allowed us to focus our scans to the inside of the Aldred Centre, near the location of the cafeteria, which resulted in 11 scans. The outdoor scans were assigned targets to aid the registration while the indoor scans used cloud based registration using specific features.
Processing
To create a combined 3D model of our building which includes the exterior and interior of the Aldred Center, Cyclone was used to add constraints such as the preplanned points on the exterior. Register and process the point cloud data using Visual Registration and Cloud to Cloud registration. Then, after the final merged scanworld is created, the entirety of the model space is clean of stray points and incorrect data. With further processing, Cyclone allows us represent point cloud data in intensity, black & white and true colour which aids in discerning particular features.
Results
The result is a synchronized image of the outside and and inside of the South side of the Aldred Centre completed by the Leica BLK360 laser scanner.he final RMS for our merged scans totaled 0.016m.
(Click to view)
True colour
Analysis
From our final RMS values we can confirm with certainty that our 3D model is , on average, accurately constrained to within 1.6 cm. The Individual error for each section may be higher or lowere than the final average RMS but this is well within allowed error.
Future Work and Comments
In summary, we selected a building for collection and 3D point cloud representation, cleaned the recovered data and created model representations. In the future, better times of day for the scans should be chosen to avoid the "ghosting" of images as well as the differing positions of certain features. Also care should be done to capture as man scans as feasible to ensure that there are no holes in the data. [1]
1301 16 Ave NW
Calgary, T2M 0L4
Canada
References
[1]C. Butterworth, "MAPS-362 3D Modeling", SAIT Polytechnic, 2018.
[2]R. Duchscher, "MAPS 365 - Photogrammetry", SAIT Polytechnic, 2018.
​
[3]G. Vosselman and H. Maas, Airborne and Terrestrial Laser Scanning. Dunbeath: Whittles Publishing, 2014
​
[4]Anon, (n.d.). Leica Cyclone - 3D Point Cloud Processing Software - Leica Geosystems - HDS. [online] Available at: http://hds.leica-geosystems.com/en/Leica-Cyclone_6515.htm.
​
[5] Leica Geosytems - [online] Available at:https://leica-geosystems.com/products/laser-scanners.
​