SFM POINT CLOUD GENERATION, SCALING, AND GEOREFERENCING APPROACHES IN AN UNDERGRADUATE FIELD COURSE

Making a point cloud is not hard to do, even with smartphones or very low end cameras. More challenging is the alignment and scaling of the point cloud. This presentation documents the challenges and successes of using low cost and open source software to teach point cloud generation in field settings. Students worked on the project in a phased approach, where they learned how to use surveying tools to locate objects in the field, then build point cloud models back in the computer lab. Two significant hurdles challenged the students. Even with background lectures on the use of ground control points to facilitate model scaling and rotation, most students had problems properly surveying GCPs, and then using GCPs to align and scale their point clouds. Perhaps one issue is the condensed nature of the exercise, which took place over two weeks (two 75 minute lectures and two 225 minute lab sessions). Some ideas for future implementation include practice labs indoors in easily controlled and manageable settings before conducting field surveys, conducting two or more such field surveys from start to finish, and extending the lab over a three week period.