2024 is a New Year, and a new me!

I started out the year going down a rabbit hole, but it was an interesting rabbit hole with a long history.

Are you familiar with shows such as Expedition Unknown or Lost Cities Revealed with Albert Lin? These are just two examples, but there are many more shows that showcase the use of Lidar to assist with finding structures, or lost civilizations. (I am more interested in Lidar as an archeological tool right now, rather than in self driving cars.)

As I am a big fan of these types of shows and thought it would be cool to play with some data, I did just that. This was not something that was done overnight… In my past, I have; purchased/played with a drone, took a few courses on GIS and QGIS, learned more technical aspects of Lidar and have a good understanding of working in 3D.

In my research of how to visualize the point cloud (which is a new term learned), I came across a variety of software. A lot of it proprietary, and most with a cost. QGIS is supposed to have the ability to read point clouds natively in version 3.32(?). Previously, it required a plugin called LAStools. I was able to load an LAZ file, but did not figure out how to manipulate it properly. So I kept looking.

Then I found it, the software that allowed me to be Josh Gates for a minute. Ok, not actually Josh, but one of the teams he has worked with. CloudCompare is a 3D point cloud and mesh processing software Open Source Project. Just as importantly, I found a great tutorial series by James Dietrich that allows me to break that threshold of “WTF” to “Ahhh, I can see that now.”

The door has been opened a little more, and I might peek down the hallway a little more. Something else that came to me, is a more “real world” understanding of Numpy. In one of the tutorials, James shows how to do a change detection, comparing 2 readings from different times. LAS files contain the point cloud data are essentially text files, that contain data in rows and columns. Which can be turned into an array for Numpy, which is used in CloudCompare. Then, BOOM! Fast math on 2 arrays! I understood the Numpy tutorials while doing them, but this helped it “click”.

The other aspect needed is the data. There is no shortage of this, but is it useful? I found a very nice site called OpenTopography that has a multitude of options. So many of which I have no idea what they are for! The majority of the data shown on the link is from the US. I have not searched further yet, but I am confident there is a lot more data for non US based data.

I am scraping the tip of the iceberg that is GIS data, and so far I am enjoying it!