Lidar Imagery


I think the engineering behind how we create Lidar (Light Detection and Ranging) based surfaces (example) is pretty eloquent and cool! It's also feasible to teach in a short video. The general concept is that we place a laser ranging instrument on a platform (usually an airplane), have a rotating mirror (as you're polling hundreds of thousands of samples per second, it would be crazy to try to spin a heavy instrument like that so fast) that the laser bounces off, travels towards the ground and reflects off a surface either partially or fully as you can recieve multiple responses. A point cloud is generated for the study area, which can then be interpolated into a surface.
The key topics to cover would be:
  • Lasers (particularly how ranging works)
  • GPS and Inertial Management Units (IMU) which determines the relative position, aspect, zenith if your platform (often an airplane)
  • Basic trigonometry (If you know the x,y,z of one point, the vector and the length of a line, then you know the x,y,z of the other point!)
I've always found that when trying to tell to people about what I do as a geographer, Lidar always catches their attention. Being able to capture a surface from the sky at 5-15 centimetre resolution is exciting to be sure! If people find this to be an interesting topic, I'll fill out quite a bit more here when I get a chance.



Please add comments, suggestions, points for discussion below. Also, feel free to modify the description above.

Comments






[Very interesting: I wonder if this could go with signals in some way; or is there a series on imaging in general. bill-engineerguy]

[It certainly could go along with signals, though a better example would be spectral remote sensing (visible light, radio, microwave). In my grad work I've come to learn that Lidar is very much a different beast from traditional spectral remote sensing techniques. In this case we're measuring distance/position and rendering topography rather than spectra over an even spatial resolution. The intensity of the signal return is of value, but to a lesser extent than the spatial component. Andrew - original poster)