Project Thermo reliably creates thermographic panoramas of indoor locations in order to identify energy inefficiencies in rooms and buildings.
I worked on this project as a volunteer at the Human Computer Interaction lab at the University of Maryland, College Park.
The software I wrote was integrated into a self-contained device which could take a series of thermographic images over the course of several days in order to map heat issues over time.
The project is a fork of OpenPano adapted for thermographic images.
After many dead-ends I ended up achieving great results by first normalizing and enhancing the images using ImageMagick, then stitching them together using OpenPano.
Pythonfor scripting everything
Image Magickfor normalizing
Exiftoolsfor reading meta-data
OpenPanofor image stitching