Intro to Computer Vision

I have started working my way through the Introduction to Computer Vision course on Udacity. I’m doing this mainly because it interests me, but also because image processing and extracting information from images comes up occasionally during my work. So far I usually just hack together a solution (if you ever want to find out the direction of sastrugi from aircraft photos come talk to me) using information from the internet or Scikit-image/Scikit-learn examples. So I wanted to learn some of the theory and look at some more advanced applications.

It has been interesting so far. In the first two sections it has covered using filters, edge-detection methods, Hough transforms to find lines, circles, and arbitrary shapes, and Fourier series. Its actually been a nice little refresher on some linear algebra that I haven’t used in a while.

I’m using Jupyter notebooks to work on some examples. They can be found here.