One of the many ‘perks’ of being an honors student in university was having to ‘honors option’ a class or two every year. For some classes this was an extra set project, but because there wern’t that many of us honors students in the CS program it often meant working with the professor to come up with a project.

This was one such project for a second year algorithms course. The goal was to find and extract lines from images. I implemented the sobel edge detection algorithm, and then used a Hough transform to convert the image to MC space and then pick likely lines.

Unfortunately the results were never that good, and the MC space diagrams were often the coolest part of the results. It served as a good introduction to image transforms, and got me thinking about other ways of analyzing images and other data which would eventually lead to a flirtation with digital signal processing.