Week 11: I ran out of puns

May 08, 2019

As promised from last week, let’s discuss results.

AlexNet ended up with a very surprising plateau of 88%. Whether it will be able to identify wetness beyond still images is a good question, but one that I can’t really test due to space limitations of my computer (I cannot download video data) and the fact that it’s summer. At this point my project is “complete,” but somewhat lackluster. hence what I will be doing moving forward, even past the senior project.

I’m going to spend a little time trying a different way of feeding data my mentor thought of – to scale a certain portion of an image up to the same size as the main image, as seen below.

This is a form of emphasis – since we are enlarging the section we care about and deeming that as the wet section. The image being as it is, there’s a lot of noise going on – sky, cars, buildings – all of which can be present or not present. The road will still be wet. Going to create a python program that does this automagically. Let’s see how that goes!

Leave a Reply

Your email address will not be published.