This week we’ve all been hard at work on our report and reviewing the feedback we got from our presentation. I have been continuing to refine the thresholding for our piece segmentation as well as playing around with some feature extraction methods. Right now I am still running into trouble with background noise of the surface we are using, but that should theoretically be fixed by the time the new surface we will be using comes in. If not I’ve done some preliminary research on how to fix it, but they would likely require tuning for that specific surface and It is a problem that would make more sense facing when we know what our final surface will be.
Here is my use of ORB feature detection method, but as you can see by the red dots it is picking up a lot of the features from my laptop case.
