MRL-SPL Vision - Matlab Spike
I know it might not sensible enough, but this is our new (color calibration free) vision. This is a Matlab spike code, which is obviously (?!?) working!
It is now learning features using Support Vector Machine (SVM), with a simple minimization. As you can see the total error is reduced and is almost convex to zero!

Comments

Post a Comment

Popular posts from this blog

Stage #1 - Replanting a pineapple

Image
I just have read an article about replanting a pineapple, and guess what; here I am! After reading couple of search queries, I decided to start this process. Let's see what happens! I have removed the rest of the fruit. Also, removed the lower leafs. According to the article [and search queries], I should wait couple of days till the lower part is completely dries. Then I can soak it into water for few weeks and hopefully the roots start to grow! Afterward, I am able to plant it in the clay. The only problem is I need somewhere with really high amount of sunlight! [Which there is none in my room...]. A Pineapple If everything goes well, this cute plant will be my friend for a while. I really enjoy having such a thing on my desk. Please let me know if you have ever tried to replant a pineapple. I am really looking for some experience in this field!
Read more

YOLO

Image
Yolo, a great library my Colleague (Ashkan) and I have found for detecting objects. YOLO stands for Y ou O nly L ook O nce, and its a deep learning computer vision library [1]. The code base is on C (not C++) and it has GPU support. These features along with the availability of designing new layers make it awesome. But the most significant feature of this library is its speed in detecting objects in real life. The default network on this project is trained to detect variety range of objects. However, the library has a huge disadvantage. It's written in C, which makes it hard to further developments. Although, there is C++ wrapper for it in [2], but the library isn't satisfying. I am still try it to fit it in my program, but after a week no progress has been attained yet... I planed to rewrite the wrapper (and in worst case the library itself) to see if I am able to  1) compile it under c++ and integrate it in my program. and 2) to see if can improve it even faster. The Y...
Read more

White Ball Detection

As the SPL rule change in 2016, the official ball has changed. Now, robots should play soccer with a black and white, foam ball. The ball, is mostly white, with black pentagons on it. Since it is the same color with goal posts, field lines and other robots, it is harder for robots to distinguish the new ball from these objects. Formerly, as well as the ball was orange, it was uniquely colored in the field. So, it could be detected simply by searching the image for specific colors, and then just by applying few filters such as width and shape. At the beginning, our strategy for detecting ball after this change, was to simply calculating image edges, and applying a Hough transform on it. Then we were expecting an acceptable outcome just by applying few filters on the result. But it didn't work! There were more problem than we expected. Calculating edge for whole image was super heavy, so we had to reduce the image resolution 4 times. Then, since the resolution was too low, the...
Read more