In this post, I will be talking about my dissertation project, which is based on Computer Vision. Lets start from the beginning, I’m working on Ball detection and distance measurement for RoboCup, specifically on NAO Robots. I’m developing this project in Java and using Java Media Framework (JMF) for visual input from webcam and grabbing frames for extracting information from them. As I mentioned earlier in this post, this program will be implemented on NAO Robots for RoboCup. It has to be really quick as ball can move around very quickly in matches. Also, it should use very low amount of resources because NAO Robot has limited resources and there will be a lot of other processing as well.
I posted ‘Real-time Shape Detection’ demonstration video long back, which is also developed in Java and follows almost same logic. However in this project, shape detection isn’t required for multiple and ‘same looking’ objects. Main target will be detecting ball only but one major problem is noise removal and high variance on lightness and surrounding colours which has high impact on the balls colour.
This is where Image Processing will be used, Noise removal and image enhancement algorithms can be applied. But Is it efficient to apply Image processing here? considering the frame processing rate as those algorithms will surely make it slow. Even with enhancement algorithms, you can’t do much about high variance on lightness. So far, all this post and development plan is focused on RGB model and even shape detection demonstration is based on RGB model as well.
The problem with RGB model is, RGB focuses on implementation details. RGB is how computer or machine treats and uses colours not how human eyes perceive colours. On the other hand, HSV focuses on actual colour components and HSV is how human eyes perceive colours. HSV can be used to find the dominating or highly influenced colour as well as its intensity.
Ironically HSV is used more than RGB for Computer Vision techniques. So right now, I’m reading and working on HSV colour model, applying it for shape detection and hoping to get better results. If all goes well, then I will be using HSV colour model for my dissertation project.
That’s it for now and I will update soon on this experiment!
Starting my robotics journey with learning and working on Aldebaran Robotics’ NAO Robot at Cognitive Robotics Research Centre, which is led by Dr. Torbjorn Dahl. It’s been more than a month since i have joined them and environment is just awesome. Everyone is really helpful and good to me. They helped me a lot while learning to work with NAO Robot and still doing, I’m the one eats their brain more than the robot does.
Dr. Torbjorn Dahl talks about Robotics, RoboCup and future developments of robots in an interview by NPORTTV
As a member of CRRC, I just started researching on Vision Processing ( Image Processing and Vision Analysis ) and Robot Localization using a different concept than the RoboCup team is currently working on. I know I’m not that capable yet and have a long way to go, But always remember “Adapt, Improvise and Overcome” that’s how it goes.
So the early future plans are:
1) A lot of work and research on Image processing algorithms.
2) Working with “Telepathy” module of Naoqi.
3) Work on computational intelligence algorithms.
Okay, Here’s the end of this update with few links:
First of all, Thanks for being so supportive, appreciating my work and posting such nice comments. Also, Sorry for not updating my blog as I’m really busy these days.
And yes, few updates from my research.. Yeah Yeah! I know, I’m busy but still, I can’t stop it, I’m addicted to it now. New things, New problems, New ways to think and Finally the New SOLUTIONS ! That’s how it goes!
Let’s talk about Sphinx, firstly, I got a partner to work on it. “Puneet Jindal” , Another Stubborn guy like me *Lol*, always ready to burn up his mind and a die hard Algo’s Lover. He’s pursuing B.Tech ( finaly year ) from NIT,Kurukshetra. We have got 85-90% accuracy on hundreds ( as the Accuracy Tracker says ) and now we are working with thousands of words to get same accuracy level on them.
Second major topic is HTML5, And I’m really loving it ! Not much to share about it. Just want to say, “HTML5 is just SO AWESOME” !
Now the upcoming topics, Optical Character Recognition (OCR), 3D Painting and Gaming/Artificial Intelligence Algo’s. I haven’t really started working on these topics, You can say that I’m having my one eye on them.