Classification

edit

Hello and thank you always for your help. Depending on a decision from the upper guys at my job, there is the probability that I won't be dealing that much with OpenCV for a while (although I have set up almost everything at home so that I can keep learning on my free time- so maybe I ll post on weekends :) ) so I would like to thank you all for all the help and insights you have given me and the things you taught me so kindly.

I would like one more question though again rather general and theoretical:

Is there the possibility of using classification with OpenCV? Classification as in the Machine Learning technique that permits a machine to "classify" an input signal (in this case an Image) as being part of a set.

To give a more concrete example: Lets say I have a camera and I point it at a banana, then the system will say "banana". If I point it at a tomato it will say "tomato". If I point it to a lettuce it will say "lettuce".

Now of course, provided that previously I have trained the system with samples of photos of a banana, a tomato and a lettuce. Say I point it to an untrained things- an apple- it will say "tomato" ergo the category to which it will most likely fit.

Take into account that the system is learning these categories and I dont explicitely explain "what is a banana"

Are there examples of these things being done with OpenCV??

Any comment, pointers or advice will be greatly appreciated